Phosphatase of regenerating liver-3 (PRL-3) is overexpressed in classical Hodgkin lymphoma and promotes survival and migration
Phosphatase of regenerating liver-3 (PRL-3) is implicated in oncogenesis of hematological and solid cancers. PRL-3 expression increases metastatic potential, invasiveness and is associated with poor prognosis. With this study, we aimed to show a possible oncogenic role of PRL-3 in classical Hodgkin lymphoma (cHL).
PRL-3 expression was measured in 25 cHL patients by immunohistochemistry and gene expression was analyzed from microdissected malignant cells. We knocked down PRL-3 in the cHL cell lines L1236 and HDLM2 and used small molecular inhibitors against PRL-3 to investigate proliferation, migration and cytokine production.
PRL-3 protein was expressed in 16% of patient samples. In three different gene expression datasets, PRL-3 was significantly overexpressed compared to normal controls. PRL-3 knockdown reduced proliferation, viability and Mcl-1 expression in L1236, but not in HDLM2 cells. Thienopyridone, a small molecule inhibitor of PRL-3, reduced proliferation of both L1236 and HDLM2. PRL-3 affected IL-13 secretion and enhanced STAT6 signaling. IL-13 stimulation partially rescued proliferation in L1236 cells after knockdown of PRL-3. PRL-3 knockdown reduced migration in both L1236 and HDLM2 cells.
PRL-3 was overexpressed in a subset of cHL patients. Inhibition of PRL-3 increased IL-13 cytokine production and reduced migration, proliferation and viability. The effects could be mediated through regulation of the anti-apoptotic molecule Mcl-1 and a feedback loop of IL-13 mediated activation of STAT6. This point to a role for PRL-3 in the pathogenesis of Hodgkin lymphoma, and PRL-3 could be a possible new drug target.
KeywordsPRL-3 Hodgkin lymphoma IL-13 STAT6 Migration Survival
phosphatase of regenerating liver-3
classical Hodgkin lymphoma
Hodgkin and Reed–Sternberg
nuclear factor κB
signal transducer and activator of transcription
acute lymphoblastic leukemia
peripheral blood mononuclear cells
myeloid cell leukemia-1
quantitative real time PCR
C-C motif chemokine ligand
granulocyte–macrophage colony-stimulating factor
C-X-C motif chemokine ligand
vascular endothelial growth factor
Lymphoma is divided in Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL), with further subclassification based on immunophenotype, morphological, genetic and clinical features . Lymphoma is the third most common childhood malignancy, and constitutes approximately 10% of all pediatric and 5% of adults cancers . HL is for most cases a curable cancer, with 5-year survival rates of 98% in children < 15 years of age and 86% in adults [3, 4]. Still, these patients can suffer from reduced life expectancy because of increased risk for secondary malignancies and toxicity from treatment . Thus, the challenge of HL is to find an optimal strategy for maintaining high cure rates and overall survival while simultaneously avoiding long-term side effects and risk for secondary malignancies associated with HL.
The hallmark of classical HL (cHL) is the malignant Hodgkin and Reed–Sternberg (HRS) cells. The HRS cells are derived from mature B cells, but they have a phenotype and expression of hematopoietic surface markers that do not resemble those of any other hematopoietic cells . Several signaling pathways are deregulated in cHL, including constitutively active nuclear factor-κB (NF-κB) and Janus kinase-signal transducer and activator of transcription (JAK-STAT) signaling pathways. A salient feature in cHL is that the malignant cells often constitute only 0.1–2% of the tumor mass. The rest is composed of non-malignant inflammatory cells like T cells, B cells, plasma cells, macrophages, natural killer (NK) cells, dendritic cells, neutrophils, eosinophils, mast cells, and fibroblasts . HRS cells secrete proliferative cytokines, like interleukin-6 (IL-6) and IL-13, which act in an autocrine and paracrine manner. The secretion of cytokines and interplay with the microenvironment, are crucial for survival of the malignant HRS cells [7, 8, 9].
We have previously found that phosphatase of regenerating liver-3 (PTP4A3/PRL-3) has oncogenic properties in multiple myeloma (MM) [10, 11] and acute lymphoblastic leukemia (ALL) . PRL-3 is up-regulated in several solid and hematological cancers, and is associated with metastasis, survival and poor prognosis [13, 14]. Little is known about PRL-3 in lymphoma, except for high mRNA-expression in some cell lines [10, 15]. No PRL-3 protein expression is found in mature healthy tissues. This seemingly restricted expression to cancerous tissues, makes PRL-3 a tempting target for cancer treatment.
With this paper, we aimed to search for PRL-3 expression in cHL and elucidate a possible survival and migratory role of PRL-3 in cHL.
The cell lines HDLM2 (ACC 17), L1236 (ACC 530), SUP-B15 (ACC 389), MHH-CALL-4 (ACC 337), CRO-AP5 (ACC 215) were from DSMZ (Brauschweig, Germany). Reh was from ATCC (Rockville, MD, USA). OCI-Ly3, OCI-Ly10, SU-DHL-4 and Karpas 422 were a kind gift from Ph.D. June H. Myklebust at the Norwegian Radium Hospital, Oslo University Hospital (Oslo, Norway). Cells were cultured in RPMI-1640 and 10–20% heat-inactivated fetal calf serum, according to supplier’s recommendation, at 37 °C in a humidified atmosphere with 5% CO2. Growth media were replenished twice weekly. Peripheral blood mononuclear cells (PBMC) was isolated from healthy volunteers using density gradient centrifugation.
Antibodies, cytokines and other reagents
Antibodies against PRL-3 (sc-130355), Myeloid cell leukemia-1 (Mcl-1) (sc-819), IL-13Rα1 (sc-398831) and IL-13Rα2 (sc-134363) were bought from Santa Cruz Biotechnology (Dallas, Texas, USA). Antibodies against phosphorylated (p) STAT1 (#9167), total STAT1 (#14994), pSTAT6 (#9361), total STAT6 (#9362) and B cell lymphoma-extra large (Bcl-xL) (#2762) were from Cell Signaling Technology (Beverly, MA, USA). Antibodies against GAPDH and PRL-3 (ab52976) were from Abcam (Cambridge, United Kingdom). Recombinant human C-C motif chemokine ligand 19 (CCL19) and IL-13 were from Peprotech (Rocky Hill, NJ, USA). The plasmid pLKO (non-silencing control) and shRNA-pLKO against PRL-3 were a kind gift from dr. Jim Lambert (University of Colorado, Denver, CO, USA) . Neutralizing antibodies against IL-6 (MAB2061) and IL-13 (MAB213) were from R&D Systems (Minneapolis, MN, USA). PRL-3 inhibitor I (#P0108) (5-[[5-Bromo-2-[(2-bromophenyl)methoxy]phenyl]methylene]-2-thioxo-4-thiazolidinone) was from Sigma-Aldrich (St. Louis, MO, USA), Analog 3 (#Z44389470) was from Enamine Store (Ukraine) and thienopyridone was a kind gift from Professor John S. Lazo (University of Virginia, Charlottesville, VA, USA).
Immunohistochemistry (IHC) and gene expression datasets
Paraffin-embedded tissue blocks from cHL patients were cut in 4 μm sections. The sections were pretreated in Target Retrieval Solution (Dako K8004, Oslo, Norway) for 20 min at 97 °C. Subsequently, sections were washed and incubated with primary antibody against PRL-3 (ab52976, 1:1000) for 40 min. Labelled polymer HRP anti-Rabbit (Dako K4011) was used for developing the stain. Microscopic images were obtained with a Lumenera Infinity 2 camera and Infinity analyze software, release 6.2 (Lumenera Corporation, Ottawa, Ontario, Canada) using a Nikon eclipse Ci microscope (Nikon Gmbh, Düsseldorf, Germany) at 400× magnification. PRL-3 expression was graded in positive or negative expression. Positive expression was defined as uniform positive staining in tumor cells compared to background staining. PRL-3 expression in normal lymph node and prostate cancer was used as a negative and positive control, respectively (Additional file 1).
We searched the open gene expression profiling source Oncomine (http://www.oncomine.org) and PubMed to identify gene expression datasets comparing PRL-3 expression in HRS cells with normal control cells.
Lentiviral transduction for PRL-3 knockdown
Stable PRL-3 knockdown in HDLM2 and L1236 was done as previously described using pLKO-shRNA against PRL-3 or pLKO (control plasmid) to establish PRL-3 knockdown (L1236 shPRL-3 and HDLM2 shPRL-3) and control (L1236 shCTRL and HDLM2 shCTRL) cells, respectively [12, 16]. Transfected cells were grown in medium containing 0.2 µg/mL puromycin for selection.
Cells were treated as indicated, then pelleted, lysed and immunoblotted as previously described . Images were acquired using LI-COR Odyssey Fc and analyzed with Image Studio Software (LI-COR, Lincoln, Nebraska). All experiments were conducted at least three times.
RNA isolation, cDNA synthesis and quantitative real time PCR (qRT-PCR)
Isolation of RNA and cDNA synthesis were performed as previously described . For relative quantification, we used the comparative 2−ΔΔCT-method, with GAPDH as endogenous reference. All samples were run in triplicate. The following primers were used: PTP4A3 (Hs02341135_m1), GAPDH (Hs99999905_m1) and IL-13 (Hs00174379_m1) (TaqMan, Applied Biosystems, ThermoFisher Scientific, Waltham, MA, USA). Analysis was performed using Step One Software 2.3 (Applied Biosystems).
Cell proliferation and viability
Cell proliferation was measured by counting cells seeded into a 24-well plate on day 1, 4 and 7 using Coulter Counter Z1 (Bechman Coulter, Fullerton, CA) or Countess Cell Counter (ThermoFisher). Viability and apoptosis were evaluated using Annexin V-FITC binding and propidium iodide (PI) uptake (APOPTEST-FITC kit; Nexins Research, Kattendijke, The Netherlands), as previously described . For L1236, PI negative or positive were used as measurement for viable or dead cells, respectively.
Relative ATP measurement
Cells (1 × 104) were seeded into 96-well clear bottom plates (Corning, Corning, NY, USA), with cytokine or inhibitor, as indicated. Cellular ATP production was estimated using CellTiter-Glo® assay as previously described . All conditions were run in triplicate, and experiments were performed three times.
Cells (4 × 105) were seeded into the upper chamber of a transwell culture plate with a polycarbonate membrane with pore size 5 µm (HDLM2) or 8 µm (L1236) (Corning). CCL19 (50 ng/mL) was used as a chemoattractant for 6 h (HDLM2) or 22 h (L1236). The cells that migrated through the membrane to the lower chamber, were counted using Coulter Counter Z1 (Bechman Coulter, Fullerton, CA). All conditions were performed in duplicate, and all experiments were performed three times.
Bio-Plex Pro Human Cytokine 27-plex Assay (Bio-Rad, Hercules, California, USA) was used according to the manufacturer’s instruction. Supernatants of cell cultures (5 × 105 cells/mL) were harvested 24 h after medium exchange, and IL-6 (DY26) and IL-13 (DY2213) secretion were measured using quantitative DuoSet ELISA (R&D systems, Minneapolis, USA), according to the manufacturer’s instructions.
Cells were allowed to adhere to Poly-l-lysine (0.01%) and then fixed with 4% paraformaldehyde before they were permeabilized with PBS, 5% human serum and 0.5% saponin (Sigma-Aldrich). PRL-3 was detected with directly conjugated PRL-3-Alexa 488 antibody (St. Cruz Biotechnology), and the nuclei were stained with Hoechst 33342 (ThermoFisher Scientific). Confocal imaging was performed using a Leica SP8 STED 3D microscope (Wetzlar, Germany). Imaris (Bitplane, Zurich, Switzerland) was used for imaging editing.
Statistical significance was determined using Student’s t test in IBM SPSS Statistics version 24.
The use and storage of patient samples was approved by the Regional Ethics Committee (Approval #28/99, 2013/1594 and 2015/700-03).
PRL-3 was expressed in cHL patients and cell lines
PRL-3 expression in microdissected HRS cells compared to normal control
Table 1: PRL-3 was overexpressed in HRS cells compared to normal control cells in three different datasets. The malignant HRS cells constitute only 0.1–2% of the tumor volume. To obtain gene expression from only the HRS cells, all three datasets had isolated HRS cells by microdissection before gene expression was performed.
In our patient cohort, there was two deaths, and we did not find any difference in clinical factors such as stage, relapse or survival when comparing patients with positive or negative expression of PRL-3. We also analyzed data from Steidl et al. , and found no significant difference in stage (Fig. 1c), survival (Fig. 1d) or relapse (not shown), based on PRL-3 expression.
PRL-3 knockdown reduced survival of L1236
In L1236 cells, knockdown of PRL-3 resulted in significantly reduced viability, but not in HDLM2 (Fig. 3b). Cell proliferation, as measured by cell counts on day 4 and 7, was reduced in L1236 shPRL-3 compared to L1236 shCTRL cells. HDLM2 shPRL-3 on the other hand, proliferated at the same rate as HDLM2 shCTRL (Fig. 3c). By western blotting, there was reduced amount of Mcl-1 protein in L1236 shPRL-3, but not in HDLM2 shPRL-3 (Fig. 3d). The amount of Bcl-xL, another important anti-apoptotic protein in cHL, was unchanged following knockdown of PRL-3 in L1236 and HDLM2. The reduced level of Mcl-1, points to a possible explanation for the reduced viability of L1236 cells after PRL-3 knockdown.
PRL-3 knockdown reduced migration of cHL cells
In addition to cancer cell survival, PRL-3 has previously been associated with metastasis and cancer cell migration. The chemokine CCL19 is a known stimulator for migration of cHL cells. We therefore examined migration of cells with or without PRL-3 knockdown. In a transwell assay, CCL19 stimulated migration of both cell lines. Knockdown of PRL-3 reduced cell migration with 56 and 51% in L1236 and HDLM2, respectively (Fig. 3e, f).
PRL-3 inhibitors reduced cell survival in cHL cells
PRL-3 regulated cytokine production in cHL cells
A common feature of cHL cells is the secretion of distinct cytokines promoting proliferation and survival. We therefore used a multiplex assay to screen PRL-3 knockdown and control cells for differences in cytokine secretion. We found that IL-6, IL-9, IL-13, granulocyte–macrophage colony-stimulating factor (GM-CSF), C-X-C motif chemokine ligand 10 (CXCL10), CCL4, CCL5 and vascular endothelial growth factor (VEGF) were differentially secreted in PRL-3 knockdown compared to control in either L1236 and/or HDLM2 (data not shown). In subsequent experiments, we chose to focus on IL-6 and IL-13, as these cytokines previously have been linked to PRL-3 or cHL pathogenesis.
Next, we stimulated cells with IL-13 and examined proliferation. Interestingly, IL-13 increased proliferation of L1236 shPRL-3 with 245% (2.5 ng/mL), and 29% in L1236 shCTRL (Fig. 5e). Proliferation of HDLM2 shPRL-3 did not respond to IL-13 stimulation, while HDLM2 shCTRL increased proliferation with 22% (Fig. 5e). Taken together, this indicates that PRL-3 regulates secretion of proliferative cytokines, and in the case of IL-6 and IL-13, by lowering the production or secretion.
PRL-3 enhanced STAT6 signaling
With this study, we aimed to explore potential roles of PRL-3 in the pathogenesis of cHL. We show that PRL-3 protein was detected in a subset of lymph node biopsies from cHL patients. In microdissected HRS cells, PRL-3 was overexpressed compared to normal control cells. Further, we show that inhibition of PRL-3 by shRNA or by small molecular inhibitors influenced cHL cell survival, reduced Mcl-1 expression and inhibited migration. PRL-3 knockdown also influenced IL-6 and IL13 secretion, and enhanced STAT6 signaling.
In our patient cohort, PRL-3 protein expression was detected in 16% of the samples. In a study by Wang et al., PRL-3 was elevated in an average of 22.3% (range 8.3–37.7%) of various cancer samples . However, hematological cancers were not included in this work. High PRL-3 expression in hematological cancers has previously been shown by us and others [10, 12, 24]. In cHL, PRL-3 was introduced as a possible oncogene by Schwering et al.  based on differentially expressed genes in L1236 compared to germinal center B cells. In the same study, they confirmed PRL-3 expression in three other cHL cell lines, including HDLM2. However, they did not compare expression in these cell lines with normal germinal center B cells. Consequently, they were not able to conclude whether PRL-3 was overexpressed in cHL in general or only in L1236. In the present study, by using open source gene expression profiling data, we show that PRL-3 is overexpressed in microdissected HRS cells, proving overexpression of PRL-3 in cHL. High PRL-3 expression is associated with metastasis in solid tumors, and poor prognosis in both solid and hematological cancers [25, 26, 27]. However, high PRL-3 expression in cHL did not correlate with clinical parameters, such as stage or survival, in either Steidl et al. , or in our cohort. In the study by Steidl et al. the HRS cells were not microdissected. To conclude whether PRL-3 is associated with negative prognosis, analysis from microdissected HRS cells should be done in a larger patient population. cHL is a disease with 86% survival for adults , and > 97% for children and adolescents . Thus, finding genes associated with prognostic outcome, without a large number of samples, is difficult.
To explore cell survival effects of PRL-3, we transduced the cHL cell lines L1236 and HDLM2 with stable shRNA, or used small molecular inhibitors, against PRL-3. We found that inhibition of PRL-3 reduced proliferation and survival of L1236 and HDLM2 cells with the notable exception of knockdown in HDLM2. Knockdown of PRL-3 in L1236 reduced the expression of Mcl-1 but not in HDLM2. This may be an explanation for decreased survival in L1236 shPRL-3, and why there was no difference in survival by PRL-3 knockdown in HDLM2. We have previously shown that Mcl-1 expression in MM correlates with PRL-3 expression [11, 18]. Zhou et al.  found increased Mcl-1 expression in AML cell lines with forced expression of PRL-3. Nagel et al.  showed that Mcl-1 was highly expressed in HL cell lines, and induced by IL-6 expression as in other malignancies as well [7, 18].
Migration of cancer cells is important for dissemination of the cancer disease. In cHL, spread of tumor cells follows a typical pattern by first involving anatomically neighboring lymph nodes before dissemination to non-adjacent sites and organs . Factors regulating spread of cHL cells are poorly understood . In a study by Linke et al.  they show that cHL cell lines are highly migratory to a CCL19 gradient by RhoA and ROCK activation. In the present study, we show that knockdown of PRL-3 in L1236 and HDLM2 reduces migration towards a CCL19 gradient. Migration and motility have been linked to PRL-3 expression in several cancers [10, 12, 25, 26, 31, 32]. PRL-3 is associated with both RhoA and ROCK activation . Taken together this indicates that PRL-3 could be important for cHL cells ability to disseminate.
A link between PRL-3 and IL-13-driven proliferation is further supported by our results from IL-13 stimulation. IL-13 increased proliferation by 245% in L1236 shPRL-3. In a work by Skinnider et al., L1236 cells were very sensitive to IL-13 neutralization and stimulation, through regulation of STAT6 activation . IL-13 stimulation increased proliferation of HDLM2 shCTRL, but not shPRL-3. The lack of increase in HDLM2 shPRL-3 is probably due to saturation with IL-13 in an autocrine loop. Kapp et al.  did not see any increase in proliferation by IL-13 stimulation in HDLM2, but they used higher concentrations and a different method. The results from the small molecular inhibitors further support that PRL-3 is important for survival of cHL cells. The cells were most sensitive towards thienopyridone, a PRL family inhibitor. Thienopyridone, and its derivate iminothienopyridinedione 13, are known as the most potent PRL-3 inhibitors [37, 38, 39, 40]. In MM and gastric cancer cell lines, sensitivity to PRL-3 inhibitor I was not related to the PRL-3 level [18, 41], as we also show here for L1236 and HDLM2. Taken together, our data suggest that PRL-3 influences survival of HL cells, probably through regulation of IL-13 secretion, phosphorylation of STAT6 and expression of Mcl-1.
In this study, we show that PRL-3 protein was expressed in 16% of cHL patients and that PRL-3 was overexpressed in HRS cells compared to normal controls. Inhibition of PRL-3 influenced cytokine production and reduced migration, proliferation and viability, probably through regulating Mcl-1 expression and IL-13-stimulated activation of STAT6.
MAH, MB, TBR designed the study. MAH, HH, PA, ENV, UMF performed experiments, acquired data or provided material. MAH, HH, PA, ENV, BL, TSS, MB, TBR analyzed and interpreted the data. MAH drafted the manuscript. All authors edited the final manuscript. All authors read and approved the final manuscript.
We want to thank Hanne Hella and Berit Størdal for technical help. The preparation and staining of biopsies for IHC, was performed by the Cellular and Molecular Imaging Core Facility (CMIC), Norwegian University of Science and Technology (NTNU). CMIC provided confocal microscopy equipment. CMIC is funded by the Faculty of Medicine at NTNU and Central Norway Regional Health Authority.
The authors declare that they have no competing interests.
Availability of data and materials
All data generated or analyzed during this study are included in this published article, or available from Gene Expression Omnibus (GSE17920, GSE39132, GSE39133, GSE39134, GSE14879, and GSE12453).
Consent for publication
Ethics approval and consent to participate
The use and storage of patient samples was approved by the Regional Ethics Committee (Approval #28/99, 2013/1594 and 2015/700-03).
This work was financially supported by Joint Research Committee between St. Olavs hospital HF and Faculty of Medicine and Health Science, NTNU.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 4.American Cancer Society. Cancer facts and figures 2017. Atlanta: American Cancer Society; 2017.Google Scholar
- 16.Gari HH, DeGala GD, Ray R, Lucia MS, Lambert JR. PRL-3 engages the focal adhesion pathway in triple-negative breast cancer cells to alter actin structure and substrate adhesion properties critical for cell migration and invasion. Cancer Lett. 2016;380:505–12.CrossRefPubMedPubMedCentralGoogle Scholar
- 36.Zhou J, Chong PS, Lu X, Cheong LL, Bi C, Liu SC, et al. Phosphatase of regenerating liver-3 is regulated by signal transducer and activator of transcription 3 in acute myeloid leukemia. Exp Hematol. 2014;42(1041–1052):e2.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.