- Email: [email protected]
A novel extracellular matrix-based leukemia model supports leukemia cells with stem cell-like characteristics
Leukemia Research 72 (2018) 105–112
Contents lists available at ScienceDirect
Leukemia Research journal homepage: www.elsevier.com/locate/leukres
Research paper
A novel extracellular matrix-based leukemia model supports leukemia cells with stem cell-like characteristics
T
Dandan Lia, Tara L. Linb, Brea Lipeb, Richard A. Hopkinsc, Heather Shinogled, ⁎ Omar S. Aljitawia,b,e, a
Department of Pathology and Laboratory Medicine, University of Kansas Medical Center, Kansas City, KS, United States Division of Hematology/Oncology and Blood and Marrow Transplantation Program, 2330 Shawnee Mission Parkway, University of Kansas Medical Center, Kansas City, KS, United States c Cardiac Surgery Research Laboratories, Children's Mercy Hospital and Clinics, Kansas City, Missouri, United States d Microscopy and Analytical Imaging Laboratory, University of Kansas, Lawrence, KS, United States e Department of Medicine, Hematology/Oncology and Bone Marrow Transplant Program, University of Rochester Medical Center, Rochester, NY, 14642, United States b
A R T I C LE I N FO
A B S T R A C T
Keywords: Extracellular matrix In vitro 3D model Leukemia Leukemia stem cell-like Decellularized Wharton's jelly matrix
Acute myeloid leukemia (AML) relapse results from the survival of chemotherapy-resistant and quiescent leukemia stem cells (LSC). These LSCs reside in the bone marrow microenvironment, comprised of other cells and extracellular matrix (ECM), which facilitates LSC quiescence through expression of cell adhesion molecules. We used decellularized Wharton’s jelly matrix (DWJM), the gelatinous material in the umbilical cord, as a scaffolding material to culture leukemia cells, because it contains many components of the bone marrow extracellular matrix, including collagen, fibronectin, lumican, and hyaluronic acid (HA). Leukemia cells cultured in DWJM demonstrated decreased proliferation without undergoing significant differentiation. After culture in DWJM, these cells also exhibited changes in morphology, acquiring a spindle-shaped appearance, and an increase in the ALDH+ cell population. When treated with a high-dose of doxorubicin, leukemia cells in DWJM demonstrated less apoptosis compared with cells in suspension. Serial colony forming unit (CFU) assays indicated that leukemia cells cultured in DWJM showed increased colony-forming ability after both primary and secondary plating. Leukemia cell culture in DWJM was associated with increased N-cadherin expression by flow cytometry. Our data suggest that DWJM could serve as an ECM-based model to study AML stem cell-like cell behavior and chemotherapy sensitivity.
1. Introduction Acute myeloid leukemia (AML) is a heterogeneous hematopoietic malignancy characterized by an aberrant clonal expansion of undifferentiated myeloid blasts. Studies have shown that leukemia stem cells (LSCs) contribute to relapse after chemotherapeutic treatment. [1] Like normal hematopoietic stem cells (HSCs), LSCs maintain their selfrenewal ability while generating clonogenic leukemic progenitors capable of producing leukemic cells [2]. Anti-proliferative chemotherapeutic agents commonly target the rapidly cycling leukemic cells, but they generally are ineffective against the quiescent LSCs, partly because of enhanced drug efflux in LSCs [3]. Therefore, it is important to develop therapeutic strategies which eliminate the LSCs in the bone marrow, where they share the “hematopoietic niche” along with normal HSCs [4]. The LSC niche, similar to the hematopoietic niche, is a 3D
⁎
microenvironment composed of bone marrow stromal cells and ECM components like collagen, fibronectin and tenascin [4]. These components create compartments that not only provide structural support to the cells in the bone marrow, but also provide chemokines and cytokines that are important in regulating LSC self-renewal, trafficking, proliferation and differentiation [5]. Currently, most leukemia in vitro studies are based on conventional two-dimensional (2D) cultures in tissue culture polystyrene (TCP) dishes/ flasks and stromal co-cultures. These models are useful in elucidating some of the molecular mechanisms of leukemia initiation and progression. However, 2D culture systems lack the leukemia-microenvironment interaction present in the 3D bone marrow microenvironment. Therefore, the LSCs in 2D culture frequently differentiate and lose their “stem-ness”. The development of a 3D model that replicates the in vivo mechanical and biochemical properties of bone
Corresponding author at: 601 Elmwood Avenue, Rochester, NY, 14642, United States. E-mail address: [email protected] (O.S. Aljitawi).
https://doi.org/10.1016/j.leukres.2018.08.012 Received 15 March 2018; Received in revised form 12 August 2018; Accepted 13 August 2018 Available online 16 August 2018 0145-2126/ © 2018 Elsevier Ltd. All rights reserved.
Leukemia Research 72 (2018) 105–112
D. Li et al.
times, 4 h prior to proliferation assessment. 10% alamarBlue (Biocentric) was added into each well. After 4 h, 100 μl of supernatant from each well was aspirated to a new well of a 96 well plate and fluorescence was measured by micro-plate reader, with excitation wavelength at 530 nm and emission wavelength at 590 nm.
marrow could allow maintenance of a true LSC-state. Our laboratory has been focused on characterizing decellularized Wharton’s jelly matrix (DWJM) and examining its potential for regenerative medicine applications. [6–8], We hypothesized that DWJM would provide a similar environment to the bone marrow ECM, because its components, such as collagen, fibronectin, hyaluronic acid, and sulphated proteoglycan [8], also exist in the bone marrow hematopoietic niche. Moreover, because cell-ECM interactions play an important role in chemoresistance in leukemia cells [9], the environment provided by DWJM could support the maintenance of LSCs. We therefore used DWJM as an ECM to examine leukemia cell-ECM interactions, hypothesizing that DWJM would support leukemia cells with LSC-like characteristics. In this model, we investigated the growth pattern of 3 human leukemia cell lines (HL60, Kasumi-1, and MV411), with a focus on proliferation, viability, morphology and myeloid differentiation. We also studied the drug resistance and stem cell characteristics of leukemia cells cultured in this model, compared to leukemia cells cultured in suspension. We found that leukemia cells cultured in our DWJM-based ECM model had LSC characteristics, suggesting that DWJM may prove useful in LSC characterization and in developing therapeutic interventions that target LSCs.
2.5. CellTrace proliferation assay To monitor the cell division of leukemia cells in suspension and in DWJM, cells were labeled with CellTrace Violet (Life Technology) before seeding. Briefly, cells were washed and resuspended with PBS at the concentration of 106cells/ml, and CellTrace Violet stock solution was added in a final concentration of 1 μl/ml. After incubation at room temperature for 20 min, 5 ml of PBS with 10% FBS were added and incubated for 5 min, followed by centrifugation to obtain pellets. Cells were resuspended in culture medium and cultured in either suspension or DWJM. Cell division was measured by flow cytometry soon after seeding and after 7 days of culture. To isolate cells from DWJM, we washed wafers in PBS, and then used collagenase II (0.05 g collagenase II in 50 ml DMEM for 1–2 hours) to digest DWJM at 37 °C. 2.6. Cell viability
2. Materials and methods Cell survival in DWJM was measured by Vi-CELL Series Cell Viability Analyzer (Beckman Coulter), which is based on Trypan Blue dye exclusion. Cells in each DWJM wafer were released by treating with 1 ml 0.002 g/ml collagenase II (Worthington) for about 2 h at 37 °C. The released cells were assessed for viability according to the manufacturer’s recommendations.
2.1. Cell culture Human AML cell lines HL60, Kasumi I and MV 411 (ATCC, Manassas, VA) were maintained in T 75 tissue culture flasks with Advanced Roswell Park Memorial Institute (RPMI) 1640 Medium (Gibco), supplemented with 5% fetal bovine serum (Sigma-Aldrich) and 1% penicillin/streptomycin (pen/strep) (Life Technologies). Cells were maintained at 37 °C in a fully humidified 5% CO2 incubator.
2.7. Histology and immunohistochemistry For morphological analysis, wafers were washed with PBS three times, fixed in 4% PFA, embedded in paraffin, sectioned and stained with hematoxylin and eosin, and visualized under the microscope using an Olympus BX40 microscope; pictures were obtained using a DP72 digital camera.
2.2. DWJM scaffold preparation The preparation of DWJM was previously described. [8,10], Briefly, to prepare DWJM scaffolds, fresh human umbilical cords were dissected after removing surrounding membranes and blood vessels. Then they were subjected to two cycles of osmotic shock in hypertonic and hypotonic solutions, followed by immersion in a non-ionic detergent Triton-x, an anionic detergent sodium lauryl succinate; finally, they underwent enzymatic digestion with recombinant endonuclease. The resulting DWJM pieces were cut into thin wafers (3 mm thick) as previously described [10] (Supplementary Figure-01A). Cartoon depiction of leukemia cells interacting with DWJM fibers (Supplementary Figure01B).
2.8. Treatment with chemotherapeutic agents After 7 days, cells cultured in suspension and cells in DWJM were treated with 50μM of doxorubicin hydrochloride (Sigma-Aldrich) for 48 h. For cells in suspension, culture medium was removed, and chemotherapeutic agents were added in fresh medium. For cells in DWJM, scaffolds were transferred into new 24-well plates and washed with PBS three times; then a chemotherapeutic agent was added to the culture medium.
2.3. Seeding DWJM wafers with AML cell lines 2.9. Apoptosis assay Before seeding, cryopreserved DWJM wafers were thawed, washed three times in phosphate buffered saline (PBS), and pre-incubated with Advanced RPMI overnight. AML cells (2*105 cells/well) were seeded into DWJM wafers in 24-well non-tissue culture treated plates with 60% area of each well covered by DWJM. Culture plates were then placed in an incubator at 37 °C with 5% CO2 and maintained in Advanced RPMI with 5% FBS for 7 days; half of the medium was changed every other day. AML cells in suspension (2*105 cells/well), cultured under the same conditions, were used as controls. In AlamarBlue assay, CellTrace proliferation and Ki67 immunohistology sample preparation, cells were maintained in RPMI 1640 (Sigma-Aldrich) with 10% FBS for HL60 and MV411 cells, and 20% FBS for Kasumi I cells.
Apoptosis in leukemia cells was measured by flow cytometry using Annexin V-Alexa 568 (Invitrogen, USA) and DAPI (Invitrogen, USA) staining. Prior to flow cytometry analysis, cells in DWJM wafers were released as described previously and 105 released cells as well as cells in suspension were stained with DAPI and Annexin V according to manufacturer’s recommendations. Data were acquired within 1 h, using LSR II (BD Biosciences), and analyzed by FlowJo software. 2.10. Aldefluor assay Aldehyde dehydrogenase (ALDH) activity was examined by using Aldefluor reagent (Stem Cell Technologies) according to the manufacturer’s protocol, followed by flow cytometry. Cells negative for propidium iodide (PI) staining were considered positive for ALDH, based on a negative control using the ALDH inhibitor diethylaminobenzaldehyde (DEAB). Data were analyzed within 1 h, using LSR II (BD
2.4. AlamarBlue assay To assess the proliferation of AML cells, DWJM wafers with cells were transferred to new 24 well plates and washed with PBS three 106
Leukemia Research 72 (2018) 105–112
D. Li et al.
Fig. 1. Leukemia cells cultured in DWJM demonstrate reduced proliferation while maintaining cell viability and undifferentiated state. (A) HL60, Kasumi 1 and MV411 cells were seeded onto DWJM wafers and allowed to adhere to DWJM for 2 days. Then DWJM wafers with attached cells were replated, and AlamarBlue assay was used to assess cell proliferation on day 2, 4, and 6. Data are normalized to the day 2 fluorescence reading. (B) Collagenase II was used to release cells embedded in DWJM, and cell viability was assessed by Vi-CELL, based on Trypan blue exclusion on days 2, 4, and 6. All values represent means ± SEM. (C) Violet intensity of HL60 and MV411 in DWJM and suspension before and 7 days after seeding. (D) Apoptosis and necrosis of HL60 and MV411 cultured in suspension or in DWJM, as measured by flow cytometry following co-staining with AnnexinV/ PI. Results are shown as density plots of one representative experiment (upper) and as summary of results of multiple experiments (bottom). Data represent means ± SEM. Experiments done in triplicate. (E) CD11b expression in HL60 and Kasumi 1 cultured in suspension and DWJM.
107
Leukemia Research 72 (2018) 105–112
D. Li et al.
Biosciences).
statistical threshold of p < 0.05.
2.11. Analysis of differentiation marker
3. Results
Expression of CD11b was measured by flow cytometry. 105 cells, cultured in suspension or in DWJM, were harvested and incubated with FITC-conjugated anti-human CD11b (Abcam) for 10 min at room temperature. After staining, cells were washed and resuspended in 400 μl PBS and analyzed by flow cytometry by using LSR II (BD Biosciences).
3.1. Leukemia cells cultured in DWJM demonstrated reduced proliferation while maintaining cell viability and undifferentiated state Since LSCs are quiescent, we first examined leukemia cell proliferation over time in our model, hypothesizing that leukemia-DWJM interactions result in decreased leukemia cell proliferation over time. Using AlamarBlue assay on days 2, 4, and 6 after leukemia cell seeding, we demonstrated decreased proliferation in all three leukemia cell lines (HL 60, Kasumi I and MV411) (Fig. 2A). Despite reduced proliferation, all three cell lines maintained nearly unchanged viability at the three time points, as measured by Trypan Blue dye exclusion (Fig. 1B). Next, we studied leukemia cell proliferation and viability in our model, compared to suspension culture systems. We found that leukemia cells cultured in DWJM demonstrated reduced proliferation measured by CellTrace proliferation assay (Fig. 1C), but no significant differences in apoptosis between the two culture conditions when assessed by Annexin V/PI flow cytometry (Fig. 1D). LSCs are expected to maintain an undifferentiated state. We next examined DWJM culture effects on leukemia cell differentiation by examining expression of CD11b, a common myeloid marker [11], using flow cytometry. Our experiments showed no increase in CD11b expression in either suspension or DWJM-culture conditions, indicating that both conditions maintain leukemia cells in an undifferentiated state (Fig. 1E). Taken together, our data suggest that leukemia cells cultured in DWJM are more quiescent than leukemia cells in suspension.
2.12. Colony forming unit (CFU) assay Cells were plated on 35 mm dishes (500 cells/dish for Kasumi I, and 300 cells/dish for HL60 and MV411) in triplicate in MethoCult® H4434 Classic methylcellulose (Stemcell Technologies, Vancouver, Canada). Cells were washed and resuspended in RPMI 1640 medium. After measuring cell number and viability by Trypan blue, cell densities were adjusted at a concentration of 10 cells/μl, and cells were added into methylcellulose, followed by addition of 1 ml cell-methylcellulose mix into each dish. After 12–14 days of incubation at 37 °C in 5% CO2, colonies consisting of > 30 cells were counted, then harvested and replated in methylcellulose. After another 12–14 days, colonies were counted. 2.13. Confocal microscopy Samples of Wharton’s Jelly matrix with incorporated MV-4-11 cells were fixed with 4% paraformaldehyde in Hanks Balanced Saline Solution (HBSS), and then cleared using pancreatin solution (0.01 g sucrose, 0.1 g pancreatin, 0.05 g saponin, 0.05% Triton-X100 diluted in HBSS) and placing it into an oven at 34 °C for seven hours. Next, samples were immersed in HBSS overnight at room temperature (RT). After permeabilization with HBSS and 0.1 g saponin (HBSS:S) for ten minutes, samples were immersed in blocking solution (3% normal goat serum and 0.05% Triton-X100 diluted in HBSS:S) and placed on a rotator for two hours at RT. Then, samples were immersed overnight in primary antibody solution of mouse monoclonal anti-Actin α-smooth muscle antibody (Sigma Aldrich) at 4 °C. Samples were rinsed with HBSS:S five times (10 min each) and then incubated with secondary antibody solution using Alexa Fluor 488 goat anti-mouse polyclonal antibody (Life Technologies, ThermoFisher Scientific). All samples were placed in a rotator during four hours of incubation at RT, and then rinsed with HBSS:S two times (ten minutes each), followed by two rinses with HSSS (ten minutes each). 10 μM of the nuclear counterstain DAPI (Life Technologies, ThermoFisher Scientific) was added to each sample for one hour at RT. After three rinses with HBSS, samples were immersed in Vectashield (Vector Laboratories, CA) mounting medium for fluorescence overnight and mounted on coverslips with silicon isolators (Grace Bio-labs, PC1R-2.5). For negative control, samples were processed following the same procedure, but without the primary antibody. Images were collected on a customized spinning disc Olympus IX-81 inverted microscope, equipped with CSU-10 (Yokogawa Electric Company); 405 nm, 488 nm, 642 nm (Coherent, Inc) and 561 nm (CrystaLaser) lasers; a Prior (Prior Scientific) stage; a Sutter (Sutter Instruments) emission filter wheel; and UPlanSApo 20 × 0.75NA air and a UPlanFL N 40 × 1.3NA oil Olympus objectives. Images were acquired using SlideBook 6 (Intelligent Imaging Innovations, Inc.), and data was deconvolved using SlideBook’s constrained iterative deconvolution algorithm.
3.2. Leukemia cells in DWJM developed spindle cell-shaped morphology Since leukemia cell interactions with ECM matrix resulted in reduced cell proliferation, we wondered if leukemia cell-matrix interactions might have caused morphologic changes in leukemia cells cultured in DWJM. By examining histology sections, we noticed that leukemia cells (HL 60 and Kasumi I) cultured in DWJM tended to change their morphology, switching from round cells, which is the morphology of AML cells in suspension, to spindle-shaped cells (Fig. 2A and B). In general, there were more spindle-shaped cells than round cells per high power field (HPF) (Fig. 2C). Round cells are mostly seen in the open spaces within the matrix, while the spindle shaped cells are mostly embedded inside the matrix. These morphologic changes were also visualized by confocal microscopy in actin-stained cells in fixed tissue (Fig. 2D). 3.3. Leukemia cells cultured in DWJM demonstrated an increase in the ALDH positive population Data suggest that AML cells with ALDH positivity are associated with quiescent state and resistance to chemotherapy treatment. [12] Since our data suggested that leukemia cells cultured in DWJM were more quiescent than leukemia cells cultured in suspension, we examined ALDH expression in our model, hypothesizing that leukemia cells cultured in DWJM were associated with increased ALDH expression compared to leukemia cells cultured in suspension. Indeed, significantly increased ALDH expression was found in both Kasumi I (∼7 fold, p < 0.05) and MV411 (∼2 fold, p < 0.05) cells cultured in DWJM compared with cells in suspension (Fig. 3A), while no differences were seen in HL60 cells (data not shown).
2.14. Statistical analysis
3.4. Leukemia cells cultured in DWJM demonstrated an increase in leukemia cell clonogenic ability
All data analyses were performed with Graphpad Prism 6 (GraphPad Software, Inc.) and presented as means ± standard deviation (SEM). Significance was determined using Student’s t-test, with a
Our data indicated that leukemia cells cultured in DWJM exhibited some stem and progenitor cell characteristics. Accordingly, we next 108
Leukemia Research 72 (2018) 105–112
D. Li et al.
Fig. 2. Leukemia cell morphologic changes during culture in DWJM. (A) Representative H&E stained histology sections of HL60 (left), Kasumi 1 (middle) and MV411 (right) cells cultured in DWJM for one week (40x). (B) Kasumi 1 cells change from round to spindle-shaped. Left: round; Middle: roundspindle; Right: spindle (100x). (C) Number of different leukemia cell shapes per high power field (HPF). Data represent means ± SEM. (D) Confocal microscopy images of leukemia cells cultured in DWJM following actin-staining (green) and Dapi nuclear staining (blue) (40×).(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).
the majority of the cultured cells. As expected, doxorubicin induced cell death mainly through apoptosis. We observed that the survival rate of cells cultured in DWJM was significantly higher than that of suspension cells in all three cell lines HL60 (∼2.5 fold, p < 0.5), Kasumi I (∼8 fold, p < 0.05), and MV411 (∼5 fold, p < 0.1) (Fig. 5 A and B), suggesting that leukemia cells cultured in DWJM are more resistant to cytotoxicity of anti-cancer drugs. To examine whether reduced uptake or increased efflux of chemotherapeutic drugs in DWJM-cultured cells might explain their relative resistance to chemotherapy, we analyzed doxorubicin fluorescence intensity in leukemia cells. We found that the accumulation of doxorubicin was significantly lower in leukemia cells cultured in DWJM compared to cells in suspension (Fig. 5C). Because N-cadherin has been reported to be a LSC marker and to be associated with stem cell drug resistance [13], we assessed N-cadherin expression in leukemia cells under DWJM and suspension conditions. We found that leukemia cells cultured in DWJM had increased N-cadherin expression compared to cells cultured in suspension (Fig. 5D).
examined the clonogenic ability of leukemia cells (which correlates with self-renewal ability) comparing cells cultured in DWJM to cells in suspension, using CFU assay with secondary replating. In all three cell lines, colony number from both primary and secondary plating showed significantly increased colony numbers in DWJM-cultured leukemia cells compared to suspension cells (Fig. 4A–C). In two cases, the increase approached statistical significance in Kasumi 1 after primary plating (Fig. 4C, p = 0.08) and MV411 after secondary plating (Fig. 4B, p = 0.051). These findings suggest that leukemia cells cultured in DWJM gain the long-term ability to self-renew.
3.5. Leukemia cells cultured in DWJM demonstrated increased drug resistance Since leukemia cells cultured in DWJM demonstrated LSC-like characteristics and since LSCs are chemoresistant, we next investigated doxorubicin’s effects on leukemia cells cultured in DWJM versus cells in suspension, using a very high dose of doxorubicin to cause apoptosis in 109
Leukemia Research 72 (2018) 105–112
D. Li et al.
Fig. 3. Leukemia cells cultured in DWJM demonstrate an increase in ALDH positive population. Representative density plots (A) and quantification (B) of ALDH positive MV411 and Kasumi 1 cells after 7 day culture in DWJM vs suspension. Data represent means ± SEM. Experiments done in at least duplicate.
Fig. 4. Leukemia cells cultured in DWJM demonstrate an increase in leukemia cell clonogenic ability. HL60 (A), MV411 (B), and Kasumi 1 (C) were cultured in DWJM or in suspension for 7 days. An equal number of cells were plated in methylcellulose for 10–14 days. Cells were replated in methylcellulose for another 10–14 days. The number of colony forming units (CFUs) after primary (upper) and secondary plating (lower) of HL60 (A), MV411 (B), and Kasumi 1 (C) cells were measured. Data represent means ± SEM. Experiments done in triplicate.*P < 0.05; **P < 0.01.
110
Leukemia Research 72 (2018) 105–112
D. Li et al.
Fig. 5. Leukemia cells cultured in DWJM demonstrate increased drug resistance. (A) Representative density plots and (B) quantification of HL-60, Kasumi I and MV411 cells undergoing apoptosis and necrosis in either DWJM or suspension, measured after 50μM doxorubicin treatment for 48 h. Data represent means ± SEM. Experiments done in triplicate. *P < 0.05; **P < 0.01. (C) Doxorubicin uptake in HL60 (left) and MV411 (right) cells after doxorubicin treatment. (D) N-cadherin expression in HL60 (left) and MV411 (right) after 7 days culture in suspension (blue) and DWJM (red).(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).
111
Leukemia Research 72 (2018) 105–112
D. Li et al.
4. Discussion
help with confocal microscopy studies. This work was partly supported by the Robert K. Dempski Cord Blood Research Fund.
In this study, we examined DWJM as a 3D ECM-based model to learn about leukemia cell behavior. Our findings support the hypothesis that an in vitro matrix modeling the in vivo ECM maintains leukemia cell quiescence and LSC traits. In support of this interpretation, we found that leukemia cells maintained in DWJM showed an increase in ALDH+ population in both Kasumi 1 and MV411 cell lines. Since ALDH activity has been reported to be increased in LSCs in bone marrow samples of AML patients [14], our findings suggest that our ECM model system favored the LSC-like phenotype. In addition, we used serial CFU assays after both primary plating and secondary replating to demonstrate increased clonogenic ability of leukemia cells cultured in DWJM, which is a feature of LSCs required for the long-term disease maintenance. Chemoresistance is a major characteristic of cancer stem cells, including LSCs, and is also an significant obstacle to successful chemotherapy [15]. Prior work has demonstrated that different components in the microenvironment play an important role in inducing cancer cell drug resistance [16]. For example, previous studies found that different types of cancer cells demonstrated increased drug resistance when cultured in 3D collagen gels [17,18]. Others have shown that lymphoma cell adhesion to fibronectin resulted in acquired resistance to mitoxantrone [19]. Hyaluronan is also associated with leukemia drug resistance [20]. Since collagen, fibronectin, and hylauronan are present in DWJM, we partially attribute the drug resistance phenotype in DWJM-cultured leukemia cells to interaction of leukemia cells with these components of DWJM. It is possible that this interaction induces a drug resistant phenotype in leukemia cells cultured in DWJM by decreasing the concentration of doxorubicin in leukemia cells, which could be related to decreased uptake or increased efflux of doxorubicin. In addition, these interactions may have resulted in enhanced N-cadherin expression by leukemia cells cultured in DWJM, a property also associated with adhesion-induced chemoresistance [13. Finally, others have shown that spindle-shaped cells in leukemia are associated with chemoresistence [21]. Our observation that our model enriched for spindle-shaped leukemia cells is consistent with this finding. In our experimental design, we compared our ECM-based culture system to traditional suspension culture conditions. However, other collagen-based 3-dimensional culture systems are available. These include collagen gel [22] and Histoculture [23,24], in which 3D tissue pieces are put in growth medium with collagen gel support or freely floating without support. Because of its ability to maintain the original tissue phenotype, Histoculture has been used in drug screening for different types of cancer in clinical trials. One advantage to our system compared to collagen-only based systems is that DWJM has other components that are present in the bone marrowECM, including fibronectin, hyaluronic acid, and sulphated proteoglycan. In addition, our DWJM-based model recapitulates some of the features of the LSC. Since currently many drugs targeting LSCs are under development, our DWJM-based model will potentially provide useful for screening of LSCtargeted therapeutics. Future studies will focus on examining DWJM vs other commercially available collagen-based 3D culture systems. In conclusion, our DWJM-based ECM model maintains leukemia cells with stem cell-like characteristics. DWJM should be further examined as a platform for leukemia research.
Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.leukres.2018.08.012. References [1] J.C. Wang, J.E. Dick, Cancer stem cells: lessons from leukemia, Trends Cell Biol. 15 (September (5)) (2005) 494–501. [2] K.J. Hope, L. Jin, J.E. Dick, Acute myeloid leukemia originates from a hierarchy of leukemic stem cell classes that differ in self-renewal capacity, Nat. Immunol. (July (7)) (2004) 738–743. [3] G.G. Wulf, R.Y. Wang, I. Kuehnle, D. Weidner, F. Marini, M.K. Brenner, et al., A leukemic stem cell with intrinsic drug efflux capacity in acute myeloid leukemia, Blood 98 (4) (2001) 1166–1173. [4] C. Lo Celso, H.E. Fleming, J.W. Wu, C.X. Zhao, S. Miake-Lye, J. Fujisaki, et al., Liveanimal tracking of individual haematopoietic stem/progenitor cells in their niche, Nature 457 (January (7225)) (2009) 92–96. [5] S.W. Lane, D.T. Scadden, D.G. Gilliland, The leukemic stem cell niche: current concepts and therapeutic opportunities, Blood 114 (August (6)) (2009) 1150–1157. [6] G.L. Converse, D. Li, E.E. Buse, R.A. Hopkins, O.S. Aljitawi, Wharton’s jelly matrix decellularization for tissue engineering applications, Methods Mol. Biol. (2017). [7] O.S. Aljitawi, Y. Xiao, D. Zhang, L. Stehno-Bittel, R. Garimella, R.A. Hopkins, et al., Generating CK19-positive cells with hair-like structures from Wharton’s jelly mesenchymal stromal cells, Stem Cells Dev. 22 (January (1)) (2013) 18–26. [8] S. Jadalannagari, G. Converse, C. McFall, E. Buse, M. Filla, M.T. Villar, et al., Decellularized Wharton’s Jelly from human umbilical cord as a novel 3D scaffolding material for tissue engineering applications, PLoS One 12 (2) (2017) e0172098. [9] D. Estrugo, A. Fischer, F. Hess, H. Scherthan, C. Belka, N. Cordes, Ligand bound beta1 integrins inhibit procaspase-8 for mediating cell adhesion-mediated drug and radiation resistance in human leukemia cells, PLoS One 2 (3) (2007) e269. [10] G.L. Converse, Wharton’s Jelly Matrix Decellularization for Tissue Engineering Applications, (2017) August. [11] S. Dziennis, R.A. Van Etten, H.L. Pahl, D.L. Morris, T.L. Rothstein, C.M. Blosch, et al., The CD11b promoter directs high-level expression of reporter genes in macrophages in transgenic mice, Blood 85 (January (2)) (1995) 319–329. [12] V.T. Hoang, E.C. Buss, W. Wang, I. Hoffmann, S. Raffel, A. Zepeda-Moreno, et al., The rarity of ALDH(+) cells is the key to separation of normal versus leukemia stem cells by ALDH activity in AML patients, Int. J. Cancer 137 (3) (2015) 525–536. [13] L. Zhi, M. Wang, Q. Rao, F. Yu, Y. Mi, J. Wang, Enrichment of N-Cadherin and Tie2bearing CD34+/CD38-/CD123+ leukemic stem cells by chemotherapy-resistance, Cancer. Lett. 296 (1) (2010) 65–73. [14] J.M. Gerber, B.D. Smith, B. Ngwang, H. Zhang, M.S. Vala, L. Morsberger, et al., A clinically relevant population of leukemic CD34(+)CD38(-) cells in acute myeloid leukemia, Blood 119 (April (15)) (2012) 3571–3577. [15] C. Holohan, S. Van Schaeybroeck, D.B. Longley, P.G. Johnston, Cancer drug resistance: an evolving paradigm. Nature reviews, Cancer 13 (October (10)) (2013) 714–726. [16] M.Y. Konopleva, C.T. Jordan, Leukemia stem cells and microenvironment: biology and therapeutic targeting, J. Clin. Oncol. 29 (February (5)) (2011) 591–599. [17] S. Dangi-Garimella, V. Sahai, K. Ebine, K. Kumar, H.G. Munshi, Three-dimensional collagen I promotes gemcitabine resistance in vitro in pancreatic cancer cells through HMGA2-dependent histone acetyltransferase expression, PLoS One 8 (5) (2013) e64566. [18] B.M. Leung, C. Moraes, S.P. Cavnar, K.E. Luker, G.D. Luker, S. Takayama, Microscale 3D collagen cell culture assays in conventional flat-bottom 384-well plates, J. Lab. Autom. 20 (April (2)) (2018) 138–145. [19] L.A. Hazlehurst, R.F. Argilagos, M. Emmons, D. Boulware, C.A. Beam, D.M. Sullivan, et al., Cell adhesion to fibronectin (CAM-DR) influences acquired mitoxantrone resistance in U937 cells, Cancer Res. 66 (Febraury (4)) (2006) 2338–2345. [20] S.L. Lompardia, D.L. Papademetrio, M. Mascaro, E.M. Alvarez, S.E. Hajos, Human leukemic cell lines synthesize hyaluronan to avoid senescence and resist chemotherapy, Glycobiology 23 (December (12)) (2013) 1463–1476. [21] L.H. Mochmann, M. Neumann, E.K. von der Heide, V. Nowak, A.A. Kuhl, J. OrtizTanchez, et al., ERG induces a mesenchymal-like state associated with chemoresistance in leukemia cells, Oncotarget 5 (2) (2014) 351–362. [22] A.E. Freeman, R.M. Hoffman, In vivo-like growth of human tumors in vitro, Proc. Natl. Acad. Sci. U. S. A. 83 (8) (1986) 2694–2698. [23] R. Hoffman, Tissue culture, 2nd edn, in: E-i-CS Maloy (Ed.), Brenner’S Encyclopedia of Genetics, vol. 7, Academic Press, 2013, pp. 73–76. [24] R.M. Hoffman, Histocultures and their use, Encyclopedia of Life Sciences (ELS), John Wiley & Sons, Ltd., Chichester, 2010 October.
Acknowledgments We thank Constance D. Baldwin, PhD for reviewing the manuscript for clarity. We thank Dr. Linheng Li for his advice in studying N-cadherin expression in our model. We also thank Dr. Fariba Behbod for her help with ALDH-related studies and Dr. Eduardo Rosa-Molinar for his
112
Contents lists available at ScienceDirect
Leukemia Research journal homepage: www.elsevier.com/locate/leukres
Research paper
A novel extracellular matrix-based leukemia model supports leukemia cells with stem cell-like characteristics
T
Dandan Lia, Tara L. Linb, Brea Lipeb, Richard A. Hopkinsc, Heather Shinogled, ⁎ Omar S. Aljitawia,b,e, a
Department of Pathology and Laboratory Medicine, University of Kansas Medical Center, Kansas City, KS, United States Division of Hematology/Oncology and Blood and Marrow Transplantation Program, 2330 Shawnee Mission Parkway, University of Kansas Medical Center, Kansas City, KS, United States c Cardiac Surgery Research Laboratories, Children's Mercy Hospital and Clinics, Kansas City, Missouri, United States d Microscopy and Analytical Imaging Laboratory, University of Kansas, Lawrence, KS, United States e Department of Medicine, Hematology/Oncology and Bone Marrow Transplant Program, University of Rochester Medical Center, Rochester, NY, 14642, United States b
A R T I C LE I N FO
A B S T R A C T
Keywords: Extracellular matrix In vitro 3D model Leukemia Leukemia stem cell-like Decellularized Wharton's jelly matrix
Acute myeloid leukemia (AML) relapse results from the survival of chemotherapy-resistant and quiescent leukemia stem cells (LSC). These LSCs reside in the bone marrow microenvironment, comprised of other cells and extracellular matrix (ECM), which facilitates LSC quiescence through expression of cell adhesion molecules. We used decellularized Wharton’s jelly matrix (DWJM), the gelatinous material in the umbilical cord, as a scaffolding material to culture leukemia cells, because it contains many components of the bone marrow extracellular matrix, including collagen, fibronectin, lumican, and hyaluronic acid (HA). Leukemia cells cultured in DWJM demonstrated decreased proliferation without undergoing significant differentiation. After culture in DWJM, these cells also exhibited changes in morphology, acquiring a spindle-shaped appearance, and an increase in the ALDH+ cell population. When treated with a high-dose of doxorubicin, leukemia cells in DWJM demonstrated less apoptosis compared with cells in suspension. Serial colony forming unit (CFU) assays indicated that leukemia cells cultured in DWJM showed increased colony-forming ability after both primary and secondary plating. Leukemia cell culture in DWJM was associated with increased N-cadherin expression by flow cytometry. Our data suggest that DWJM could serve as an ECM-based model to study AML stem cell-like cell behavior and chemotherapy sensitivity.
1. Introduction Acute myeloid leukemia (AML) is a heterogeneous hematopoietic malignancy characterized by an aberrant clonal expansion of undifferentiated myeloid blasts. Studies have shown that leukemia stem cells (LSCs) contribute to relapse after chemotherapeutic treatment. [1] Like normal hematopoietic stem cells (HSCs), LSCs maintain their selfrenewal ability while generating clonogenic leukemic progenitors capable of producing leukemic cells [2]. Anti-proliferative chemotherapeutic agents commonly target the rapidly cycling leukemic cells, but they generally are ineffective against the quiescent LSCs, partly because of enhanced drug efflux in LSCs [3]. Therefore, it is important to develop therapeutic strategies which eliminate the LSCs in the bone marrow, where they share the “hematopoietic niche” along with normal HSCs [4]. The LSC niche, similar to the hematopoietic niche, is a 3D
⁎
microenvironment composed of bone marrow stromal cells and ECM components like collagen, fibronectin and tenascin [4]. These components create compartments that not only provide structural support to the cells in the bone marrow, but also provide chemokines and cytokines that are important in regulating LSC self-renewal, trafficking, proliferation and differentiation [5]. Currently, most leukemia in vitro studies are based on conventional two-dimensional (2D) cultures in tissue culture polystyrene (TCP) dishes/ flasks and stromal co-cultures. These models are useful in elucidating some of the molecular mechanisms of leukemia initiation and progression. However, 2D culture systems lack the leukemia-microenvironment interaction present in the 3D bone marrow microenvironment. Therefore, the LSCs in 2D culture frequently differentiate and lose their “stem-ness”. The development of a 3D model that replicates the in vivo mechanical and biochemical properties of bone
Corresponding author at: 601 Elmwood Avenue, Rochester, NY, 14642, United States. E-mail address: [email protected] (O.S. Aljitawi).
https://doi.org/10.1016/j.leukres.2018.08.012 Received 15 March 2018; Received in revised form 12 August 2018; Accepted 13 August 2018 Available online 16 August 2018 0145-2126/ © 2018 Elsevier Ltd. All rights reserved.
Leukemia Research 72 (2018) 105–112
D. Li et al.
times, 4 h prior to proliferation assessment. 10% alamarBlue (Biocentric) was added into each well. After 4 h, 100 μl of supernatant from each well was aspirated to a new well of a 96 well plate and fluorescence was measured by micro-plate reader, with excitation wavelength at 530 nm and emission wavelength at 590 nm.
marrow could allow maintenance of a true LSC-state. Our laboratory has been focused on characterizing decellularized Wharton’s jelly matrix (DWJM) and examining its potential for regenerative medicine applications. [6–8], We hypothesized that DWJM would provide a similar environment to the bone marrow ECM, because its components, such as collagen, fibronectin, hyaluronic acid, and sulphated proteoglycan [8], also exist in the bone marrow hematopoietic niche. Moreover, because cell-ECM interactions play an important role in chemoresistance in leukemia cells [9], the environment provided by DWJM could support the maintenance of LSCs. We therefore used DWJM as an ECM to examine leukemia cell-ECM interactions, hypothesizing that DWJM would support leukemia cells with LSC-like characteristics. In this model, we investigated the growth pattern of 3 human leukemia cell lines (HL60, Kasumi-1, and MV411), with a focus on proliferation, viability, morphology and myeloid differentiation. We also studied the drug resistance and stem cell characteristics of leukemia cells cultured in this model, compared to leukemia cells cultured in suspension. We found that leukemia cells cultured in our DWJM-based ECM model had LSC characteristics, suggesting that DWJM may prove useful in LSC characterization and in developing therapeutic interventions that target LSCs.
2.5. CellTrace proliferation assay To monitor the cell division of leukemia cells in suspension and in DWJM, cells were labeled with CellTrace Violet (Life Technology) before seeding. Briefly, cells were washed and resuspended with PBS at the concentration of 106cells/ml, and CellTrace Violet stock solution was added in a final concentration of 1 μl/ml. After incubation at room temperature for 20 min, 5 ml of PBS with 10% FBS were added and incubated for 5 min, followed by centrifugation to obtain pellets. Cells were resuspended in culture medium and cultured in either suspension or DWJM. Cell division was measured by flow cytometry soon after seeding and after 7 days of culture. To isolate cells from DWJM, we washed wafers in PBS, and then used collagenase II (0.05 g collagenase II in 50 ml DMEM for 1–2 hours) to digest DWJM at 37 °C. 2.6. Cell viability
2. Materials and methods Cell survival in DWJM was measured by Vi-CELL Series Cell Viability Analyzer (Beckman Coulter), which is based on Trypan Blue dye exclusion. Cells in each DWJM wafer were released by treating with 1 ml 0.002 g/ml collagenase II (Worthington) for about 2 h at 37 °C. The released cells were assessed for viability according to the manufacturer’s recommendations.
2.1. Cell culture Human AML cell lines HL60, Kasumi I and MV 411 (ATCC, Manassas, VA) were maintained in T 75 tissue culture flasks with Advanced Roswell Park Memorial Institute (RPMI) 1640 Medium (Gibco), supplemented with 5% fetal bovine serum (Sigma-Aldrich) and 1% penicillin/streptomycin (pen/strep) (Life Technologies). Cells were maintained at 37 °C in a fully humidified 5% CO2 incubator.
2.7. Histology and immunohistochemistry For morphological analysis, wafers were washed with PBS three times, fixed in 4% PFA, embedded in paraffin, sectioned and stained with hematoxylin and eosin, and visualized under the microscope using an Olympus BX40 microscope; pictures were obtained using a DP72 digital camera.
2.2. DWJM scaffold preparation The preparation of DWJM was previously described. [8,10], Briefly, to prepare DWJM scaffolds, fresh human umbilical cords were dissected after removing surrounding membranes and blood vessels. Then they were subjected to two cycles of osmotic shock in hypertonic and hypotonic solutions, followed by immersion in a non-ionic detergent Triton-x, an anionic detergent sodium lauryl succinate; finally, they underwent enzymatic digestion with recombinant endonuclease. The resulting DWJM pieces were cut into thin wafers (3 mm thick) as previously described [10] (Supplementary Figure-01A). Cartoon depiction of leukemia cells interacting with DWJM fibers (Supplementary Figure01B).
2.8. Treatment with chemotherapeutic agents After 7 days, cells cultured in suspension and cells in DWJM were treated with 50μM of doxorubicin hydrochloride (Sigma-Aldrich) for 48 h. For cells in suspension, culture medium was removed, and chemotherapeutic agents were added in fresh medium. For cells in DWJM, scaffolds were transferred into new 24-well plates and washed with PBS three times; then a chemotherapeutic agent was added to the culture medium.
2.3. Seeding DWJM wafers with AML cell lines 2.9. Apoptosis assay Before seeding, cryopreserved DWJM wafers were thawed, washed three times in phosphate buffered saline (PBS), and pre-incubated with Advanced RPMI overnight. AML cells (2*105 cells/well) were seeded into DWJM wafers in 24-well non-tissue culture treated plates with 60% area of each well covered by DWJM. Culture plates were then placed in an incubator at 37 °C with 5% CO2 and maintained in Advanced RPMI with 5% FBS for 7 days; half of the medium was changed every other day. AML cells in suspension (2*105 cells/well), cultured under the same conditions, were used as controls. In AlamarBlue assay, CellTrace proliferation and Ki67 immunohistology sample preparation, cells were maintained in RPMI 1640 (Sigma-Aldrich) with 10% FBS for HL60 and MV411 cells, and 20% FBS for Kasumi I cells.
Apoptosis in leukemia cells was measured by flow cytometry using Annexin V-Alexa 568 (Invitrogen, USA) and DAPI (Invitrogen, USA) staining. Prior to flow cytometry analysis, cells in DWJM wafers were released as described previously and 105 released cells as well as cells in suspension were stained with DAPI and Annexin V according to manufacturer’s recommendations. Data were acquired within 1 h, using LSR II (BD Biosciences), and analyzed by FlowJo software. 2.10. Aldefluor assay Aldehyde dehydrogenase (ALDH) activity was examined by using Aldefluor reagent (Stem Cell Technologies) according to the manufacturer’s protocol, followed by flow cytometry. Cells negative for propidium iodide (PI) staining were considered positive for ALDH, based on a negative control using the ALDH inhibitor diethylaminobenzaldehyde (DEAB). Data were analyzed within 1 h, using LSR II (BD
2.4. AlamarBlue assay To assess the proliferation of AML cells, DWJM wafers with cells were transferred to new 24 well plates and washed with PBS three 106
Leukemia Research 72 (2018) 105–112
D. Li et al.
Fig. 1. Leukemia cells cultured in DWJM demonstrate reduced proliferation while maintaining cell viability and undifferentiated state. (A) HL60, Kasumi 1 and MV411 cells were seeded onto DWJM wafers and allowed to adhere to DWJM for 2 days. Then DWJM wafers with attached cells were replated, and AlamarBlue assay was used to assess cell proliferation on day 2, 4, and 6. Data are normalized to the day 2 fluorescence reading. (B) Collagenase II was used to release cells embedded in DWJM, and cell viability was assessed by Vi-CELL, based on Trypan blue exclusion on days 2, 4, and 6. All values represent means ± SEM. (C) Violet intensity of HL60 and MV411 in DWJM and suspension before and 7 days after seeding. (D) Apoptosis and necrosis of HL60 and MV411 cultured in suspension or in DWJM, as measured by flow cytometry following co-staining with AnnexinV/ PI. Results are shown as density plots of one representative experiment (upper) and as summary of results of multiple experiments (bottom). Data represent means ± SEM. Experiments done in triplicate. (E) CD11b expression in HL60 and Kasumi 1 cultured in suspension and DWJM.
107
Leukemia Research 72 (2018) 105–112
D. Li et al.
Biosciences).
statistical threshold of p < 0.05.
2.11. Analysis of differentiation marker
3. Results
Expression of CD11b was measured by flow cytometry. 105 cells, cultured in suspension or in DWJM, were harvested and incubated with FITC-conjugated anti-human CD11b (Abcam) for 10 min at room temperature. After staining, cells were washed and resuspended in 400 μl PBS and analyzed by flow cytometry by using LSR II (BD Biosciences).
3.1. Leukemia cells cultured in DWJM demonstrated reduced proliferation while maintaining cell viability and undifferentiated state Since LSCs are quiescent, we first examined leukemia cell proliferation over time in our model, hypothesizing that leukemia-DWJM interactions result in decreased leukemia cell proliferation over time. Using AlamarBlue assay on days 2, 4, and 6 after leukemia cell seeding, we demonstrated decreased proliferation in all three leukemia cell lines (HL 60, Kasumi I and MV411) (Fig. 2A). Despite reduced proliferation, all three cell lines maintained nearly unchanged viability at the three time points, as measured by Trypan Blue dye exclusion (Fig. 1B). Next, we studied leukemia cell proliferation and viability in our model, compared to suspension culture systems. We found that leukemia cells cultured in DWJM demonstrated reduced proliferation measured by CellTrace proliferation assay (Fig. 1C), but no significant differences in apoptosis between the two culture conditions when assessed by Annexin V/PI flow cytometry (Fig. 1D). LSCs are expected to maintain an undifferentiated state. We next examined DWJM culture effects on leukemia cell differentiation by examining expression of CD11b, a common myeloid marker [11], using flow cytometry. Our experiments showed no increase in CD11b expression in either suspension or DWJM-culture conditions, indicating that both conditions maintain leukemia cells in an undifferentiated state (Fig. 1E). Taken together, our data suggest that leukemia cells cultured in DWJM are more quiescent than leukemia cells in suspension.
2.12. Colony forming unit (CFU) assay Cells were plated on 35 mm dishes (500 cells/dish for Kasumi I, and 300 cells/dish for HL60 and MV411) in triplicate in MethoCult® H4434 Classic methylcellulose (Stemcell Technologies, Vancouver, Canada). Cells were washed and resuspended in RPMI 1640 medium. After measuring cell number and viability by Trypan blue, cell densities were adjusted at a concentration of 10 cells/μl, and cells were added into methylcellulose, followed by addition of 1 ml cell-methylcellulose mix into each dish. After 12–14 days of incubation at 37 °C in 5% CO2, colonies consisting of > 30 cells were counted, then harvested and replated in methylcellulose. After another 12–14 days, colonies were counted. 2.13. Confocal microscopy Samples of Wharton’s Jelly matrix with incorporated MV-4-11 cells were fixed with 4% paraformaldehyde in Hanks Balanced Saline Solution (HBSS), and then cleared using pancreatin solution (0.01 g sucrose, 0.1 g pancreatin, 0.05 g saponin, 0.05% Triton-X100 diluted in HBSS) and placing it into an oven at 34 °C for seven hours. Next, samples were immersed in HBSS overnight at room temperature (RT). After permeabilization with HBSS and 0.1 g saponin (HBSS:S) for ten minutes, samples were immersed in blocking solution (3% normal goat serum and 0.05% Triton-X100 diluted in HBSS:S) and placed on a rotator for two hours at RT. Then, samples were immersed overnight in primary antibody solution of mouse monoclonal anti-Actin α-smooth muscle antibody (Sigma Aldrich) at 4 °C. Samples were rinsed with HBSS:S five times (10 min each) and then incubated with secondary antibody solution using Alexa Fluor 488 goat anti-mouse polyclonal antibody (Life Technologies, ThermoFisher Scientific). All samples were placed in a rotator during four hours of incubation at RT, and then rinsed with HBSS:S two times (ten minutes each), followed by two rinses with HSSS (ten minutes each). 10 μM of the nuclear counterstain DAPI (Life Technologies, ThermoFisher Scientific) was added to each sample for one hour at RT. After three rinses with HBSS, samples were immersed in Vectashield (Vector Laboratories, CA) mounting medium for fluorescence overnight and mounted on coverslips with silicon isolators (Grace Bio-labs, PC1R-2.5). For negative control, samples were processed following the same procedure, but without the primary antibody. Images were collected on a customized spinning disc Olympus IX-81 inverted microscope, equipped with CSU-10 (Yokogawa Electric Company); 405 nm, 488 nm, 642 nm (Coherent, Inc) and 561 nm (CrystaLaser) lasers; a Prior (Prior Scientific) stage; a Sutter (Sutter Instruments) emission filter wheel; and UPlanSApo 20 × 0.75NA air and a UPlanFL N 40 × 1.3NA oil Olympus objectives. Images were acquired using SlideBook 6 (Intelligent Imaging Innovations, Inc.), and data was deconvolved using SlideBook’s constrained iterative deconvolution algorithm.
3.2. Leukemia cells in DWJM developed spindle cell-shaped morphology Since leukemia cell interactions with ECM matrix resulted in reduced cell proliferation, we wondered if leukemia cell-matrix interactions might have caused morphologic changes in leukemia cells cultured in DWJM. By examining histology sections, we noticed that leukemia cells (HL 60 and Kasumi I) cultured in DWJM tended to change their morphology, switching from round cells, which is the morphology of AML cells in suspension, to spindle-shaped cells (Fig. 2A and B). In general, there were more spindle-shaped cells than round cells per high power field (HPF) (Fig. 2C). Round cells are mostly seen in the open spaces within the matrix, while the spindle shaped cells are mostly embedded inside the matrix. These morphologic changes were also visualized by confocal microscopy in actin-stained cells in fixed tissue (Fig. 2D). 3.3. Leukemia cells cultured in DWJM demonstrated an increase in the ALDH positive population Data suggest that AML cells with ALDH positivity are associated with quiescent state and resistance to chemotherapy treatment. [12] Since our data suggested that leukemia cells cultured in DWJM were more quiescent than leukemia cells cultured in suspension, we examined ALDH expression in our model, hypothesizing that leukemia cells cultured in DWJM were associated with increased ALDH expression compared to leukemia cells cultured in suspension. Indeed, significantly increased ALDH expression was found in both Kasumi I (∼7 fold, p < 0.05) and MV411 (∼2 fold, p < 0.05) cells cultured in DWJM compared with cells in suspension (Fig. 3A), while no differences were seen in HL60 cells (data not shown).
2.14. Statistical analysis
3.4. Leukemia cells cultured in DWJM demonstrated an increase in leukemia cell clonogenic ability
All data analyses were performed with Graphpad Prism 6 (GraphPad Software, Inc.) and presented as means ± standard deviation (SEM). Significance was determined using Student’s t-test, with a
Our data indicated that leukemia cells cultured in DWJM exhibited some stem and progenitor cell characteristics. Accordingly, we next 108
Leukemia Research 72 (2018) 105–112
D. Li et al.
Fig. 2. Leukemia cell morphologic changes during culture in DWJM. (A) Representative H&E stained histology sections of HL60 (left), Kasumi 1 (middle) and MV411 (right) cells cultured in DWJM for one week (40x). (B) Kasumi 1 cells change from round to spindle-shaped. Left: round; Middle: roundspindle; Right: spindle (100x). (C) Number of different leukemia cell shapes per high power field (HPF). Data represent means ± SEM. (D) Confocal microscopy images of leukemia cells cultured in DWJM following actin-staining (green) and Dapi nuclear staining (blue) (40×).(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).
the majority of the cultured cells. As expected, doxorubicin induced cell death mainly through apoptosis. We observed that the survival rate of cells cultured in DWJM was significantly higher than that of suspension cells in all three cell lines HL60 (∼2.5 fold, p < 0.5), Kasumi I (∼8 fold, p < 0.05), and MV411 (∼5 fold, p < 0.1) (Fig. 5 A and B), suggesting that leukemia cells cultured in DWJM are more resistant to cytotoxicity of anti-cancer drugs. To examine whether reduced uptake or increased efflux of chemotherapeutic drugs in DWJM-cultured cells might explain their relative resistance to chemotherapy, we analyzed doxorubicin fluorescence intensity in leukemia cells. We found that the accumulation of doxorubicin was significantly lower in leukemia cells cultured in DWJM compared to cells in suspension (Fig. 5C). Because N-cadherin has been reported to be a LSC marker and to be associated with stem cell drug resistance [13], we assessed N-cadherin expression in leukemia cells under DWJM and suspension conditions. We found that leukemia cells cultured in DWJM had increased N-cadherin expression compared to cells cultured in suspension (Fig. 5D).
examined the clonogenic ability of leukemia cells (which correlates with self-renewal ability) comparing cells cultured in DWJM to cells in suspension, using CFU assay with secondary replating. In all three cell lines, colony number from both primary and secondary plating showed significantly increased colony numbers in DWJM-cultured leukemia cells compared to suspension cells (Fig. 4A–C). In two cases, the increase approached statistical significance in Kasumi 1 after primary plating (Fig. 4C, p = 0.08) and MV411 after secondary plating (Fig. 4B, p = 0.051). These findings suggest that leukemia cells cultured in DWJM gain the long-term ability to self-renew.
3.5. Leukemia cells cultured in DWJM demonstrated increased drug resistance Since leukemia cells cultured in DWJM demonstrated LSC-like characteristics and since LSCs are chemoresistant, we next investigated doxorubicin’s effects on leukemia cells cultured in DWJM versus cells in suspension, using a very high dose of doxorubicin to cause apoptosis in 109
Leukemia Research 72 (2018) 105–112
D. Li et al.
Fig. 3. Leukemia cells cultured in DWJM demonstrate an increase in ALDH positive population. Representative density plots (A) and quantification (B) of ALDH positive MV411 and Kasumi 1 cells after 7 day culture in DWJM vs suspension. Data represent means ± SEM. Experiments done in at least duplicate.
Fig. 4. Leukemia cells cultured in DWJM demonstrate an increase in leukemia cell clonogenic ability. HL60 (A), MV411 (B), and Kasumi 1 (C) were cultured in DWJM or in suspension for 7 days. An equal number of cells were plated in methylcellulose for 10–14 days. Cells were replated in methylcellulose for another 10–14 days. The number of colony forming units (CFUs) after primary (upper) and secondary plating (lower) of HL60 (A), MV411 (B), and Kasumi 1 (C) cells were measured. Data represent means ± SEM. Experiments done in triplicate.*P < 0.05; **P < 0.01.
110
Leukemia Research 72 (2018) 105–112
D. Li et al.
Fig. 5. Leukemia cells cultured in DWJM demonstrate increased drug resistance. (A) Representative density plots and (B) quantification of HL-60, Kasumi I and MV411 cells undergoing apoptosis and necrosis in either DWJM or suspension, measured after 50μM doxorubicin treatment for 48 h. Data represent means ± SEM. Experiments done in triplicate. *P < 0.05; **P < 0.01. (C) Doxorubicin uptake in HL60 (left) and MV411 (right) cells after doxorubicin treatment. (D) N-cadherin expression in HL60 (left) and MV411 (right) after 7 days culture in suspension (blue) and DWJM (red).(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).
111
Leukemia Research 72 (2018) 105–112
D. Li et al.
4. Discussion
help with confocal microscopy studies. This work was partly supported by the Robert K. Dempski Cord Blood Research Fund.
In this study, we examined DWJM as a 3D ECM-based model to learn about leukemia cell behavior. Our findings support the hypothesis that an in vitro matrix modeling the in vivo ECM maintains leukemia cell quiescence and LSC traits. In support of this interpretation, we found that leukemia cells maintained in DWJM showed an increase in ALDH+ population in both Kasumi 1 and MV411 cell lines. Since ALDH activity has been reported to be increased in LSCs in bone marrow samples of AML patients [14], our findings suggest that our ECM model system favored the LSC-like phenotype. In addition, we used serial CFU assays after both primary plating and secondary replating to demonstrate increased clonogenic ability of leukemia cells cultured in DWJM, which is a feature of LSCs required for the long-term disease maintenance. Chemoresistance is a major characteristic of cancer stem cells, including LSCs, and is also an significant obstacle to successful chemotherapy [15]. Prior work has demonstrated that different components in the microenvironment play an important role in inducing cancer cell drug resistance [16]. For example, previous studies found that different types of cancer cells demonstrated increased drug resistance when cultured in 3D collagen gels [17,18]. Others have shown that lymphoma cell adhesion to fibronectin resulted in acquired resistance to mitoxantrone [19]. Hyaluronan is also associated with leukemia drug resistance [20]. Since collagen, fibronectin, and hylauronan are present in DWJM, we partially attribute the drug resistance phenotype in DWJM-cultured leukemia cells to interaction of leukemia cells with these components of DWJM. It is possible that this interaction induces a drug resistant phenotype in leukemia cells cultured in DWJM by decreasing the concentration of doxorubicin in leukemia cells, which could be related to decreased uptake or increased efflux of doxorubicin. In addition, these interactions may have resulted in enhanced N-cadherin expression by leukemia cells cultured in DWJM, a property also associated with adhesion-induced chemoresistance [13. Finally, others have shown that spindle-shaped cells in leukemia are associated with chemoresistence [21]. Our observation that our model enriched for spindle-shaped leukemia cells is consistent with this finding. In our experimental design, we compared our ECM-based culture system to traditional suspension culture conditions. However, other collagen-based 3-dimensional culture systems are available. These include collagen gel [22] and Histoculture [23,24], in which 3D tissue pieces are put in growth medium with collagen gel support or freely floating without support. Because of its ability to maintain the original tissue phenotype, Histoculture has been used in drug screening for different types of cancer in clinical trials. One advantage to our system compared to collagen-only based systems is that DWJM has other components that are present in the bone marrowECM, including fibronectin, hyaluronic acid, and sulphated proteoglycan. In addition, our DWJM-based model recapitulates some of the features of the LSC. Since currently many drugs targeting LSCs are under development, our DWJM-based model will potentially provide useful for screening of LSCtargeted therapeutics. Future studies will focus on examining DWJM vs other commercially available collagen-based 3D culture systems. In conclusion, our DWJM-based ECM model maintains leukemia cells with stem cell-like characteristics. DWJM should be further examined as a platform for leukemia research.
Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.leukres.2018.08.012. References [1] J.C. Wang, J.E. Dick, Cancer stem cells: lessons from leukemia, Trends Cell Biol. 15 (September (5)) (2005) 494–501. [2] K.J. Hope, L. Jin, J.E. Dick, Acute myeloid leukemia originates from a hierarchy of leukemic stem cell classes that differ in self-renewal capacity, Nat. Immunol. (July (7)) (2004) 738–743. [3] G.G. Wulf, R.Y. Wang, I. Kuehnle, D. Weidner, F. Marini, M.K. Brenner, et al., A leukemic stem cell with intrinsic drug efflux capacity in acute myeloid leukemia, Blood 98 (4) (2001) 1166–1173. [4] C. Lo Celso, H.E. Fleming, J.W. Wu, C.X. Zhao, S. Miake-Lye, J. Fujisaki, et al., Liveanimal tracking of individual haematopoietic stem/progenitor cells in their niche, Nature 457 (January (7225)) (2009) 92–96. [5] S.W. Lane, D.T. Scadden, D.G. Gilliland, The leukemic stem cell niche: current concepts and therapeutic opportunities, Blood 114 (August (6)) (2009) 1150–1157. [6] G.L. Converse, D. Li, E.E. Buse, R.A. Hopkins, O.S. Aljitawi, Wharton’s jelly matrix decellularization for tissue engineering applications, Methods Mol. Biol. (2017). [7] O.S. Aljitawi, Y. Xiao, D. Zhang, L. Stehno-Bittel, R. Garimella, R.A. Hopkins, et al., Generating CK19-positive cells with hair-like structures from Wharton’s jelly mesenchymal stromal cells, Stem Cells Dev. 22 (January (1)) (2013) 18–26. [8] S. Jadalannagari, G. Converse, C. McFall, E. Buse, M. Filla, M.T. Villar, et al., Decellularized Wharton’s Jelly from human umbilical cord as a novel 3D scaffolding material for tissue engineering applications, PLoS One 12 (2) (2017) e0172098. [9] D. Estrugo, A. Fischer, F. Hess, H. Scherthan, C. Belka, N. Cordes, Ligand bound beta1 integrins inhibit procaspase-8 for mediating cell adhesion-mediated drug and radiation resistance in human leukemia cells, PLoS One 2 (3) (2007) e269. [10] G.L. Converse, Wharton’s Jelly Matrix Decellularization for Tissue Engineering Applications, (2017) August. [11] S. Dziennis, R.A. Van Etten, H.L. Pahl, D.L. Morris, T.L. Rothstein, C.M. Blosch, et al., The CD11b promoter directs high-level expression of reporter genes in macrophages in transgenic mice, Blood 85 (January (2)) (1995) 319–329. [12] V.T. Hoang, E.C. Buss, W. Wang, I. Hoffmann, S. Raffel, A. Zepeda-Moreno, et al., The rarity of ALDH(+) cells is the key to separation of normal versus leukemia stem cells by ALDH activity in AML patients, Int. J. Cancer 137 (3) (2015) 525–536. [13] L. Zhi, M. Wang, Q. Rao, F. Yu, Y. Mi, J. Wang, Enrichment of N-Cadherin and Tie2bearing CD34+/CD38-/CD123+ leukemic stem cells by chemotherapy-resistance, Cancer. Lett. 296 (1) (2010) 65–73. [14] J.M. Gerber, B.D. Smith, B. Ngwang, H. Zhang, M.S. Vala, L. Morsberger, et al., A clinically relevant population of leukemic CD34(+)CD38(-) cells in acute myeloid leukemia, Blood 119 (April (15)) (2012) 3571–3577. [15] C. Holohan, S. Van Schaeybroeck, D.B. Longley, P.G. Johnston, Cancer drug resistance: an evolving paradigm. Nature reviews, Cancer 13 (October (10)) (2013) 714–726. [16] M.Y. Konopleva, C.T. Jordan, Leukemia stem cells and microenvironment: biology and therapeutic targeting, J. Clin. Oncol. 29 (February (5)) (2011) 591–599. [17] S. Dangi-Garimella, V. Sahai, K. Ebine, K. Kumar, H.G. Munshi, Three-dimensional collagen I promotes gemcitabine resistance in vitro in pancreatic cancer cells through HMGA2-dependent histone acetyltransferase expression, PLoS One 8 (5) (2013) e64566. [18] B.M. Leung, C. Moraes, S.P. Cavnar, K.E. Luker, G.D. Luker, S. Takayama, Microscale 3D collagen cell culture assays in conventional flat-bottom 384-well plates, J. Lab. Autom. 20 (April (2)) (2018) 138–145. [19] L.A. Hazlehurst, R.F. Argilagos, M. Emmons, D. Boulware, C.A. Beam, D.M. Sullivan, et al., Cell adhesion to fibronectin (CAM-DR) influences acquired mitoxantrone resistance in U937 cells, Cancer Res. 66 (Febraury (4)) (2006) 2338–2345. [20] S.L. Lompardia, D.L. Papademetrio, M. Mascaro, E.M. Alvarez, S.E. Hajos, Human leukemic cell lines synthesize hyaluronan to avoid senescence and resist chemotherapy, Glycobiology 23 (December (12)) (2013) 1463–1476. [21] L.H. Mochmann, M. Neumann, E.K. von der Heide, V. Nowak, A.A. Kuhl, J. OrtizTanchez, et al., ERG induces a mesenchymal-like state associated with chemoresistance in leukemia cells, Oncotarget 5 (2) (2014) 351–362. [22] A.E. Freeman, R.M. Hoffman, In vivo-like growth of human tumors in vitro, Proc. Natl. Acad. Sci. U. S. A. 83 (8) (1986) 2694–2698. [23] R. Hoffman, Tissue culture, 2nd edn, in: E-i-CS Maloy (Ed.), Brenner’S Encyclopedia of Genetics, vol. 7, Academic Press, 2013, pp. 73–76. [24] R.M. Hoffman, Histocultures and their use, Encyclopedia of Life Sciences (ELS), John Wiley & Sons, Ltd., Chichester, 2010 October.
Acknowledgments We thank Constance D. Baldwin, PhD for reviewing the manuscript for clarity. We thank Dr. Linheng Li for his advice in studying N-cadherin expression in our model. We also thank Dr. Fariba Behbod for her help with ALDH-related studies and Dr. Eduardo Rosa-Molinar for his
112