Cuesta‑Casanovas et al. Cancer Cell International Cancer Cell International (2023) 23:97 https://doi.org/10.1186/s12935-023-02944-4 RESEARCH Open Access Prolactin receptor signaling induces acquisition of chemoresistance and reduces clonogenicity in acute myeloid leukemia Laia Cuesta‑Casanovas1,2, Jennifer Delgado‑Martínez1,3, Josep M. Cornet‑Masana1, José M. Carbó1, Antònia Banús‑Mulet1, Francesca Guijarro1,4,5,6, Jordi Esteve1,4,5,6 and Ruth M. Risueño1* Abstract Background Development of precision medicine requires the identification of easily detectable and druggable biomarkers. Despite recent targeted drug approvals, prognosis of acute myeloid leukemia (AML) patients needs to be greatly improved, as relapse and refractory disease are still difficult to manage. Thus, new therapeutic approaches are needed. Based on in silico-generated preliminary data and the literature, the role of the prolactin (PRL)-mediated signaling was interrogated in AML. Methods Protein expression and cell viability were determined by flow cytometry. Repopulation capacity was stud‑ ied in murine xenotransplantation assays. Gene expression was measured by qPCR and luciferase-reporters. SA-β-Gal staining was used as a senescence marker. Results The prolactin receptor (PRLR) was upregulated in AML cells, as compared to their healthy counterpart. The genetic and molecular inhibition of this receptor reduced the colony-forming potential. Disruption of the PRLR signaling, either using a mutant PRL or a dominant-negative isoform of PRLR, reduced the leukemia burden in vivo, in xenotransplantation assays. The expression levels of PRLR directly correlated with resistance to cytarabine. Indeed, acquired cytarabine resistance was accompanied with the induction of PRLR surface expression. The signaling associ‑ ated to PRLR in AML was mainly mediated by Stat5, in contrast to the residual function of Stat3. In concordance, Stat5 mRNA was significantly overexpressed at mRNA levels in relapse AML samples. A senescence-like phenotype, meas‑ ured by SA-β-gal staining, was induced upon enforced expression of PRLR in AML cells, partially dependent on ATR. Similar to the previously described chemoresistance-induced senescence in AML, no cell cycle arrest was observed. Additionally, the therapeutic potential of PRLR in AML was genetically validated. Conclusions These results support the role of PRLR as a therapeutic target for AML and the further development of drug discovery programs searching for specific PRLR inhibitors. Keywords Prolactin receptor, Leukemia, Senescence, Biomarker, Differentiation *Correspondence: Ruth M. Risueño 5 Faculty of Medicine, University of Barcelona, Barcelona, Spain risueno@carrerasresearch.org 6 1 Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Josep Carreras Leukaemia Research Institute (IJC), Campus ICO‑GTP, Crta Barcelona, Spain Can Ruti, Camí de les Escoles, s/n, 08916 Badalona, Barcelona, Spain 2 Faculty of Biosciences, Autonomous University of Barcelona, Barcelona, Spain 3 Faculty of Pharmacy, University of Barcelona, Barcelona, Spain 4 Department of Hematology, Hospital Clínic, Barcelona, Spain © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Cuesta‑Casanovas et al. Cancer Cell International (2023) 23:97 Page 2 of 18 Background Methods The prolactin (PRL) system has emerged as a relevant Primary samples player in neoplasia, especially in breast and prostate Primary AML (n = 61) and MDS samples (n = 39) were cancers. Although the precise role of PRL-mediated obtained from patients diagnosed at Hospital Clínic of PRL receptor (PRLR) signaling is still controversial in Barcelona (Spain) and Hospital Germans Trias i Pujol breast cancer cells, PRLR activation seems to be impor- (Badalona, Spain), following the 2016 WHO classification tant in tumor initiation, whereas its inhibition may miti- [15] (Table 1 and Additional file 3: Table S1). Healthy- gate aggressiveness and/or dissemination of stablished donor buffy coats (n = 7) and umbilical cord blood units tumors (reviewed in [1]). PRLR is a class I hematopoietic (n = 4) were provided by Banc de Sang i Teixits (BST, Bar- cytokine receptor, that lacks intrinsic enzymatic activ- celona, Spain). ity. The Jak2-Stat5 axis is the main PRLR-mediated sig- nal transducer [2]. Contrary to PRL, multiple isoforms of mRNA expression from public repositories PRLR result from alternative splicing of the primary tran- The analyses from public repositories were mainly per- script that mainly affects the intracellular domain and, formed in the E2: Genomics Analysis and Visualization consequently, the intracellular signaling, although the Platform (https://​hgser​ver1.​amc.​nl/​cgi-​bin/​r2/​main.​ physiological role of each isoform is still controversial. cgi), using the public database GSE13159 (MILE study) PRL-induced Stat5 activation is associated to high-grade [16, 17]. Probes analysed were: PRLR, 211917_s_at; PRL, prostate tumors and aggressiveness, and PRLR inhibition 205445_at; Stat3, 208991_at; Stat5a, 203010_at. Diagno- reduced tumor burden [3, 4]. Despite promising preclini- sis vs. relapse analysis was performed in the public data- cal data, the clinical-grade anti-PRLR antibody LFA102 base GSE66525 [18]. showed no relevant clinical antitumor activity in clini- cal trials in either tumor type [5]. Overexpression of PRL Cytotoxicity and proliferation assays and/or PRLR were also described in glioblastoma [6], 1.5 × ­105 cells/mL or 2.5 × ­106 primary AML cells/mL gynecological cancers [7–9], prostate cancer [10] among were cultured in 96-well plates and all drugs (Addi- other tumor types, reinforcing the notion of PRL:PRLR tional file 2: Methods) were added at the indicated axis as a broad neoplastic feature providing proliferation concentrations, as previously described [19]. For prolif- and/or survival signaling. eration assays, AML cells were stained with 1 µg/mL DiI Acute Myeloid Leukemia (AML) is a blood neopla- (Thermo Fisher Scientific) [20]. sia characterized by the rapid expansion of transformed immature myeloid progenitors in bone marrow and Clonogenicity assays peripheral organs, marked by poor prognosis and fre- 1 × ­103 cells of AML cell lines or 50 × ­103 primary AML quent relapses. Cytotoxic therapy remains the backbone cells or lineage-depleted umbilical cord blood cells were therapy for the last decades, despite recently approved treated at the indicated concentration for 18 h and cul- targeted-pharmacological approaches [11]. Thus, identi- tured in 1 mL of MethoCult H4034 Optimum (StemCell fication of new biomarkers with therapeutic potential is Technologies) in 24-well plates, as described previously urgently needed. AML cell lines express PRLR, and its [19]. activation enhances migration, adhesion and prolifera- tion in vitro [12]. Epidemiologically, hyperprolactinemia PRLR‑silencing with CRISPR/Cas9 technology is associated with an elevated risk for hematopoietic can- Specific sgRNA sequence with the target sequences cer [13], suggesting that high PRL levels in plasma might GAG​CCA​AGA​CGC​TCA​CCA​CTAGG and GTT​CGC​ correlate with transformation events in the hematopoi- TGC​AAA​CCA​GAC​CATGG were obtained from Chop- etic system [14]. Chop (https://​chopc​hop.​cbu.​uib.​no/) and cloned into de Based on the literature and on a preliminary in silico pLentiGuide-Puro lentiviral vector (Addgene #52963) analysis, PRLR was identified as a differentially expressed according to manufacturer’s indications. Cells were co- biomarker in AML cells, at mRNA and surface protein transduced with pLentiCas9-Blast (Addgene #52962), expression levels. Inhibition of this receptor decreased and selected with puromycin (Sigma-Aldrich) and blasti- the clonogenicity and engraftment potential. PRLR cidin (Sigma-Aldrich). Clones were selected based on the expression levels directly correlated with resistance to surface expression of PRLR by flow cytometry. cytarabine, the backbone chemotherapeutics used in AML, and the induction of a senescence-like pheno- In vivo studies type. The genetic validation of PRLR as a therapeutic tar- 6/8 week-old NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ (NSG, get supports the further development of PRLR-targeted Jackson Laboratories) mice were myeloablated with pharmacological approaches for AML. intraperitoneal 30 mg/kg busulfan (Sigma-Aldrich). Table 1 AML patients’ information AML Type Sex Age WHO subtype BPB (%) BBM (%) Karyotype Additional molecular features Risk CR #1 BM M 49 Therapy related AML 42 26 46–47, XY, del(5)(q22q34),del(6) None Adv Yes (q22q25),del(7)(q22q23),-8,- 9,add(11)(q23),+i(11)(q11),- 16,+mar1,+mar2,+mar3[cp8] #2 PB M 64 AML without maturation 1 35 46,XY[20] None Int No #3 N/D F 45 AML treatment-related 22 44 46,XX[20] None Int No #4 PB F 51 AML with mutated NPM1 26 63 N/D NPM1 mut Fav No #5 N/D M 61 AML with mutated NPM1 33 72 46,XY[20] NPM1 mut, DNMT3A mut Fav Yes #6 N/D M 51 AML with myelodysplasia-related changes 88 93 46,XY,-5,+mar1[4]/46,XY,-5,+mar2[4] None Adv Yes Cuesta‑Casanovas et al. Cancer Cell International #7 N/D F 54 AML with myelodysplasia-related changes 47 83 48,XX,+add(13)(q34),+add(13)(q34) None Int Yes [8]/46,XX[5] #8 LA F 44 AML with mutated NPM1 76 80 N/D NPM1 mut Fav No #9 LA F 67 MPAL, T/Myeloid, NOS 53 30 46,XX[21] None Int Yes #10 PB F 80 AML with mutated NPM1 53 95 46,XX[20] FLT3-ITD, NPM1 mut Int No (2023) 23:97 #11 BM M 41 AML with t(3;3)(q21.3;q26.2); GATA2. 57 73 46,XY,t(3;3)(q21;q26)[1]/45,X,-Y,t(3;3) None Adv Yes MECOM (q21;q26)[19] #12 BM M 45 N/D N/D 0 46,XY[20] FLT3-ITD Adv No #13 BM M 71 AML with mutated NPM1 0 3 Normal NPM1 mut Fav No #14 BM M 27 N/D 0 2 46,XY[20] None Int No #15 BM M 27 AML with myelodysplasia-related changes 3 34 46,XY[37] FLT3-ITD Int No #16 PB M 69 AML with myelodysplasia-related changes 23 3 43,XY,-4,add(4)(q?35),del(5)(q13q33),- None Adv N/D 7,der(12)t(12;?14)(p?12;q?12),-14,-15,del(20) (q11q13),-21,+mar1,+mar2[22] #17 BM F 45 AML with myelodysplasia-related changes 0 20 46,XX[20] None Int No #18 BM F 82 AML with mutated NPM1 79 83 46,XX,der(12)t(1;12)(q21;q24.3)[4]/46,XX[12] NPM1 mut Fav Yes #19 BM M 78 N/D 0 0 Normal None Int No #20 BM M 73 N/D 0 27 Complex None Adv No #21 BM M 70 N/D 0 5 46,XY[20] None Int No #22 BM M 78 N/D 0 0 Normal None Int No #23 BM M 27 N/D 0 6 46,XY[20] None Int No #24 BM M 47 APL with PML-RARA​ 71 0 46,XY[20] N/D Fav No #25 BM M 71 AML with mutated NPM1 N/D 0 Normal NPM1 mut Fav No #26 BM M 68 AML with myelodysplasia-related changes N/D N/D N/D N/D N/D N/D #27 N/D N/D N/D N/D N/D N/D N/D N/D N/D N/D #28 PB M 75 N/D N/D 98 46,XY,inv(3)(p25q21),del(11)(p11.1p15) None N/D N/D [cp14] #29 BM M 64 AML with mutated NPM1 56 78 46,XY[20] NPM1 mut Fav No Page 3 of 18 Table 1 (continued) AML Type Sex Age WHO subtype BPB (%) BBM (%) Karyotype Additional molecular features Risk CR #30 PB F 47 AML with inv(3)(q21.3q26.2); GATA2, 3 23 46,XX,inv(3)(q21q26.2)[17]/46,XX[30] None Adv Yes MECOM #31 BM F 90 AML with myelodysplasia-related changes 92 77 42,XX,del(5)(q22q34),del(7)(q22q32),add(8) TP53 mut Adv N/D (q24),-15,-16,add(17)(p13),-18,-19,-20,-21,- 22,-22,+r(?),+mar[cp18]* #32 PB F 90 AML with myelodysplasia-related changes 92 77 42,XX,del(5)(q22q34),del(7)(q22q32),add(8) TP53 mut Adv N/D (q24),-15,-16,add(17)(p13),-18,-19,-20,-21,- 22,-22,+r(?),+mar[cp18]* #33 LA F 63 AML with mutated NPM1 100 94 N/D FLT3-ITD, DNMT3A mut, IDH2mut Int No #34 N/D N/D N/D N/D N/D N/D N/D N/D N/D N/D Cuesta‑Casanovas et al. Cancer Cell International #35 N/D N/D N/D N/D N/D N/D N/D N/D N/D N/D #36 PB F 64 AML without differentiation 48 73 46,XX[30] MLL-PTD, IDH2 mut, DNMT3A mut Adv No #37 PB F 69 AML with mutated NPM1 68 66 47,XX,+8[20] NRAS mut, RUNX1 mut, CEBPA mut Adv No #38 BM M 62 AML with inv(3)(q21.3q26.2); GATA2, 18 61 46,XY,inv(3)(q21q26.2)[8]/46,XY[20] DNMT3A mut, IDH1 mut Adv No MECOM (2023) 23:97 #39 PB F 77 AML with mutated NPM1 95 85 N/D FLT3-ITD, TET2 mut Int N/D #40 PB M 62 AML with mutated RUNX1 49 81 6,XY,del(7)(?)[19]/46,XY[1] None Adv No #41 N/D N/D N/D N/D N/D N/D N/D N/D N/D #42 PB M 63 AML with myelodysplasia-related changes 43 30 N/D TP53 mut Adv Yes #43 BM M 27 N/D 0 1 46,XY[20] BCR/ABL absence Int No #44 BM M 57 N/D 43 33 46,XY[5] None Int No #45 BM M 71 AML with mutated NPM1 N/D 1 Normal NPM1 mut Fav No #46 BM M 73 N/D 0 27 Complex None Adv No #47 BM M 43 N/D 29 6 46,XY[30] None Int Yes #48 BM M 78 AML with myelodysplasia-related changes 0 N/D N/D N/D N/D N/D #49 LA M 47 AML with mutated NPM1 23 48 Normal DNMT3A mut, IDH2 mut, TP53 mut Adv No #50 LA M 40 AML with mutated NPM1 96 93 N/D IDH1 mut, NPM1 mut, FLT3-ITD Adv N/D #51 LA M 37 AML with inv(16)(p13.1q22); CBFB-MYH11 98 91 47,XY,+9,inv(16)(p13;q22),der(17)t(11;17) CEPBA mut, FLT3 mut, WT1 mut Fav No (q13;q25)[10] #52 LA F 21 AML NOS, monoblastic 98 98 46,XX,t(1;20;11’) KMT2A rearrangement Adv No (p36q21;q11.2;q23),t(10;inv(11)(q13q23)) (p12;q13)[20] #53 LA M 65 AML with mutated NPM1 62 80 46,XY[20] NPM1 mut, FLT3-ITD Fav No #54 PB M 71 AML with mutated NPM1 86 84 N/D NPM1 mut, FLT3-ITD Int N/D #55 PB M 55 AML with mutated NPM1 85 76 46,XY[20] NPM1 mut, IDH1 mut, DNMT3A mut, TET2 Fav N/D mut, PTPN11 mut Page 4 of 18 Cuesta‑Casanovas et al. Cancer Cell International Table 1 (continued) AML Type Sex Age WHO subtype BPB (%) BBM (%) Karyotype Additional molecular features Risk CR #56 PB F 68 AML with mutated NPM1 54 66 46,XY[20] NPM1 mut, DNMT3A mut, FLT3-TKD, IDH1 Fav No mut, NRAS mut (2023) 23:97 #57 PB + BM F 74 AML with biallelic mutation of CEBPA 20 23 46,XX[20] CEBPA, TET2 Fav No #58 PB F N/D N/D N/D N/D N/D N/D N/D N/D #59 PB M 84 AML with myelodysplasia-related changes 84 66 46,XY,del(20)(q11q13)[5]/48,XY,+8,del(20) JAK2 mut, IDH2 mut, SRSF2 mut, ETV6-MN1 Int N/D (q11q13),+21[15] rearrangement #60 PB + BM M 72 AML with mutated NPM1 16 38 46,XY[20] NPM1 mut, FLT3-ITD, IDH2 mut, RUNX1 mut, Fav N/D WT1 mut, IDH1 mut, DNMT3A mut #61 LA N/D N/D N/D N/D N/D N/D N/D N/D N/D AML patients’ information Age in years; Adv adverse, AML acute myeloid leukemia, BBM blasts in bone marrow, BM bone marrow, BPM blasts in peripheral blood, CR chemoresistance, F female, Fav favorable, Int intermediate, M male, N/D not determined, PB peripheral blood, Risk cytogenetic risk group according to 2017 ELN recommendations [53], WHO subtype WHO subtype according with the 2016 WHO classification [15] Page 5 of 18 Cuesta‑Casanovas et al. Cancer Cell International (2023) 23:97 Page 6 of 18 1 × ­106 MonoMac-1 cells transduced with pLL-EF1α- Interestingly, not only AML, but also the AML-related rFLuc-T2A-GFP (rFLuc, System Bioscience #LL410PA-1) myeloid neoplasia Myelodysplastic syndromes (MDS) were intravenously injected. Leukemia burden was deter- overexpress PRLR on the cell surface, observing a wide mined by bioluminescence (IVIS Lumina III In vivo range of expression levels, at a comparable frequency rate Imaging System, Perkin Elmer). Images were analysed (AML: 32.278 ± 3.591%; MDS: 32.947 ± 3.708%). Similar with Aura Imaging software (v3.2). For competition results were obtained at RNA level (Additional file 1: Fig. assays, an equivalent number of MonoMac-1 cells trans- S1B). PRLR positivity was residual in both healthy blood duced with pULTRA-Smurf (Addgene #48974) as a con- donor cells sources analysed: peripheral blood mononu- trol and test-transduced MonoMac-1 were intravenously clear cells (PB-MNCs) and hematopoietic progenitor/ injected. Engraftment was determined by flow cytometry stem cells (lineage-depleted umbilical cord blood cells) at day 14. (Fig. 1B). Significant differences were also observed at a total PRLR protein level (Fig. 1C). Remarkably, PRLR was Luciferase assays expressed on several low-frequent hematopoietic sub- CISH reporter plasmid was kindly supplied by Dr. Clev- types, such as CD11c-positive dendritic cells or CD14- enger [21]. pGL4.33[luc2P/SRE/Hygro] (MEK-Erk positive monocytes/macrophages (Fig. 1D). A surface response element, Promega #E1340); pGL4.29[luc2P/ protein expression screening in a wide panel of cell lines CRE/Hygro] (cAMP response element, Promega derived from different hematological neoplasias revealed #E8471); 4xM67 pTATA TK-Luc (Stat3 response element, that PRLR is expressed on myeloid-associated tumors, Addgene #8688) are commercially available. pRL-SV40 mainly AML (U-937, SKM-1, MonoMac-1), while this (Addgene #E2231) was used as transfection efficiency receptor is generally absent in B or T lymphoid leukemias control. (Fig. 1E). As PRL can act as a survival and proliferation fac- β‑Galactosidase staining tor [22, 23], several AML cell lines were challenged 5 × ­105 cells/mL were attached to poly-l-lysine (50 µg/ with the natural ligand PRL and the engineered antago- mL) (Sigma)-coated 96-well plates and stained with the nist del 1-9-G129R-hPRL (G129R) [24]. Cell viability β-Galactosidase Staining Kit (Cell Signal) following the remained unchanged after a 72 h-high dose-treatment manufacture’s recommendations. Samples were observed with PRL or its antagonist G129R, regardless the PRLR at 40× and images were analysed and quantified with surface expression level (Fig. 2A), as previously demon- ImageJ (v1.8.0_172). strated [12]. As primary AML cells expressed higher lev- els of PRLR on the surface than AML cell lines (Fig. 1B, Results E), the long isoform of PRLR (PRLR wt) was ectopically As the axis PRL:PRLR regulates key processes in several expressed on SKM-1, MonoMac-1 and HL-60 AML cell solid tumors, the role of this signaling pathway was eval- lines. While SKM-1 expressed intermediate surface levels uated in the transformation events associated to AML. of endogenous PRLR, MonoMac-1 and HL-60 expressed As a first approach, several AML and healthy blood sub- low levels (Fig. 1E). Interestingly, enforced expression of populations were interrogated for the expression of a PRLR in SKM-1 resulted in a notable overexpression of PRLR-associated gene signature. The different AML cell the receptor; while MonoMac-1 and HL-60 presented fractions clustered together, suggesting that the activa- more modest levels, despite transduction (Fig. 2B and tion status of this signaling pathway was similar in this Additional file 1: Fig. S2A). In contrast to the parent cell population, as compared to healthy blood subpopula- lines, PRLR wt-overexpressing MonoMac-1 and HL-60 tions (Additional file 1: Fig. S1A), indicating that AML cells responded to the presence of PRL by inducing cell cells shared a common altered gene signature associated proliferation. However, the antagonist G129R spared to PRLR. In concordance with this preliminary analysis AML cell lines in terms of proliferation rate (Fig. 2C). To and previous reports [12], PRLR mRNA was differentially further explore the effect of PRL in patient cells, primary expressed in AML (2.544 ± 0.066) samples as compared AML samples (n = 19) from each representative subgroup to healthy blood donors (2.053 ± 0.174), whereas no sig- were treated with PRL and G129R and no significant nificant differences were found in the natural ligand PRL changes were observed in terms of cell viability after a gene expression, at an autocrine level (Fig. 1A). Next, 72 h treatment (Fig. 2D). Similarly, either PRL or G129R the surface expression of the protein was determined by failed to affect survival and/or proliferation of healthy flow cytometry, as a surrogate of the functional recep- donor total PB-MNCs (n = 7), CD13-positive myeloid, tor. Similar to the mRNA data, primary patient AML CD3-positive T cells or CD19-positive B cells (Fig. 2E). samples expressed higher levels of the PRLR in plasma Since relapse and chemorefractoriness are the main clini- membrane (Fig. 1B), within the blast gate (SSC-CD45dim). cal challenges encountered in AML and both properties Cuesta‑Casanovas et al. Cancer Cell International (2023) 23:97 Page 7 of 18 Fig. 1 Prolactin receptor is overexpressed in hematological malignancies. A PRLR (211917_s_at) and PRL (205445_at) mRNA expression in healthy blood donor cells (grey) and AML cells (red) from the GSE13159 database. Ns, not significant; **p < 0.01 (unpaired t test). B PRLR surface expression in ­CD45+ cell population of healthy donors (PB-MNCs n = 10, and HSPC n = 3, grey), AML (n = 59, red), and MDS (n = 39, blue) patient cells were analysed by flow cytometry *p < 0.05 (one-way ANOVA, Dunnett’s multiple comparison test). C PRLR protein expression in AML samples (n = 2) and healthy blood donor cells (n = 2) were analysed by Western Blot; GAPDH was used as loading control. D PRLR surface expression in different hematopoietic subpopulations of healthy blood cells (n = 10, grey) and in C­ D45+ cell population of neoplastic blood cells [AML (n = 59, red) and MDS (n = 39, blue)] were analysed as in B. E PRLR surface expression in a panel of cell lines derived from hematological malignancies (n = 19) and classified by phenotype [myeloid (orange), lymphoid B (green) and lymphoid T (blue)] analysed as in B. Bars represent the mean ± SEM (duplicates) (See figure on next page.) Fig. 2 Inhibition of PRLR decreases AML clonogenicity, sparing the healthy counterpart. A AML cell lines (HL-60, MonoMac-1, SKM-1, KG-1, U-937, Kasumi-1) (n = 6) were treated with the vehicle (grey), PRL (blue) or G129R (red) at 500 ng/mL for 72 h and cell viability was analysed by flow cytometry (one-way ANOVA, Dunnett’s multiple comparison test, triplicates). B PRLR-transduced MonoMac-1 (MM), HL-60 and SKM-1 cells were validated by qPCR (mRNA expression, bars represent ­2−ΔCt ± SEM) and Western Blot with GAPDH as loading control (protein, a representative membrane is shown, and arrows showed the interesting band). Grey, control; Orange, PRLRwt-transduced cells. ***p < 0.001; ****p < 0.0001 (two-way ANOVA, Šidák’s multiple comparison test, triplicates). C Parental and PRLR-overexpressing MonoMac-1 (MM), HL-60 and SKM-1 cells were treated during 4 days with the vehicle (grey), PRL (blue) or G129R (red) at 500 ng/mL and proliferation (DiI mean fluorescence intensity) was assessed by flow cytometry. **p < 0.01; ***p < 0.001 (two-way ANOVA, Šidák’s multiple comparison test, duplicates). D AML patient cells (n = 19) were treated with the vehicle (C, grey), PRL (blue) or G129R (red) at 250 and 500 ng/mL and cell viability was assessed after 72 h in the C­ D45+ cell population by flow cytometry. **p < 0.01 (non-parametric one-way ANOVA, Holm-Šidák’s multiple comparison test). E Healthy donor PB-MNCs (n = 7) were treated with the vehicle (grey), PRL (blue) or G129R (red) at 250 and 500 ng/mL and cell viability was assessed after 72 h in C­ D45+ total blood population, C ­ D13+ myeloid cells, ­CD3+ T cells and C ­ D19+ B cells by flow cytometry. Statistics were done as in D. F AML patient cells (n = 10) were treated with the vehicle (C, grey), PRL (blue) or G129R (red) at 250 and 500 ng/mL for 18 h, colonies were counted after 14 days based on cellularity and morphology. Statistics were done as in D. **p < 0.01. G Lineage-depleted umbilical cord blood cells (n = 4) were treated with the vehicle (grey), PRL (blue) or G129R (red) at 250 or 500 ng/mL for 18 h, colonies were counted after 14 days based on cellularity and morphology. Statistics were done as in D. H PRLR expression of parental U-937 (grey) and three different PRLR-directed CRISPR/Cas9-transduced clones (orange) were phenotyped for PRLR surface expression by flow cytometry (one-way ANOVA, Dunnett’s multiple comparison test, triplicates). I Parental cells (grey) and transduced clones (orange) were cultured in a semisolid medium enriched with cytokines; colonies were counted after 7 days based on cellularity and morphology. **p < 0.01 (one-way ANOVA, Dunnett’s multiple comparison test, n = 6). In all graphs, bars represent the mean ± SEM of each individual experiment Cuesta‑Casanovas et al. Cancer Cell International (2023) 23:97 Page 8 of 18 Fig. 2 (See legend on previous page.) reside in the most primitive leukemic cell fraction [25, However, a 33% reduction was observed in the number 26], clonogenicity was evaluated upon treatment, as a of colonies formed when AML samples were treated with surrogate assay for stem cell functionality. The potential the G129R. Interestingly, all tested AML samples were of primary AML samples (n = 10) to generate blast colo- responsive to the treatment (Fig. 2F). In concordance nies in a semisolid medium, in the presence of instructive with the lack of effect in cell viability (Fig. 2E), the fre- cytokines, remained unaffected in the presence of PRL. quency of both total CFUs and each colony subtype were Cuesta‑Casanovas et al. Cancer Cell International (2023) 23:97 Page 9 of 18 equivalent in lineage-depleted umbilical cord blood cells PRL wt induced the transcriptional activation of Stat5- (n = 4), regardless of treatment (Fig. 2G). To validate the regulated genes (CISH promoter activity), in contrast to role of PRLR in clonogenicity, the U-937 AML cells, that PRL mut that failed to induce any transcriptional activity express the highest level of this receptor in the surface (Fig. 3B). In concordance with the in vivo data (Fig. 3A) (Fig. 1E), were edited with a CRIPR/Cas9-based lentiviral and the colony-forming assay (Fig. 2F), PRL mut-express- system to knock out PRLR up to 2.57-fold (Fig. 2H). In ing AML cells displayed a reduced clonogenic capacity, concordance with the results in Fig. 2F, PRLR downregu- while PRL wt induced little (if any) differences in terms lation resulted in a dramatic decrease of the colony for- of blast colony formation (Fig. 3C). Alternative splic- mation capacity (Fig. 2I). Thus, PRLR-mediated signaling ing-generated PRLR isoforms differ in the intracellular seemed to be relevant for self-renewal and differentiation domain and in their signaling transduction upon activa- capacities, whereas the importance in survival and prolif- tion. The wildtype or long isoform (PRLR wt, 80–85 kDa), eration was limited. predominant in physiological conditions, activates the Next, the role of PRL in a relevant AML in vivo model Jak2/Stat5 pathway upon PRL ligand recognition. In was evaluated. Adult NSG mice were pharmacologically contrast, the short isoform (PRLR short, 45 kDa), whose conditioned and transplanted with rFLuc-AML Mono- physiological distribution is controversial, behaves as Mac-1 cells. Mice were treated for 5 days with 0.2 mg/ a dominant negative in some cellular context [29]. Both kg PRL or G129R, once the engraftment was established. PRLR wt and PRLR short were transduced into a P ­ RLRint The engraftment levels were monitored by biolumines- AML cell line, MonoMac-1, to further genetically vali- cence throughout the length of the experiment (schemed date the role of PRLR-mediated signaling in the engraft- in Additional file 1: Fig. S2B). As endogenous murine ment potential (Additional file 1: Fig. S2C). As expected, PRL level in females is greatly increased, compared to Stat5-mediated transcriptional activation was induced males [27], engraftment was analysed within each gender. by PRL when PRLR wt was present, in a dose–response Moreover, murine PRL not only has more than 50-fold manner, and no PRL-induced response was observed in lower potency towards human PRLR than human PRL, PRLR short (Fig. 3D), confirming the dominant-nega- but also is an effective competitive inhibitor of human tive role of the PRLR short in this system. A competitive PRL [28]. The treatment with human PRL increased the repopulation assay was chosen to evaluate the relative engraftment level in males; however, no significant differ- engraftment potential of the transduced AML cells (GFP- ences were observed in females (Fig. 3A), probably due tagged PRLR/PRL isoforms) against the parental leuke- to the differences in murine PRL plasma levels between mic cells (smurf-expressing empty vector). Whereas the genders, suggesting that high endogenous mPRL may be leukemic repopulation capacity of AML remained unaf- disrupting the action of exogenous hPRLR. In concord- fected by the enforced expression of PRL wt or PRLR wt, ance, the treatment with the antagonist G129R decreased PRL-neglected signaling AML cells (PRL mut or PRLR the engraftment potential of AML, especially in females short) consistently underperformed, both in bone mar- (Fig. 3A). To limit the effect of the endogenous murine row and spleen (Fig. 3E), supporting the role of PRLR- PRL [24] in the xenotransplantation assays, AML cell mediated signaling in the leukemia engraftment potential lines were stably transduced with the wildtype and the in vivo. Interestingly, the migration capacity to periph- mutant G129R PRL (PRL wt and PRL mut, respectively); eral organs of both PRL mut- and PRLR short-expressing thus, an autocrine effect was expected without any inter- AML cells was severely diminished, as observed in the species interference. As a validation of the system [24], leukemia levels in spleen. In concordance, PRL acted as a (See figure on next page.) Fig. 3 PRLR signaling modulates the in vivo AML regeneration capacity. A rFluc-transduced MonoMac-1 (MM) were intravenously injected into adult conditioned NSG mice (n = 34). At day 4, mice were treated 5 days with the vehicle (grey), PRL (blue) or G129R (red) at 0.2 mg/kg. Engraftment was detected by bioluminescence. A representative bioluminescence image at day 13 is presented. *p < 0.05; **p < 0.01; ****p < 0.0001 (one-way ANOVA, Tukey’s multiple comparison test). B Activity of the Stat5-reporter CISH was evaluated in the presence of the empty vector (grey), PRL wt (orange) or PRL mut (blue) PRLR-expressing HEK293T cells. ***p < 0.001 (one-way ANOVA, Tukey’s multiple comparison test, triplicates). C Parental (grey) and PRL wt- (orange) or PRL mut- (blue) transduced MonoMac-1 (MM, n = 7) and HL-60 (n = 5) were cultured in methylcellulose and colonies were counted after 7 days. *p < 0.05; **p < 0.01 (non-parametric one-way ANOVA, Tukey’s multiple comparison test). D Activation of Stat5 reporter CISH was assessed in PRLR wt- (orange) and PRLR short- (blue) transfected HEK293T cells treated with increasing doses of PRL. *p < 0.05; **p < 0.01 (one-way ANOVA, Šidák’s multiple comparison test, triplicates). E An equivalent number of pSmurf-transduced MonoMac-1 (blue), and empty control or PRL wt/PRL mut (n = 28) or PRLR wt/PRLR short (n = 31) MonoMac-1 cells (yellow) were intravenously injected into adult conditioned NSG mice. After 15 days, ­CD45+ AML cells from bone marrow and spleen were analysed by flow cytometry. *p < 0.05; ***p < 0.001; ****p < 0.0001 (non-parametric one-way ANOVA, Tukey’s multiple comparison test). F Migration capacity was determined using a transwell chamber assays. Growth medium containing 10% FBS was placed in the lower chamber and MonoMac-1 (MM) cells were incubated with the vehicle (grey) or PRL at 500 ng/mL (blue). Cells were allowed to migrate into the low chamber for 48 h. Migrated cells were quantified by flow cytometry. *p < 0.05 (unpaired t test, triplicates). In all graphs, bars represent the mean ± SEM Cuesta‑Casanovas et al. Cancer Cell International (2023) 23:97 Page 10 of 18 Fig. 3 (See legend on previous page.) Cuesta‑Casanovas et al. Cancer Cell International (2023) 23:97 Page 11 of 18 chemoattractant for AML cells in transwell cell migration S3B). In order to explore the dependency of AML cells assays (Fig. 3F). on the Jak/Stat pathway in the context of the PRLR, the To decipher the signaling pathway associated to PRLR ­PRLRint AML cell lines MonoMac-1 and the ­ PRLRlow in this cellular context, key response elements were inter- AML cell lines HL-60 were treated with specific Stat3 rogated for their response to PRL. As expected, and and Stat5 inhibitors. Sensitivity to the Stat3 inhibitor unlike Stat3 (pTATA), Jak2/Stat5 is activated upon PRL was equivalent in both cell lines (EC50 = 133 µM and recognition, inducing Stat5-mediated gene expression 146 µM, respectively) (Fig. 4E), suggesting that PRLR is (CISH) (Fig. 4A), activating Stat5 by phosphorylation not involved in this signaling pathway, in concordance (Fig. 4B and Additional file 1: Fig. S3A) and translocating with the lack of transcriptional activity upon PRL activa- phospho-Stat5 to the nucleus (Figs. 4C). Despite produc- tion (Fig. 4A). The cytotoxic effect induced by the inhi- tive recognition of the PRLR, G129R behaved as an inac- bition of Jak2 was also similar in both AML cell lines tive antagonist (Fig. 4D), suggesting that G129R acted (Fig. 4E), discarding a major contribution of Jak2 to the as a competitive PRL antagonist [30], rather than as an PRLR-mediated signaling. In contrast, a lower EC50 was inverse agonist. Next, the potential regulation of MEK/ observed in P ­ RLRint AML cells (MonoMac-1), as com- Erk- (CRE reporter) and cAMP-mediated (SRE reporter) pared to ­PRLRlow AML cells (HL-60), when challenged signaling pathways was evaluated and no modulation with the Stat5 (30 µM vs. 213 µM) inhibitor (Fig. 4E and was observed in response to PRL (Additional file 1: Fig. Additional file 1: Fig. S3C). Thus, the sensitivity to the Fig. 4 In AML cells, PRLR induces signaling through Jak2/Stat5. A CISH and pTATA reporters were transfected in HEK293T cells and treated with the vehicle (grey), PRL (blue) or G129R (red) at 500 ng/mL, in starving conditions. The luciferase activity (normalized against renilla signal) is shown in a representative replicate. **p < 0.01 (two-way ANOVA, Tukey’s multiple comparison test, triplicates). B PRLR-transduced MonoMac-1 (MM PRLR wt) cells were treated with the vehicle, PRL (blue) or G129R (red) at 250 and 500 ng/mL for 2 h and total protein were extracted. Total and phosphorylated Stat3 and Stat5 were analysed by Western Blot; GAPDH was used as loading control and a representative membrane is shown (n = 3). C PRLR-transduced and control MonoMac-1 (MM) were treated for 2 h with the vehicle, PRL or G129R at 500 ng/mL. Nuclear protein fraction was extracted and total and phosphorylated Stat5 were analysed by Western Blot; Lamin B1 was used as loading control and a representative membrane is shown (n = 3). D PRLR wt-expressing HEK293T cells were transfected with the CISH reporter and treated with increasing doses of PRL (blue) or G129R (red) for 24 h. Luciferase activity is shown. *p < 0.05 (paired t test, triplicates). E MonoMac-1 and HL-60 cells were treated with an inhibitor of Stat3, Stat5 and Jak2 for 72 h at different doses and viability was analysed by flow cytometry. EC50 value is shown in the table, triplicates. F PRLR-transduced and control MonoMac-1 (MM) cells were treated for 72 h with a Stat5 inhibitor (10 and 50 µM) alone (green) or in combination with 500 ng/mL PRL (squared green). Viability was assessed by flow cytometry. *p < 0.05 (one-way ANOVA. Dunnet’s multiple comparison test, triplicates). G Stat3 (209991_at) and Stat5 (203010_at) mRNA expression in healthy donors (grey) and AML patients (red) from GSE13159 database. Ns, not significant; ****p < 0.0001 (unpaired t test). In all graphs, bars represent the mean ± SEM Cuesta‑Casanovas et al. Cancer Cell International (2023) 23:97 Page 12 of 18 Stat5 inhibitor was related to the PRLR expression lev- senescence phenotype in relationship to PRLR expres- els. Indeed, the cytotoxicity induced by the Stat5 inhibi- sion, X-gal-based β-galactosidase staining was exam- tor was partially reverted in the presence of PRL (Fig. 4F), ined as a canonical biomarker for this state [32]. As in AML cell lines, suggesting that Stat5-activated sign- demonstrated by the increase of senescence-associated aling pathway was regulated by PRL. Moreover, Stat5 β-gal (SA-β-gal) staining (Fig. 6A), PRLR overexpres- mRNA was overexpressed in AML patient samples sion induced a senescence-like phenotype; although no (GSE13159 [16, 17]), as compared to healthy donors, consistent differences were observed in the proliferation while Stat3 mRNA levels are equivalent in both popula- rate (Additional file 1: Fig. S5A), p21 or p16 upregulation tions (Fig. 4G), highlighting the pivotal role of Stat5 in (Fig. 6B and Additional file 1: Fig. S5B), or cell cycle status the PRLR-mediated leukemic features. (Additional file 1: Fig. S5C), precluding cell cycle arrest To evaluate the role of PRLR signaling during leuke- [33], and suggesting the acquisition of a premature senes- mogenesis, paired diagnosis-relapse AML samples were cence-like phenotype. Interestingly, basal SA-β-gal stain- analyzed for their PRLR and Stat5 mRNA levels to track ing was associated with PRLR expression (Fig. 6C). These the expression of the receptor and the main secondary findings were further validated using a CRISPR/Cas9- messenger through the course of the disease. Stat5 gene based strategy to downregulate the expression of PRLR expression increased in relapse disease, although the (Fig. 2H), reducing the SA-β-gal positivity (Fig. 6D). expression of PRLR mRNA decreased (Fig. 5A); how- The highest the expression of PRLR, the highest the ever, surface expression of the receptor was similar at resistance to cytarabine and the highest the senescence diagnosis and relapse in a limited patient cohort (Addi- marker staining. In AML, induction of senescence as tional file 1: Fig. S4A). Interestingly, sensitivity to cytara- well as tumor survival and chemotherapy persistence are bine, the backbone of most chemotherapeutic regimens dependent on ATR activity. Previous studies showed that for AML, correlated with PRLR surface expression. ATR inhibition synergized with nucleoside analogues, The highest the endogenous expression of PRLR corre- like cytarabine, to eradicate AML in xenotransplantation sponded to the highest resistance to cytarabine treatment mouse models [34]. Indeed, an ATR-dependent senes- (Fig. 5B). While enforced expression of PRL in AML cells cence-like resilient state was demonstrated to be respon- failed to confer chemoresistance to cytarabine, over- sible for AML relapse, at least partially. This resilient expression of PRLR significantly augmented the EC50 state had a variable duration and conferred an optimized of cytarabine in AML cells (Fig. 5C), suggesting that fitness and enhanced repopulation capacity [25]. In con- the PRLR-mediated signaling was responsible for over- cordance, the presence of elimusertib, an ATR inhibitor, coming the cytotoxicity effect produced by cytarabine. increased the sensitivity to PRLR-overexpressing AML Besides, acquired resistance to cytarabine in AML cell cells to cytarabine treatment, as compared to the parental lines by long-term drug treatment (Additional file 1: Fig. cell line (Fig. 6E), suggesting that ATR played a key role S4B) resulted in the upregulation of PRLR at surface and in the chemoresistance observed upon enforced expres- total protein expression levels (Fig. 5D). At a gene expres- sion of PRLR wt. sion level, overexpression of PRLR induced the down- regulation of key cytarabine transporters (hENT2 and Discussion hENT3) and cytarabine activators (dCK and NDK), while Refractory and relapsed (R/R) AML is still highly chal- inactivators such as PN-I, CDA and NT5E were upregu- lenging from a clinical point of view, as poor response lated (Fig. 5E). Thus, the expression of PRLR induced a rates and, consequently, unfavorable prognosis are chemoresistance-associated gene signature [31]. expected. Given the need for more effective treatment Recently, chemoresistant AML cells have been asso- options in R/R AML, improving the efficacy of salvage ciated with a senescence-like phenotype, maintaining chemotherapy by combining targeted therapy or optimiz- the leukemia repopulation potential [25]. To explore the ing drug delivery had been explored with limited success (See figure on next page.) Fig. 5 PRLR confers chemoresistance to AML. A PRLR and Stat5a mRNA expression at diagnosis (blue) and relapse (red) in AML patients from GSE66525 database. *p < 0.05 (unpaired t test). B AML cell lines were treated for 72 h with cytarabine (AraC) at the indicated doses. Viability was assessed by flow cytometry and a representative result was shown. The EC50 values were correlated with the PRLR surface expression analysed by flow cytometry. *p < 0.05 (simple linear regression, triplicates). Green, HL-60; Red, SKM-1; Blue, U-937; Black, MonoMac-1 (MM); Brown, KG-1. C PRLR-transduced (orange), PRL-transduced (orange) and control (black) MonoMac-1 (MM) cells were treated with AraC for 72 h. Viability was analysed by flow cytometry and representative results were shown (n = 3). D PRLR expression in AraC-resistant (AraC R, red) and the parental (grey) cell lines HL-60 and SKM-1 were analysed by flow cytometry (protein surface expression) and western blot (total protein level). *p < 0.05; **p < 0.01 (paired t test, triplicates). E Expression of genes associted to cytarabine transporters (green), activators (blue) and inactivators (yellow and purple) were determined by qPCR in PRLR wt-tranduced HL-60 and SKM-1 cells and compared to parental cells. *p < 0.05 (one-way ANOVA, Dunnett’s multiple comparison test, triplicates). In all graphs, bars represent the mean ± SEM Cuesta‑Casanovas et al. Cancer Cell International (2023) 23:97 Page 13 of 18 Fig. 5 (See legend on previous page.) Cuesta‑Casanovas et al. Cancer Cell International (2023) 23:97 Page 14 of 18 [11]. Despite inconsistencies regarding the response rates with survival/proliferation of AML cells [46]. From an for most of the investigational agents [hypomethylating epidemiology standpoint, leukemia is the second most agents (decitabine or azacitidine) [35], Bcl-2 inhibitor frequent cancer associated to hyperprolactinemia in (venetoclax) [35], tyrosine kinase inhibitors (sorafenib females [13]. Here, the therapeutic potential of PRLR in [36], pazopanib [37], quizartinib [38] or crenolanib AML is corroborated, and associated with Stat5 activa- [39]), or MAPK (binimetinib [40] or trametinib [41]) tion and chemoresistance phenotypes. Abnormal activa- and hedgehog (glasdegib) [42] modulators] and com- tion of Stat5 via phosphorylation is frequently observed binations, durable responses were achieved in certain in AML cells [47], induces proliferation of leukemic cells patients; though it will be important to identify biomark- [48] and desensitizes to certain drugs [49]. A STAT5- ers that predict individual patient response. Nevertheless, associated signature correlated to unfavorable clinical disruptive therapeutic approaches are urgently needed as outcome in AML patients [50]. Indeed, pharmacological the chemotherapy sphere might have reached its limits inhibition of Stat5 reduced leukemia burden in an in vivo and the therapeutic window of most targeted therapies is FLT3-ITD + AML model [51]. Although promising, no fairly narrow. clinical-grade Stat5 small molecule inhibitors have been Senescence, chemoresistance and cancer stemness proven in clinical trials. Clinical experiences from Stat3 are closely interrelated. A senescence-like phenotype is inhibitors suggested low specificity and wide side effects observed in resistant AML cells upon chemotherapy [25, [52], probably due to a poor therapeutic window based 26, 43], associated with an increase in colony-forming on the pleiotropic effect of Stat3. As PRLR is differen- and engraftment potential [25]. After recovery, these tially expressed and activated in leukemic cells, targeting cells are enriched in leukemic stem cells [25, 44], link- directly this receptor might provide a safe therapeutic ing the senescence program to the acquisition of cancer window to further explore this approach in clinical devel- stemness properties. PRLR overexpression correlated opment. LFA102, a clinical-grade PRLR neutralizing anti- to both the senescence-like phenotype, as measured body, failed to provide any clinical benefit in breast and by SA-β-Gal staining, and the resistance to cytarabine. prostate cancer patients [5], although insufficient dose Disruption of the PRLR signaling also affected leukemic exposure and inhibition efficacy in vivo might account stemness, as measured in clonogenic assays. Similar to for these negative results. Despite the dedication of great the chemoresistance-induced senescence resilient pheno- efforts from the scientific community in developing type where ATR catalytic activity enables the induction small molecules with antagonistic effect on PRLR, and of this cellular state, tumor survival and persistence [25], based on the key role of this receptor in leukemogen- ATR was also involved in the PRLR-mediated resistance esis, designing novel therapeutic strategies to specifically to cytarabine. Interestingly, PRLR-induced senescence- disrupt the PRLR-mediated signaling in leukemic cells like phenotype failed to induce cell arrest, as proliferation might provide a promising approach to overcome relapse was not affected, in concordance with previous reports and refractory episodes in AML. [25, 44]. Personalized medicine requires the identification of Conclusions new functional biomarkers, such as PRLR. Expression of PRLR was described in a limited cohort (9 patients) • Prolactin Receptor is differentially expressed in AML of monoblast-like M4 AML patients several decades and its expression positively correlates with chem- ago [45] and autocrine/paracrine PRL was associated oresistance to cytarabine. (See figure on next page.) Fig. 6 PRLR induces a senescence-like phenotype that is associated with chemoresistance. A SA-β-galactosidase staining in PRLR wt-tranduced (orange) and parental control (grey) HL-60 and SKM-1. *p < 0.05 (unpaired t test, n = 4). B p21 mRNA expression in PRLR-tranduced (orange) and parental control (grey) HL-60 and SKM-1 cells analysed by qPCR. Bars represented ­2ΔCt ± SEM of three independent experiments. ***p < 0.001 (unpaired t test). C SA-β-galactosidase staining in HL-60 (n = 4, green), SKM-1 (n = 4, blue), MonoMac-1 (MM, n = 6, grey) and U-937 (n = 3, purple) cells. *p < 0.05 (one-way ANOVA, Tukey’s multiple comparison test). D SA-β-galactosidase staining in CRISPR/Cas9-tranduced (orange) and parental control (grey) U-937 clones (n = 3, one-way ANOVA, Dunnett’s multiple comparison test). E PRLR wt-transduced (orange) and parental control (grey) HL-60 and SKM-1 cells were treated for 48 h with increasing cytarabine doses alone (AraC, black) or in combination with 20 nM elimusertib (red). Viability was analysed by flow cytometry and representative results were graphicated. EC50 value was calculated (left), as well as the fold change in the EC50 (1-(EC50 AraC/cotreatment EC50) (n = 4, right); ***p < 0.001 (unpaired t test). In all graphs, bars represent the mean ± SEM Cuesta‑Casanovas et al. Cancer Cell International (2023) 23:97 Page 15 of 18 Fig. 6 (See legend on previous page.) Cuesta‑Casanovas et al. Cancer Cell International (2023) 23:97 Page 16 of 18 • Inhibition of the PRLR-mediated signaling reduces the data. FG and JE, provided patient samples and clinical data. RMR, designed the clonogenicity and the engraftment potential of the project, analyzed the data, interpreted results and wrote the manuscript. AML in xenotransplantation mouse models. All authors discussed the results and commented on the manuscript. All authors read and approved the final manuscript. Funding This work was funded by the Deutsche José Carreras Leukämie-Stiftung (RMR), Fundació Marató TV3 (Grant Number: 201930-30) (RMR), the Josep Carreras Abbreviations International Leukaemia Foundation (RMR), l’Obra Social “La Caixa”-Fundació AML Acute myeloid leukemia Bancària “La Caixa” (RMR), and CERCA Programme/Generalitat de Catalunya PB-MNC Peripheral blood mononuclear cell (IJC). PLR Prolactin PRLR Prolactin receptor Availability of data and materials For original data or detailed protocols, please contact Dr. Risueño risueno@ carrerasresearch.org. Supplementary Information The online version contains supplementary material available at https://​doi.​ Declarations org/​10.​1186/​s12935-​023-​02944-4. Ethics approval and consent to participate Additional file 1: Figure S1. PRLR signaling pathway is differentially This work was approved by the Ethical Committee of Hospital Clínic (Barce‑ expressed in AML in comparison to HSCs and early precursors. A mRNA lona, Spain) and Hospital Germans Trias i Pujol (Badalona, Spain). Animal work expression profiles related to PRLR signalingwere selected and hierarchy was approved by the Animal Experimentation Ethics Committee of CMCiB grouped without supervision, previous normalization by RMA. AML (Badalona, Spain). patient samples enrich in LSCs were marked. B PRLRmRNA expression in healthy blood donor cells, ALL, AML, CLL, CML, and MDSfrom the Consent for publication GSR13159 database. **p < 0.01; ***p < 0.001; ****p < 0.0001.. Figure S2. Val‑ Not applicable. idation of PRLR-transduced cells and AML murine model. A PRLR surface expression in PRLR wt-transducedand parental controlMonoMac-1, HL-60 Competing interests and SKM-1 cells by flow cytometry. A representative histogram is shown. RMR is a shareholder of Leukos Biotech. JMCM, JMC, RMR are inventors in B rFLuc-transduced MonoMac-1cells were intravenously injected into patents related to leukemia treatments. The other authors have nothing to adult conditioned NSG mice. At day 4, 6, 8, 11 and 13 mice were treated declare. intraperitoneally with the vehicle, PRL or G129R at 0.2 mg/kg. Engraft‑ ment was followed by bioluminescence at day 4, 6, 8, 11, 13 and 14. C PRLR wt-and PRLR short-transducedMonoMac-1cells were validated by Received: 21 February 2023 Accepted: 11 May 2023 qPCRand Western Blotwith GAPDH as loading control and a representa‑ tive membrane is shown. *p < 0.05; ****p < 0.0001. Figure S3. PRLR signals through Jak2/Stat5. A PRLR-transduced MonoMac-1 cellswere treated with the vehicle, PRLor G129Rat different doses for 2 h and total protein lysates were obtained. Phosphorylated and total Stat5 were analysed by References Western Blot; GAPDH was used as loading control. A representative mem‑ 1. Goffin V, Touraine P. The prolactin receptor as a therapeutic target in brane is shown. B CRE and SRE reporters were transfected in HEK293T cells human diseases: browsing new potential indications. Expert Opin Ther and treated with the vehicle, PRLor G129Rat 500 ng/mL. The luciferase Targets. 2015;19:1229–44. activity is shown in a representative replicate. **p < 0.01. C MonoMac-1and 2. Bole-Feysot C, Goffin V, Edery M, Binart N, Kelly PA. Prolactin (PRL) and HL-60were treated with a Stat3, Stat5 and Jak2 inhibitors for 72 h at dif‑ its receptor: actions, signal transduction pathways and phenotypes ferent doses and viability was analysed by flow cytometry. A representa‑ observed in PRL receptor knockout mice. Endocr Rev. 1998;19:225–68. tive result was shown. Bars represent the mean ± SEM. Figure S4. PRLR 3. Thomas LN, Merrimen J, Bell DG, Rendon R, Goffin V, Too CKL. Car‑ confers chemoresistance to AML. A PRLR surface expression analysis in boxypeptidase-D is elevated in prostate cancer and its anti-apoptotic non-refractoryvs. refractoryAML patient samplesby flow cytometry. B Cyt‑ activity is abolished by combined androgen and prolactin receptor arabine-resistantand the parentalcell lines HL-60 and SKM-1were validated targeting. Prostate. 2014;74:732–42. by flow cytometry. A representative result is shown. Figure S5. PRLR over‑ 4. Rouet V, Bogorad RL, Kayser C, Kessal K, Genestie C, Bardier A, et al. expression does not affect AML proliferation capacity. A Proliferationwas Local prolactin is a target to prevent expansion of basal/stem cells in assessed in PRLR-transducedand parental controlHL-60 and SKM-1 cells by prostate tumors. Proc Natl Acad Sci USA. 2010;107:15199–204. flow cytometry. B p16 mRNA expression in PRLR-tranducedand parental 5. Agarwal N, Machiels J-P, Suárez C, Lewis N, Higgins M, Wisinski K, controlHL-60 and SKM-1 cells analysed by qPCR. C Cell cycle analysis of et al. Phase I study of the prolactin receptor antagonist LFA102 in PRLR-transduced and parental control HL-60 and SKM-1 analysed by flow metastatic breast and castration-resistant prostate cancer. Oncologist. cytometry. A representative histogram is shown. Figure S6. A graphical 2016;21:535–536i. summary of the role of PRLR-PRL signaling in AML. 6. Asad AS, Nicola Candia AJ, Gonzalez N, Zuccato CF, Abt A, Orrillo SJ, Additional file 2. Methods. et al. Prolactin and its receptor as therapeutic targets in glioblastoma multiforme. Sci Rep. 2019;9:19578. Additional file 3: Table S1. MDS patients’ information. 7. Levina VV, Nolen B, Su Y, Godwin AK, Fishman D, Liu J, et al. Bio‑ logical significance of prolactin in gynecologic cancers. Cancer Res. 2009;69:5226–33. Acknowledgements 8. Tan D, Chen KHE, Khoo T, Walker AM. Prolactin increases survival and We would like to thank all members of the Risueño laboratory for helpful migration of ovarian cancer cells: importance of prolactin receptor type discussions and technical support. and therapeutic potential of S179D and G129R receptor antagonists. Cancer Lett. 2011;310:101–8. Author contributions 9. Karthikeyan S, Russo A, Dean M, Lantvit DD, Endsley M, Burdette JE. Prol‑ LCC, designed and performed experiments, analyzed the data, and wrote the actin signaling drives tumorigenesis in human high grade serous ovarian manuscript. JDM, JMCM, JMC and ABM, performed experiments and analyzed Cuesta‑Casanovas et al. Cancer Cell International (2023) 23:97 Page 17 of 18 cancer cells and in a spontaneous fallopian tube derived model. Cancer 30. Bernichtein S, Kayser C, Dillner K, Moulin S, Kopchick JJ, Martial JA, et al. Lett. 2018;433:221–31. Development of pure prolactin receptor antagonists. J Biol Chem. 10. Goffin V, Hoang DT, Bogorad RL, Nevalainen MT. Prolactin regulation 2003;278:35988–99. of the prostate gland: a female player in a male game. Nat Rev Urol. 31. Fajardo-Orduña GR, Ledesma-Martínez E, Aguiñiga-Sánchez I, Mora- 2011;8:597–607. García MDL, Weiss-Steider B, Santiago-Osorio E. Inhibitors of chemore‑ 11. Koenig K, Mims A. Relapsed or primary refractory AML: moving past MEC sistance pathways in combination with Ara-C to overcome multidrug and FLAG-ida. Curr Opin Hematol. 2020;27:108–14. resistance in AML. A mini review. Int J Mol Sci. 2021;22:4955. 12. Abdelbaset-Ismail A, Borkowska S, Janowska-Wieczorek A, Tonn T, Rodri‑ 32. Dimri GP, Lee X, Basile G, Acosta M, Scott G, Roskelley C, et al. A biomarker guez C, Moniuszko M, et al. Novel evidence that pituitary gonadotropins that identifies senescent human cells in culture and in aging skin in vivo. directly stimulate human leukemic cells-studies of myeloid cell lines and Proc Natl Acad Sci USA. 1995;92:9363–7. primary patient AML and CML cells. Oncotarget. 2016;7:3033–46. 33. Caliò A, Zamò A, Ponzoni M, Zanolin ME, Ferreri AJM, Pedron S, et al. Cel‑ 13. Berinder K, Akre O, Granathand F, Hulting AL. Cancer risk in hyperprol‑ lular senescence markers p16 INK4a and p21 CIP1/WAF are predictors of actinemia patients: a population-based cohort study. Eur J Endocrinol. hodgkin lymphoma outcome. Clin Cancer Res. 2015;21:5164–72. 2011;165:209–15. 34. Fordham SE, Blair HJ, Elstob CJ, Plummer R, Drew Y, Curtin NJ, et al. Inhibi‑ 14. Hatfill SJ, Kirby R, Hanley M, Rybicki E, Böhm L. Hyperprolactinemia tion of ATR acutely sensitizes acute myeloid leukemia cells to nucleoside in acute myeloid leukemia and indication of ectopic expression of analogs that target ribonucleotide reductase. Blood Adv. 2018;2:1157–69. human prolactin in blast cells of a patient of subtype M4. Leuk Res. 35. Aldoss I, Yang D, Aribi A, Ali H, Sandhu K, Al Malki MM, et al. Efficacy of 1990;14:57–62. the combination of venetoclax and hypomethylating agents in relapsed/ 15. Arber DA, Orazi A, Hasserjian R, Thiele J, Borowitz MJ, Le Beau MM, et al. refractory acute myeloid leukemia. Haematologica. 2018;103:e404–7. The 2016 revision to the World Health Organization classification of 36. Ravandi F, Alattar ML, Grunwald MR, Rudek MA, Rajkhowa T, Richie MA, myeloid neoplasms and acute leukemia. Blood. 2016;127:2391–405. et al. Phase 2 study of azacytidine plus sorafenib in patients with acute 16. Haferlach T, Kohlmann A, Wieczorek L, Basso G, Te Kronnie G, Béné MC, myeloid leukemia and FLT-3 internal tandem duplication mutation. et al. Clinical utility of microarray-based gene expression profiling in the Blood. 2013;121:4655–62. diagnosis and subclassification of leukemia: report from the interna‑ 37. Kessler T, Koschmieder S, Schliemann C, Crysandt M, Mikesch JH, von tional microarray innovations in leukemia study group. J Clin Oncol. Stillfried S, et al. Phase II clinical trial of pazopanib in patients with acute 2010;28:2529–37. myeloid leukemia (AML), relapsed or refractory or at initial diagno‑ 17. Kohlmann A, Kipps TJ, Rassenti LZ, Downing JR, Shurtleff SA, Mills KI, sis without an intensive treatment option (PazoAML). Ann Hematol. et al. An international standardization programme towards the applica‑ 2019;98:1393–401. tion of gene expression profiling in routine leukaemia diagnostics: the 38. Cortes JE, Khaled S, Martinelli G, Perl AE, Ganguly S, Russell N, et al. microarray innovations in LEukemia study prephase. Br J Haematol. Quizartinib versus salvage chemotherapy in relapsed or refractory FLT3- 2008;142:802–7. ITD acute myeloid leukaemia (QuANTUM-R): a multicentre, randomised, 18. Hackl H, Steinleitner K, Lind K, Hofer S, Tosic N, Pavlovic S, et al. A gene controlled, open-label, phase 3 trial. Lancet Oncol. 2019;20:984–97. expression profile associated with relapse of cytogenetically normal 39. Randhawa JK, Kantarjian HM, Borthakur G, Thompson PA, Konopleva acute myeloid leukemia is enriched for leukemia stem cell genes. Leuk M, Daver N, et al. Results of a phase II study of crenolanib in relapsed/ Lymphoma. 2015;56:1126–8. refractory acute myeloid leukemia patients (Pts) with activating FLT3 19. Moreno-Martínez D, Nomdedeu M, Lara-Castillo MC, Etxabe A, Pratco‑ mutations. Blood. 2014;124:389–389. rona M, Tesi N, et al. XIAP inhibitors induce differentiation and impair 40. Maiti A, Naqvi K, Kadia TM, Borthakur G, Takahashi K, Bose P, et al. Phase II clonogenic capacity of acute myeloid leukemia stem cells. Oncotarget. trial of MEK inhibitor binimetinib (MEK162) in RAS-mutant acute myeloid 2014;5:4337–46. leukemia. Clin Lymphoma Myeloma Leuk. 2019;19:142-1481.e1. 20. Etxabe A, Lara-Castillo MC, Cornet-Masana JM, Banús-Mulet A, Nomde‑ 41. Borthakur G, Popplewell L, Boyiadzis M, Foran J, Platzbecker U, Vey N, deu M, Torrente MA, et al. Inhibition of serotonin receptor type 1 in acute et al. Activity of the oral mitogen-activated protein kinase kinase inhibitor myeloid leukemia impairs leukemia stem cell functionality: a promising trametinib in RAS-mutant relapsed or refractory myeloid malignancies. novel therapeutic target. Leukemia. 2017;31:2288–302. Cancer. 2016;122:1871–9. 21. Fang F, Antico G, Zheng J, Clevenger CV. Quantification of PRL/Stat5 42. Zucenka A, Maneikis K, Pugaciute B, Ringeleviciute U, Dapkeviciute A, signaling with a novel pGL4-CISH reporter. BMC Biotechnol. 2008;8:11. Davainis L, et al. Glasdegib in combination with low-dose Cytarabine for 22. DeVito WJ, Avakian C, Stone S, Okulicz WC. Prolactin-stimulated mitogen‑ the outpatient treatment of relapsed or refractory acute myeloid leuke‑ esis of cultured astrocytes is mediated by a protein kinase C-dependent mia in unfit patients. Ann Hematol. 2021;100:1195–202. mechanism. J Neurochem. 1993;60:835–42. 43. Boyd AL, Aslostovar L, Reid J, Ye W, Tanasijevic B, Porras DP, et al. 23. Leff MA, Buckley DJ, Krumenacker JS, Reed JC, Miyashita T, Buckley AR. Identification of chemotherapy-induced leukemic-regenerating cells Rapid modulation of the apoptosis regulatory genes, bcl-2 and bax by reveals a transient vulnerability of human AML recurrence. Cancer Cell. prolactin in rat Nb2 lymphoma cells. Endocrinology. 1996;137:5456–62. 2018;34:483-498.e5. 24. Goffin V, Kinet S, Ferrag F, Binart N, Martial JA, Kelly PA. Antagonistic 44. Milanovic M, Fan DNY, Belenki D, Däbritz JHM, Zhao Z, Yu Y, et al. Senes‑ properties of human prolactin analogs that show paradoxical agonistic cence-associated reprogramming promotes cancer stemness. Nature. activity in the Nb2 bioassay. J Biol Chem. 1996;271:16573–21659. 2018;553:96–100. 25. Duy C, Li M, Teater M, Meydan C, Garrett-Bakelman FE, Lee TC, et al. 45. Oberholtzer E, Contarini M, Veglia F, Cossarizza A, Franceschi C, Geuna M, Chemotherapy induces senescence-like resilient cells capable of initiat‑ et al. Prolactin increases the susceptibility of primary leukemia cells to NK ing AML recurrence. Cancer Discov. 2021;11:1542–61. and LAK effectors. Adv Neuroimmunol. 1996;6:233–47. 26. Farge T, Saland E, de Toni F, Aroua N, Hosseini M, Perry R, et al. Chemo‑ 46. Kooijman R, Gerlo S, Coppens A, Hooghe-Peters EL. Myeloid leukemic therapy-resistant human acute myeloid leukemia cells are not enriched cells express and secrete bioactive pituitary-sized 23 kDa prolactin. J for leukemic stem cells but require oxidative metabolism. Cancer Discov. Neuroimmunol. 2000;110:252–8. 2017;7:716–35. 47. Birkenkamp K, Geugien M, Lemmink H, Kruijer W, Vellenga E. Regulation 27. Patil MJ, Green DP, Henry MA, Akopian AN. Sex-dependent roles of of constitutive STAT5 phosphorylation in acute myeloid leukemia blasts. prolactin and prolactin receptor in postoperative pain and hyperalgesia Leukemia. 2001;15:1923–31. in mice. Neuroscience. 2013;253:132–41. 48. Hoelbl A, Schuster C, Kovacic B, Zhu B, Wickre M, Hoelzl MA, et al. Stat5 is 28. Utama FE, Tran TH, Ryder A, LeBaron MJ, Parlow AF, Rui H. Insensitivity indispensable for the maintenance of Bcr/Abl-positive leukaemia. EMBO of human prolactin receptors to nonhuman prolactins: relevance for Mol Med. 2010;2:98–110. experimental modeling of prolactin receptor-expressing human cells. 49. Warsch W, Kollmann K, Eckelhart E, Fajmann S, Cerny-Reiterer S, Hölbl A, Endocrinology. 2009;150:1782–90. et al. High STAT5 levels mediate imatinib resistance and indicate disease 29. Abramicheva PA, Smirnova OV. Prolactin receptor isoforms as the basis of progression in chronic myeloid leukemia. Blood. 2011;117:3409–20. tissue-specific action of prolactin in the norm and pathology. Biochemis‑ try. 2019;84:329–45. Cuesta‑Casanovas et al. Cancer Cell International (2023) 23:97 Page 18 of 18 50. Tang Y, Xiao S, Wang Z, Liang Y, Xing Y, Wu J, et al. A prognostic model for acute myeloid leukemia based on IL-2/STAT5 pathway-related genes. Front Oncol. 2022;12:38. 51. Wingelhofer B, Maurer B, Heyes EC, Cumaraswamy AA, Berger-Becvar A, de Araujo ED, et al. Pharmacologic inhibition of STAT5 in acute myeloid leukemia. Leukemia. 2018;32:1135–46. 52. Ogura M, Uchida T, Terui Y, Hayakawa F, Kobayashi Y, Taniwaki M, et al. Phase I study of OPB-51602, an oral inhibitor of signal transducer and activator of transcription 3, in patients with relapsed/refractory hemato‑ logical malignancies. Cancer Sci. 2015;106:896–901. 53. Döhner H, Estey E, Grimwade D, Amadori S, Appelbaum FR, Büchner T, et al. Diagnosis and management of AML in adults: 2017 ELN recommen‑ dations from an international expert panel. Blood. 2017;129:424–47. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Ready to submit your research ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions