**PP2A Deficiency Enhances Carcinogenesis of Lgr5**<sup>+</sup> **Intestinal Stem Cells Both in Organoids and In Vivo**

#### **Yu-Ting Yen 1,2, May Chien 1,2, Yung-Chih Lai 2, Dao-Peng Chen 3, Cheng-Ming Chuong 2,4, Mien-Chie Hung 5,6 and Shih-Chieh Hung 1,2,7,\***


Received: 16 November 2019; Accepted: 28 December 2019; Published: 30 December 2019

**Abstract:** In most cancers, cellular origin and the contribution of intrinsic and extrinsic factors toward transformation remain elusive. Cell specific carcinogenesis models are currently unavailable. To investigate cellular origin in carcinogenesis, we developed a tumorigenesis model based on a combination of carcinogenesis and genetically engineered mouse models. We show in organoids that treatment of any of three carcinogens, DMBA, MNU, or PhIP, with protein phosphatase 2A (PP2A) knockout induced tumorigenesis in Lgr5<sup>+</sup> intestinal lineage, but not in differentiated cells. These transformed cells increased in stem cell signature, were upregulated in EMT markers, and acquired tumorigenecity. A mechanistic approach demonstrated that tumorigenesis was dependent on Wnt, PI3K, and RAS-MAPK activation. In vivo combination with carcinogen and PP2A depletion also led to tumor formation. Using whole-exome sequencing, we demonstrate that these intestinal tumors display mutation landscape and core driver pathways resembling human intestinal tumor in The Cancer Genome Atlas (TCGA). These data provide a basis for understanding the interplay between extrinsic carcinogen and intrinsic genetic modification and suggest that PP2A functions as a tumor suppressor in intestine carcinogenesis.

**Keywords:** carcinogen; protein phosphatase 2A (PP2A); intestinal tumor; intestinal organoid; Lgr5<sup>+</sup> crypt stem cell

#### **1. Introduction**

The cells of origin in most cancers have remained unknown. Chemical carcinogenesis mouse models recapitulating most of human cancers that are induced by exposure to environmental carcinogens [1], however, is difficult to be achieved in a cell-specific manner. Therefore, the current strategies to investigate the cellular origins of cancers are using genetically engineered mouse models (GEMMs), with either transgenic or conditionally targeted gene technologies to induce tumor in different cellular contexts [2]. Moreover, both models take a long time to develop cancer, limiting progress in the cancer research field.

The most applied animal model for studying intestinal tumorigenesis is based on activating mutations in the Wnt pathway, which relies on adenomatous polyposis coli (Apc) depletion [3] and beta-catenin (CTNNB1) activation [4], leading to beta-catenin stabilization and constitutive transcription of its down-stream genes. Recent progress in the understanding of the cell of origin of intestinal tumor was made using this model, although several inconsistencies were observed. After in vivo Apc depletion in leucine-rich-repeat containing G-protein-coupled receptor 5 (Lgr5)<sup>+</sup> crypt stem cells, tumor formation occurred within 3–5 weeks [3]. However, Apc depletion or being combined with KrasG12D mutation in progenitor and differentiated cells did not induce tumor formation [5,6]. However, tumor-initiating mutations can occur in Lgr5<sup>+</sup> crypt stem cells and in differentiated Lgr5<sup>−</sup> cells [4], indicating that the two hypotheses are not mutually exclusive. Ablation of Lgr5<sup>+</sup> cells in orthotopically transplanted tumors, generated by genetic modification in differentiated villus cells, suppressed tumor growth [7]. Interestingly, Lgr5<sup>+</sup> cells reappeared and tumors recurred when ablation was terminated 4 days later. The generation of Lgr5<sup>+</sup> cells from Lgr5<sup>−</sup> cells after Lgr5<sup>+</sup> ablation was also observed in the xenograft mouse model of human colon cancer stem cells (CSCs) [8]. However, the specific mechanism of Lgr5<sup>+</sup> cell generation from remaining Lgr5<sup>−</sup> cells remains unclear.

Aberrant activation of signal transduction pathway, a dynamic process involving an 'on/off' switch, can transform a normal cell to be malignant or further render cancer cells with the capacities for therapy resistance. Activating mutations in genes encoding kinases or signaling molecules, such as RAS and PI3K, switch on the signaling, continuously activating a survival and/or proliferation pathway, while activations of phosphatases, such as the serine/threonine phosphatase PP2A family, switch off the signaling [9]. Previous efforts through high-throughput screens of tyrosine kinome and tyrosine phosphatome have identified several driver or passenger mutations in a spectrum of malignancies, including intestinal tumor [10–12]. However, there are few if any studies focusing on the altered signalings driven by serine/threonine kinase mutations [13]. Human intestinal tumors contain active mutations in genes encoding proteins involved in the WNT, MAPK, TGF-beta, and PI3K pathways [14]. Ingenuity pathway analysis (IPA) of TCGA-COAD revealed PP2A complex and its subunits, such as PPP2R1A and PPP2CA, are intercalated among several driver mutation pathways (Figure S1A–C). Moreover, endogenous PP2A inhibitors, SET and CIP2A [15], are highly expressed in intestinal tumors in comparison to their matched normal tissue samples (Figure S1D). We showed that PP2A was suppressed in intestinal tumor stem cells (CSCs), thereby activating its substrate kinases to enhance survival under hypoxia and serum depletion [16], thereby increasing resistance to anti-angiogenesis therapy [17]. Our recent studies also demonstrated that reduced PP2A activity in colorectal and lung CSCs enhances suspension survival and induces tumor initiation [18], revealing the tumor suppressive role of PP2A [19]. Although higher numbers of Apc, p53, KrasG12D, and Smad4 driver mutations may be required for human colorectal tumorigenesis, there are some intestinal tumors carry only one or no alteration in these driver mutations [20]. For example, gene fusions involving R-spondin 1 occurring in 10% of intestinal tumor are mutually exclusive with active Wnt signaling caused by *APC* or *CTNNB1* mutations [20]. The emerging novel intestinal tumorigenesis animal models should allow for elucidating the molecular mechanisms of these cancers.

Given that cancer is the product of complex interactions between the genetic and environmental predisposition factors, the combined use of chemical carcinogens that switch on kinases and GEMM with phosphatase deficiency is a logical approach for examining the complex interplay between genetic susceptibility and environmental exposure [21]. To investigate the cell origin of intestinal tumor, we first combined treatment with carcinogen 7,12-dimethylbenzanthracene (DMBA) that has previously been known to induce rodent s in the presence of 1,2-dimethyl-hydrazine [22] and PP2A inhibition via okadaic acid (OA) treatment or genetic deficiency. DMBA not only activates multiple mutations in different codons of ras [23] but also induces activation in other pathways, such as Notch [24], providing a screening approach for identifying key kinases or molecules. Besides DMBA, we also investigated the effects of *N*-methyl-*N*-nitrosourea (MNU) and 2-amino-1-methyl-6-phenylimidazo-[4,5-b]pyridine (PhIP) on intestinal carcinogenesis. MNU is one of the direct alkylating agents, which does not

require metabolic activation for initiating carcinogenesis [25]. PhIP has received considerable attention because it has multi-organ targets and it was developed upon broiling of fish and meat [26]. Moreover, we established primary intestinal organoid models that recapitulate the rodent intestinal tumorigenesis paradigm [27]. These rodent intestinal tumorigenesis models are useful in the development of new strategies for targeting rodent intestinal CSCs and treatment of intestinal tumor.

#### **2. Materials and Methods**

#### *2.1. Mouse Colonies*

*Ppp2r1aflox*/*flox* mice, carrying conditional alleles with loxP sites flanking exon 5–6 of *Ppp2r1a*, were purchased from the Jackson Laboratory and crossed to *Lgr5-EGFP-CreERT2* or *Villin-Cre* mice to generate *Lgr5-EGFP-CreERT2; Ppp2r1aflox*/*flox* or *Villin-Cre; Ppp2r1a flox*/*flox* mice. NOD/SCID mice were purchased from Lasco Co., Ltd. (Taiwan). All animal studies and care of live animals were approved and performed following the guidelines made by the China Medical University Institutional Animal Care and Use Committee 2016-398-1; 2017-239.

#### *2.2. Mouse Intestinal Organoid Cell Isolation, Culture, and Passage*

Organoid culture was preformed according to a protocol modified from previously described methods [28]. In brief, the intestines were dissected, opened longitudinally and cut into small (2 mm) pieces. The tissues were rocked in dissociation reagent and incubated at room temperature (15–25 ◦C) for 15 min. The tissues were then mixed and filtered through a 70 μm sterile cell strainer. The crypts were collected by centrifugation at 140× *g* for 5 min at 4 ◦C. Approximately 500 crypts were suspended in 50 μL growth factor reduced phenol-free Matrigel (BD Biosciences, San Jose, CA, USA). Next, a 50 μL droplet of Matrigel/crypt mix was placed and polymerized in the center well of a 48-well plate. The basic culture medium (Dulbecco's modified Eagle's medium/F12 supplemented with penicillin/streptomycin), was supplemented with 50 ng/mL murine recombinant epidermal growth factor (EGF; Peprotech, Hamburg, Germany), Noggin (5% final volume) and R-spondin 1 (5% final volume) called ''ENR" medium. Medium change was performed every 3–4 days. Each condition was examined in triplicate with multiple (>15) organoids in each sample. Each experiment was repeated twice.

#### *2.3. Dysplasia Index*

Histologic changes were scored blindly on the levels of four histological characteristics as previously described [27]: nuclear grade (enlarged nuclei with diffuse membrane irregularities and prominent nucleoli); stratification; mitoses and invasion (>2 foci). The dysplasia index was evaluated by all microscopic fields containing viable organoids with 5 fields per sample (*n*).

#### *2.4. Primary Organoid Transplantation*

For transplantation, cells from passage 7; day 50 organoid cultures were collected. Dissociated cells were pelleted by centrifugation and resuspended with Matrigel (50% Matrigel (BD), in a total volume of 100 μL containing indicated cell numbers and injected s.c. into NOD-SCID mice.

#### *2.5. Immunofluorescence and Immunohistochemistry*

Freshly isolated intestines were prepared according to a protocol modified from previously described methods [27]. The intestines were then applied for immunostaining. For immunostaining, the organoid cells were rinsed three times in ice-cold PBS. The organoid cells were spun down at 900 rpm for 10 min at 4 ◦C. Sections were deparaffinized and stained with H&E for the initial histology analysis. The immunofluorescence was performed on paraffin-embedded sections (5 μm). The permeabilized organoid cell samples were incubated with primary antibodies overnight at 4 ◦C. The samples were incubated with anti-PPP2R1A (GTX102206; GeneTex, Hsinchu City, Taiwan),

anti-CK20 (GTX110600 Genetex), anti-Lgr5 (GTX50839 Genetex), anti-SMA (Abcam, ab5694, Cambridge, MA, USA), anti-beta-catenin (BD Transduction Labs, San Jose, CA, USA; 610154); the secondary antibodies used were DyLight® 650 Conjugated goat anti-rabbit (cat no. A120-101D5; Bethyl Laboratories Inc., Montgomery, TX, USA) and Goat anti-Rabbit IgG Antibody-FITC (Bethyl cat no. A120-101F) and DAPI (Molecular Probes) for 1 h at room temperature. The slides were mounted with SlowFade (SlowFade® AntiFade Kit, Molecular Probes, Waltham, MA, USA) followed by covering with a coverslip, and the edges were sealed to prevent drying. The specimens were examined with a Zeiss 710 Laser Scanning confocal microscope (Zeiss, Oberkochen, Germany).

Intestinal tissue was fixed and processed into paraffin blocks according to standard procedures. beta-catenin immunohistochemistry was performed as previously described [3]. Immunohistochemistry protocol hold as following: freshly isolated intestines were flushed with 10% formalin in PBS and fixed by incubation in a 10-fold excess of formalin overnight at room temperature. The formalin was removed and the intestines washed twice in PBS at room temperature. The intestines were then transferred to a tissue cassette and dehydrated by serial immersion in 20-fold volumes of 70, 96 and 100% EtOH for 2 h each at 4 ◦C. Excess ethanol was removed by incubation in xylene for 1.5 h room temperature and the cassettes then immersed in liquid paraffin (56 ◦C) overnight. Paraffin blocks were prepared using standard methods and 4μm tissue sections generated. These sections were de-waxed by immersion in xylene (2 × 5 min) and hydrated by serial immersion in 100% EtOH (2 × 1 min), 96% EtOH (2 × 1 min), 70% EtOH (2 × 1 min) and distilled water (2 × 1 min). Endogenous peroxidase activity was blocked by immersing the slides in peroxidase blocking buffer (0.040 M citric acid, 0.121 M disodium hydrogen phosphate, 0.030 M sodium azide, 1.5% hydrogen peroxide) for 15 min at room temperature. For beta-catenin, antigen retrieval involved 20 min boiling in Tris-EDTA pH 9.0, and blocking buffer (1% BSA in PBS) added to the slides for 30 min at room temperature. For beta-catenin (BD Transduction Labs, 610154), staining involved 1/100 dilution in blocking buffer (0.05% BSA in PBS) for 2 h at room temperature.

The slides were then rinsed in PBS and secondary antibody added (polymer HRP-labeled anti-mouse/rabbit, Envision) for 30 min at room temperature (Dako, Trappes, France). Slides were again washed in PBS and bound peroxidase detected by adding DAB substrate for 10 min at room temperature. Slides were then washed 2× in PBS and nuclei counterstained with Mayer's hematoxylin for 2 min, followed by two rinses in distilled water. Sections were dehydrated by serial immersion for 1 min each in 50 and 60% EtOH, followed by 2 min each in 70, 96, and 100% EtOH and xylene. Slides were mounted in mounting medium and a coverslip placed over the tissue section.

For immunohistochemistry (IHC) analysis, nuclei expressing beta-catenin after IHC staining were counted under 200× magnification. The Histological score (H-score) was determined based on the intensity and percentage of nucleus staining at each intensity [29], and calculated as follows: H-score = (nucleus showing highly beta-catenin expression) × 3 + (nucleus showing beta-catenin expression) × 2 + (nucleus showing weak beta-catenin expression) × 1.

#### *2.6. Viral Infection of Organoids*

For in vitro deficiency of the *Ppp2r1a*, organoid cultures containing floxed *Ppp2r1a* alleles were infected with adenovirus-encoding Cre recombinase (Ad-Cre) (Vector Biolabs, Philadelphia, PA, USA) at a titer of 100 multiplicity of infection (MOI) [27].

#### *2.7. Tamoxifen Induction*

Mice aged 6–8 weeks were injected intraperitoneally with a single 200 μL dose of tamoxifen in sunflower oil at 10 mg/mL.

#### *2.8. Organoid Disaggregation, FACS, and Immunoblotting*

Organoid cultures were recovered and dissociated from collagen gel by collagenase IV incubation, followed by incubation with 0.05% trypsin and EDTA. After extensive washing with 10% FBS, cells

were filtered with 40-μm cell strainers (BD Falcon) Pellets were resuspended with FACS staining solution (5% FCS in PBS). Stringent wash was applied using ice-cold PBS, followed by isolation of Lgr5−EGFP<sup>+</sup> cells using an FACSAria II (BD) [30]. For immunoblotting, the organoid cells were lysed in lysis buffer (1% Triton X-100, 150 mmol/L NaCl, 10 mmol/L Tris pH 7.4, 1 mmol/L EDTA pH 8.0, protease inhibitor cocktail) and then sonicated. The protein concentration was then measured. Next, equal amounts of protein (20 μg/well) were separated by SDS-polyacrylamide gel electrophoresis, transferred to nitrocellulose, and immunoblotted with primary antibodies. The membranes were blocked with CISblock buffer purchased from Cis-biotechnology, Taiwan. The following antibodies were used: anti-phospho-AKT (Ser-473), anti-AKT; anti-phospho-ERK1/2 (Thr-202/Tyr-204), anti-ERK, and anti-PP2A from Cell Signaling; anti-Lgr5 and anti-alkaline phosphate intestinal (Alpi) (Genetex, epitope C-terminus), beta-catenin (BD Transduction Labs, 610154), anti-Villin (Santa Cruz, Dallas, TX, USA), and anti-beta-actin and GAPDH (Sigma-Aldrich, St. Louis, MO, USA). Following the primary antibody incubation, the nitrocellulose membranes were incubated with secondary antibodies and visualized by ECL.

#### *2.9. Antibody Arrays of Mouse AKT Pathway Phosphorylation*

The RayBio™ Mouse AKT Pathway Phosphorylation Array Kit (cat. no. AAH-AKT1-2) was purchased and preformed according to a protocol modified from RayBiotech Inc. (Norcross, GA, USA). Briefly, the array membranes were blocked with blocking buffer for 30 min at room temperature. The membranes were then incubated with 2 mL of lysate prepared from organoid cultures with different treatments after normalization with equal amounts of protein. After extensive washing with wash buffer I (3 washings of 5 min each), and wash buffer II (3 washings of 5 min each) to remove unbound materials, the membranes were incubated with the Detection Antibody Cocktail for 1.5 to 2 h at room temperature, followed by wash with wash buffer I and II. Then the membranes were incubated with HRP-Anti-Rabbit IgG for 2 h at room temperature. The unbound HRP antibody was washed out with wash buffer I and II. Finally, each array membrane was exposed to X-ray film using a chemiluminescence detection system (Perkin Elmer, Wellesley, MA, USA).

#### *2.10. Transcriptome Analysis*

RNA was extracted from organoid culture using an RNeasy Kit (Qiagen, Hilden, Germany). RNA integrity was assessed using the RNA Nano6000 assay kit (Agilent Technologies, Santa Clara, CA, USA). For RNA-seq, library preparation and sequencing were performed by Novogene Technology. The output data (FASTQ files) were mapped to the target genome (TopHat v2.0.12), which can generate a database of splice junctions based on the gene model annotation file. HTSeq v0.6.1 was used to count the reads numbers mapped to each gene. Then the FPKM of each gene was calculated based on the length of the gene and reads count mapped to this gene. Differential expression heatmap results and biological variability were analyzed by ClustVis free web server [31] and gene set enrichment analysis (GSEA) [32], respectively. Data were submitted and approved by Gene Expression Omnibus (GEO; accession number GSE120241).

#### *2.11. Whole-Exome Sequencing, Alignment, and Annotation*

Exome sequences were captured with SureSelectXT Mouse All Exon Kit (G7550E-001, Agilent, CA, USA) following the standard protocols. The products of exome capture should pass criteria: the length of fragments: 300 ± 30 bp and total amount: >600 ng. After exome capturing, the index-tagged samples were pooled and sequenced on Illumina HiSeq 2000. Burrows-Wheeler Alignment (v0.7.12) was employed to align reads to the reference genome (mm10) with default parameters. Aligned reads were sorted by picard-tools (v1.8). The duplicated reads were marked by picard-tools. Indel Realignment were performed with GenomeAnalysisTK (v3.5) using mm10 dbsnp database as known sites. Base quality score recalibration was also performed with GenomeAnalysisTK (v3.5) using mm10 dbsnp database. SNPs and indels were called by GenomeAnalysisTK HaplotypeCaller (v3.5), which

used default parameters. Whole exome sequencing raw data was submitted to SRA database (SRA; http://trace.ncbi.nlm.nih.gov/Traces/sra/, accession number SRP162613)

#### *2.12. Statistics*

The *p*-values were determined using two-tailed Student's *t*-test (t groups) and One-way ANOVA (>2 groups). A *p*-value less than 0.05 was considered significant.

#### **3. Results**

#### *3.1. Combination of DMBA and OA Treatment Induces Dysplasia and Oncogenic Transformation in Organoid Culture*

We chose an organoid culture system supported by epidermal growth factor (EGF), Noggin and R-spondin 1 (ENR) medium to investigate whether DMBA or/and OA could induce oncogenic transformation. As previously described [27], small intestine or colon organoids predominantly exhibited a well-organized, stereotyped epithelial single-layer organization at 7 days of culture, and maintained the similar morphology over a 50-day period of culture (Figure 1). At day 7, DMBA did not affect colony (organoid)-forming efficiency. OA induced a slight increase in both colony-forming efficiency, while a combination of DMBA and OA induced a large and significant increase in colony-forming efficiency (Figure 1A). At day 50, DMBA did not affect organoid morphology, OA induced mild enlargement in part of the epithelial layer with crowded nuclei, while a combination of DMBA and OA induced a very large malformation involving the entire epithelium with a confluent sheet of nuclear pleomorphism (Figure 1B), similar to that observed only when combining *Apc*, *p53*, *KrasG12D*, and *Smad4* mutations in differentiated villus cells [27]. Histological analysis revealed that organoids treated with DMBA alone had a single-layer epithelium, similar to the control. OA-treated organoids showed multi-cell-layer-changes in only a small part of the epithelium, while those treated with DMBA in combination with OA showed multi-cell-layer-changes with loss of the cell border in nearly the entire epithelium (Figure 1C), similar to the histology achieved only by quadruple mutants, *Apc*/*KrasG12D*/*p53*/*Smad4* [27]. A dysplasia index quantification of proliferation, nuclear atypia, invasion, and cellular stratification in organoids indicated that DMBA did not induce dysplasia compared to the control, OA induced a marginal increase in dysplasia, while the combination of DMBA and OA induced a large and significant increase in dysplasia (Figure 1D). Furthermore, the combination of DMBA and OA, but not DMBA or OA alone, endowed organoids with robust in vivo tumorigenicity, forming alpha-smooth muscle actin (SMA)<sup>+</sup> and CK20<sup>+</sup> intestinal tumor after subcutaneous transplantation (Figure 1E–G).

**Figure 1.** Combination of DMBA and okadaic acid (OA) induces dysplasia and oncogenic transformation in wild-type intestinal organoid culture. In vitro culture of wild type intestinal organoids without (control) or with DMBA or/and tamoxifen (TAM) in the presence of EGF, Noggin, and R-spondin 1 (500 single cells/well). (**A**) Colony (organoid)-forming efficiency was calculated at day 7. Experiment has been carried out in triplicate and each time 100 organoids were counted in each group. (**B**) Bright-field of organoid culture at day 50. Scale bar, 100 μm. (**C**) H&E staining and histologic characterization of cystic stratified epithelium with nuclear pleomorphism (arrow). Scale bar, 50 μm. (**D**) Dysplasia index at day 50 (experiment were repeated twice with *n* = 3 microscopic fields containing viable organoids). (**E**) Dissociated cells in Matrigel (500,000 cells/100 μL) were injected s.c. into NOD-SCID mice. In vivo tumor formation 45 days later (For those without tumor formation, observation was extended for up to 3 months, experiment were repeated twice with *n* = 3). (**F**) Tumor size 45 days after s.c. implantation (*n* = 8). (**G**) H&E staining and immunofluorescence of CK20 and SMA for tumor sections. Scale bar, 100 μm. \*, *p* < 0.05; \*\*\*, *p* < 0.0001 as determined with one-way ANOVA.

*3.2. Combination of DMBA Treatment and PP2A Deficiency Induces Dysplasia and Oncogenic Transformation in Organoid Culture*

We further characterized the transformation effect of DMBA or/and adenovirus carrying recombinase (Ad-Cre-GFP)-mediated-PP2A deficiency in *Ppp2r1aflox*/*flox* mice-derived organoids. Similarly, DMBA had no significant effects, Ad-Cre-GFP mediated-PP2A deficiency had mild effects, and the combination of DMBA and Ad-Cre-GFP mediated-PP2A deficiency had large effects on early organoid forming efficiency, late organoid morphology, histological changes, and dysplasia (Figure 2A–D). Furthermore, only the combination of DMBA and Ad-Cre-GFP-mediated-PP2A deficiency fully endowed organoids with robust in vivo tumorigenicity (Figure 2E–G).

**Figure 2.** Combination of DMBA and Ad-Cre induces dysplasia and oncogenic transformation in Ppp2r1aflox/flox intestinal organoid culture. In vitro culture of Ppp2r1aflox/flox intestinal organoids without (control) or with DMBA or/and Ad-Cre-GFP (Ad-Cre) infection in the presence of EGF, Noggin and R-spondin 1 (500 single cells/well). (**A**) Colony (organoid)-forming efficiency was calculated at day 7. At least 100 organoids were counted in each group. (**B**) Bright-field and fluorescence images of organoid culture at day 50. Scale bar, 100 μm. (**C**) H&E staining and histologic characterization of cystic stratified epithelium with nuclear pleomorphism (arrow indicated). Scale bar, 50 μm. (**D**) Dysplasia index at day 50 (experiments were repeated twice with *n* = 3 microscopic fields containing viable organoids). (**E**) Dissociated cells in Matrigel (500,000 cells/100 μL) were injected s.c. into NOD-SCID mice. In vivo tumor formation 45 days later (for those without tumor formation, observation was extended for up to 3 months, experiments were repeated twice with *n* = 3). (**F**) Tumor size 45 days after s.c. implantation (*n* = 3). (**G**) H&E staining and immunofluorescence of CK20 and SMA for tumor sections. Scale bar, 100 μm. \*, *p* < 0.05; \*\*\*, *p* < 0.0001 as determined with one-way ANOVA.

#### *3.3. Combination of DMBA Treatment and PP2A Deficiency in Lgr5*<sup>+</sup> *Rather than in Di*ff*erentiated Villus Cells Induces Dysplasia and Oncogenic Transformation in Organoid Culture*

To investigate whether Lgr5<sup>+</sup> crypt stem cells or differentiated villus cells serve as the cell of origin of tumors, we treated organoids from *Lgr5-EGFP-CreERT2*; *Ppp2r1aflox*/*flox* mice with DMBA or/and tamoxifen. Similarly, DMBA did not have significant effects, tamoxifen mediated-PP2A deficiency had mild effects, and the combination of DMBA and tamoxifen-mediated-PP2A deficiency had great effects on early organoid forming efficiency, late organoid morphology, histological changes, and dysplasia (Figure 3A–E). Furthermore, only the combination of DMBA and tamoxifen-mediated-PP2A deficiency endowed organoids with in vivo tumorigenicity (Figure 3F,G). Interestingly, the combination of DMBA and PP2A deficiency in organoids derived from *Villin-Cre; Ppp2r1aflox*/*flox* mice did not affect early organoid forming efficiency, late organoid morphology, histological changes, and dysplasia (Figure S2A–E), and failed to induce in vivo tumorigenicity (Figure S2F). Of note, the recombinase activity in Villin-Cre mice is gradually reduced from villus to crypt [33], nevertheless, Ppp2r1a protein is only deleted in sorted villus cells but not in sorted Lgr5<sup>+</sup> cells (Figure S2A). These data suggest that Lgr5<sup>+</sup> crypt stem cells but not differentiated villus cells serve as the cell of origin of intestinal tumor.

**Figure 3.** Combination of DMBA and TAM induces dysplasia and oncogenic transformation in *Lgr5-EGFP-CreERT2*; *Ppp2r1aflox*/*flox* intestinal organoid culture. In vitro culture of *Lgr5-EGFP-CreERT2; Ppp2r1aflox*/*flox* intestinal organoids without (control) or with DMBA or/and tamoxifen (TAM) in the presence of EGF, Noggin and R-spondin 1 (500 single cells/well). (**A**) Colony (organoid)-forming efficiency was calculated at day 7. At least 100 organoids were counted in each group. (**B**) Bright-field and fluorescence images of organoid culture at day 50. Scale bar, 100 μm. (**C**) Fluorescence-activated cell sorting (FACS) isolation of Lgr5<sup>+</sup> and Lgr5<sup>−</sup> populations. After FACS, PPP2R1A protein levels were detected by western blot. (**D**) H&E staining and histologic characterization of cystic stratified epithelium with nuclear pleomorphism (arrow). Scale bar, 50 μm. (**E**) Dysplasia index at day 50 (experiments were repeated twice with *n* = 3 microscopic fields containing viable organoids). (**F**) Dissociated cells in Matrigel (500,000 cells/100 μL) were injected s.c. into NOD-SCID mice. In vivo tumor formed 45 days later (for those without tumor formation, observation was extended for up to 3 months, experiments were repeated twice with *n* = 3). (**G**) H&E staining and immunofluorescence of CK20 and SMA for tumor sections. Scale bar, 100 μm. \*, *p* < 0.05; \*\*\*, *p* < 0.0001 as determined with one-way ANOVA.

#### *3.4. Combination of DMBA Treatment and PP2A Deficiency in Lgr5*<sup>+</sup> *Cells Induces Upregulation in Stem Cell and EMT Markers, Downregulation in Di*ff*erentiated Markers, and Tumorigenicity in Organoid Culture*

Flow cytometric analysis (Figure 4A) revealed that DMBA did not increase the Lgr5<sup>+</sup> cell ratio (as assayed by Lgr5-EGFP). PP2A deficiency induced a marginal increase, while the combination of DMBA and PP2A deficiency induced a significant increase in the Lgr5<sup>+</sup> cell ratio. Comparative gene expression analysis of RNA samples isolated from organoid culture of *Lgr5-EGFP-CreERT2; Ppp2r1aflox*/*flox* mice 50 days after DMBA and tamoxifen administration revealed marked upregulation of stem cell genes, such as *Lgr5*, *CD44*, *Ephb3*, *Egr2*, *Notch1*, and *Sox4*; as well as EMT markers, such as *Snail1*, *Snail2*, *Twist1*, *fibronectin*, and *vimentin*; and a marked downregulation of genes associated with differentiated cells, such as Paneth, enterocyte, goblet, and secretory cells compared to other treatment groups (Figure 4B). Furthermore, a small number (103) of Lgr5<sup>+</sup> but not Lgr5<sup>−</sup> cells isolated from organoid culture treated with DMBA and tamoxifen possessed in vivo tumorigenicity (Figure 4C). Collectively, these data suggest that the combination of DMBA and PP2A deficiency converted Lgr5<sup>+</sup> crypt stem cells into CSCs.

**Figure 4.** Combination of DMBA and PP2A deficient induces upregulation in stem cell and EMT markers and downregulation in differentiated markers in intestinal organoid. (**A**) In vitro culture of *Lgr5-EGFP-CreERT2; Ppp2r1aflox*/*flox* intestinal organoids. Merged bright-field and fluorescence images of organoid culture treated without (control) or with DMBA or/and tamoxifen (TAM) for 50 days in the presence of EGF, Noggin and R-spondin 1. *Lgr5*-*EGFP* were denoted as arrows. Scale bar, 100 μm. Flow cytometric analysis and quantification of GFP expression (bottom panel). No fluorescence organoid culture was serve as negative control to decide the threshold. \*, *p* < 0.05 as determined with one-way ANOVA. (**B**) RNA-*sequencing* (seq) analysis of transcriptomes for 50-day organoids. The upper heat map shows clustering to previously reported RNA-seq data of sorted ISC (Intestinal Stem Cell); reserve ISC; paneth cell; LRCs (label retaining cells); tuft; enterocytes; goblet and secretory progenitor cells. The lower heat map shows clustering to EMT (epithelial–mesenchymal transition). (**C**) Tumor incidence in limiting dilution assay. Tumorigenic potential characterization of indicated numbers of Lgr5<sup>+</sup> and Lgr5− cells from the 50-day organoid culture. Dissociated cells in Matrigel (100 μL) were injected s.c. into NOD-SCID mice. Incidence of tumor formation was calculated more than 2 months.

#### *3.5. Combination of DMBA Treatment and PP2A Deficiency Generates CSCs through PI3K, ERK, and Wnt Activation*

To understand the molecular characteristics of the genes and pathways involved in organoid culture of *Lgr5-EGFP-CreERT2*; *Ppp2r1aflox*/*flox* mice after DMBA and tamoxifen administration, RNA-seq and gene set enrichment analysis (GSEA) were performed. Similar to previous findings that intestinal tumor begins with specific molecular alterations in Wnt-beta-catenin pathway [13,23], Wnt signaling was upregulated upon oncogenic transformation of organoid culture (Figure 5A). Western blotting of nuclear proteins (Figure 5B) and immunofluorescence (Figure 5C) revealed that nuclear accumulation of beta-catenin was predominantly observed in organoids with both of DMBA treatment and PP2A deficiency. Screening with a serine/threonine phosphorylation protein array (Figure S3) followed by confirmation with western blotting further revealed that PI3K/AKT/GSK-3beta and Raf/ERK were

activated in organoids with both DMBA treatment and PP2A depletion compared to other treatment groups (Figure 5D,E). Interestingly, treatment with the PI3K inhibitor LY294002, MEK inhibitor PD98059, and Wnt inhibitor DKK1 reduced the formation of malformed organoids, inhibited dysplasia (Figure 5F,G), and completely blocked in vivo tumorigenicity (Figure 5H). Notably, not only the WNT signaling was the most, but also the PI3K and RAS-MAPK (ERK) signalings were common altered pathways in human intestinal tumor, as revealed by The Cancer Genome Atlas (TCGA) Project [14]. These data suggest that CSC generation by DMBA treatment and PP2A deficiency depends on the activation of PI3K, ERK, and Wnt signals.

**Figure 5.** *Cont.*

**Figure 5.** Beta-catenin activation caused by PI3K and ERK mediates dysplasia and oncogenic transformation in organoid culture. *Lgr5-EGFP-CreERT2; Ppp2r1aflox*/*flox* intestinal organoids were treated without (control) or with DMBA or/and tamoxifen (TAM) for 50 days in the presence of EGF, Noggin and R-spondin 1. (**A**) Gene set enrichment analysis (GSEA) shows "WNT SIGNALING PATHWAY" for the organoids treated with DMBA and TAM group versus control group, *p* < 0.05. (**B**) Western blotting of nuclear and cytoplasmic fractions. GAPDH and Histone H3 were used as protein loading controls for the cytoplasmic and nuclear fractions, respectively. (**C**) Immunofluorescence of Lgr5-EGFP and beta-catenin expression in organoid (denoted by arrows). Scale bar, 100 μm. (**D**) Graphs of mouse AKT pathway phosphorylation protein expression array (original data in Extended Data Figure 3) and densitometric analyses, *n* = 1. (**E**) Western blotting. (**F**) Bright-field, fluorescence images, H&E, PAS staining and beta-catenin immunofluorescence images. (**G**) Western blot analysis of whole cell lysate and nuclear fraction from organoid cultures treated with EGF, Noggin, R-spondin 1 (ENR), DMBA, and TAM for 50 days in the absence (Vehicle) or presence of indicated inhibitor treatment. (**H**) Dissociated cells in Matrigel (500,000 cells/100 μL) were injected s.c. into NOD-SCID mice. In vivo tumor formed 45 days later (For those without tumor formation, observation was extended for up to 3 months, *n* = 8). \*\*\*, *p* < 0.0001 as determined with one-way ANOVA.

#### *3.6. Lgr5*<sup>+</sup> *CSCs Are R-Spondin 1-Dependent*

CSCs isolated from human intestinal tumor specimens express the Lgr5 crypt marker [34]. R-spondin is expressed by the intestinal stroma and is differentially upregulated during *Citrobacter rodentium*- and dextran sulfate sodium (DSS)-induced colitis in mice, which reflects human ulcerative colitis, a precancerous stage [35]. These data suggest a role for R-spondin 1 and its receptor Lgr5 in the maintenance of undifferentiated status and tumorigenesis of human colorectal CSCs. In contrast to tumorigenesis initiated by dedifferentiation (tumor generated from intestinal epithelial cell of *villin-creERT2*/*APC lox*/*lox*/*K-rasG12D*/+ mice) [4], where tumor cells were Lgr5<sup>−</sup> and generated independently of R-spondin 1, tumor cells in the current study were Lgr5<sup>+</sup> and generated dependently on R-spondin 1 (Figure 6), suggesting that tumor cells generated by DMBA treatment and PP2A deficiency in mouse Lgr5<sup>+</sup> cells were indeed intestinal CSCs and could serve as surrogates of human

colorectal CSCs [34]. Previous reports using tumorigenesis models generated by genetic manipulation in differentiated villus cells, even with low efficiency in tumorigenesis [27], also demonstrated important roles for Lgr5<sup>+</sup> cells in tumorigenesis/metastasis [7,8].

**Figure 6.** R-spondin 1 drives Wnt-dependent dysplasia and oncogenic transformation in the intestinal organoid culture. In vitro culture of *Lgr5-EGFP-CreERT2; Ppp2r1aflox*/*flox* intestinal organoids (500 single cells/well) treated with DMBA and tamoxifen (TAM) in the presence of EGF, Noggin, and with (ENR) or without R-spondin-1 (EN). (**A**) Schematic illustration of the experimental design. (**B**) Colony (organoid)-forming efficiency at day 7. At least 100 organoids were counted in each group. (**C**) Bright-field images at day 50. Scale bar, 100 μm. (**D**) H&E staining and histologic characterization of cystic stratified epithelium with nuclear pleomorphism (arrow). Scale bar, 50 μm. (**E**) Dysplasia index at day 50 (experiment was repeated twice with *n* = 3 microscopic fields containing viable organoids). (**F**) Incidence of tumor formation at indicated time periods, and (**G**) tumor size in NOD-SCID mice after s.c. injection 45 days later of indicated cell numbers from 50-day organoid. \*\*, *p* < 0.01 as determined with Student's *t*-test.

#### *3.7. Combination of DMBA Treatment with PP2A Deficiency in Lgr5*<sup>+</sup> *Also Induces Tumor Formation In Vivo*

We investigated whether tumorigenesis generated by DMBA treatment and PP2A deficiency in mouse Lgr5<sup>+</sup> cells reflects tumor formation in vivo. *Lgr5-EGFP-CreERT2; Ppp2r1aflox*/*flox* mice treated with DMBA and tamoxifen formed adenocarcinoma in the small intestine and colorectal regions 36 days later. Histological analysis revealed that mice receiving DMBA and tamoxifen increased the incidence of multiple foci adenoma compared to other groups (Figure 7A). IHC analysis revealed that beta-catenin accumulated in the nuclei of tumor cells mainly in the crypt area (Figure 7B) and quantitative evaluation of nucleus beta-catenin accumulation [29] further showed that nucleus beta-catenin H-score was significantly greater in mice receiving DMBA and tamoxifen compared to other groups (Figure 7C,D). The efficiency and rapidity were much greater than tumorigenesis generated by *Apc* deficiency in either Lgr5<sup>+</sup> crypt stem cell- [3] or differentiated villus cell-based models [4].

**Figure 7.** Combination of DMBA treatment with PP2A deficient in Lgr5<sup>+</sup> drives intestinal neoplasia in both the small intestine and colon. (**A**) H&E staining and (**B**) Beta-catenin IHC were performed on the serial sections. Multiple beta-cateninhigh adenomas were observed throughout the colon 36 days after induction. (**C**) Quantitative analysis of nucleus beta-catenin H-score from (**B**). (**D**) High level expression of beta-catenin was apparent in the transformed stem cell compartment (Peyer's patches are stained blue). Multiple beta-cateninhigh transformed cells were observed throughout the intestinal 36 day after induction (magnified at right panel, denoted by arrows). Original magnifications: (**A**) left, 4×; (**A**) right, 10×; (**B**) left, 20×; (**B**) right, 40×; (**D**) center, 20×; (**D**) right, 40×; Scale bar, 100 μm. \*\*, *p* < 0.01 as determined with one-way ANOVA.

#### *3.8. Not Only DMBA but Also Other Carcinogens Induce Tumors from Lgr5*<sup>+</sup> *Intestinal Stem Cells of PP2A Deficient Mice*

To demonstrate carcinogen-induced tumor from Lgr5<sup>+</sup> intestinal stem cells of PP2A deficient mice was not limited to DMBA, *N*-methyl-*N*-nitrosourea (MNU) and 2-amino-1-methyl-6-phenylimidazo-[4,5-b]pyridine (PhIP) were also administered in combination with tamoxifen treatment in organoid cultures derived from *Lgr5-EGFP-CreERT2; Ppp2r1aflox*/*flox* mice and from *Villin-Cre; Ppp2r1aflox*/*flox* mice. Interestingly, increased early organoid forming efficiency, late organoid with irregular nuclei and prominent nucleoli morphology and dysplasia, and in vivo tumorigenicity were only observed in organoid cultures derived from *Lgr5-EGFP-CreERT2; Ppp2r1aflox*/*flox* mice but not from *Villin-Cre; Ppp2r1aflox*/*flox* mice (Figure S4A,B). More importantly, we found that tumorigenicity induced by combination of MUN or PhIP with PP2A deficiency also depended on the activation of PI3K, ERK, and Wnt signals (Figure S4C–F). Similarly, combination of MUN or PhIP treatment with PP2A deficiency in Lgr5<sup>+</sup> cells also induced tumor formation in vivo (Figure S4G).

#### *3.9. Mutational Landscapes of Intestinal Tumors Derived from Lgr5*<sup>+</sup> *Cells Treated with Carcinogen and PP2A Deficiency*

To gain insight of the genetic alterations that drove these pathways in the models, we performed whole-exome sequencing (WES) (Figure S5A). Somatic variants in each chemical induced tumor organoid sample were identified with a tumor-control paired strategy by removing the variants in their paired control samples and the variants affecting protein coding sequence were further filtrated (Figure S5B). The numbers of somatic mutations, including synonymous and nonsynonymous mutations, for organoid cultures with combined DMBA, MUN, or PhIP treatment with PP2A deficiency in Lgr5<sup>+</sup> cells were shown in Figure S5C. IPA analysis of 270 mutated genes shared by DMBA, MUN, and PhIP treatment with PP2A deficiency in Lgr5<sup>+</sup> cells (Figure S5D) revealed 10 top significantly enriched pathways (Figure S5E), including the intestinal tumor metastasis signaling (*p*-value = 2.12 <sup>×</sup> 10−57) and Wnt/beta-catenin signaling (*p*-value = 1.40 <sup>×</sup> 10−55). Furthermore, beta-catenin (*CTNNB1*) seemed to be a common downstream hub as identified by the Path Explorer function in IPA (Figure S5F). There were five core pathways related with human intestinal tumor found by TCGA, including p53, RAS-MAPK, PI3K, TGF-beta, and WNT pathways (Figure 8) [14]. We identified several recurrent mutations with FDR < 0.1, including *Braf*, *erbb2*, *kras*, *pten*, *Smad2*, *Smad4*, *Apc*, *DKK2*, *Wnt4*, *Wnt5a*, and *Wnt5b* (Figure 8). Based on these analyses, the mouse intestinal tumor organoid models based on combined use of chemical carcinogen and genetic modification can recapitulate the human colorectal cancer development process in response to complex interactions between the genotype and environmental factor.

**Figure 8.** Landscape of somatic mutant genes in three carcinogen treated organoid cultures. Data matrix shows number of somatic mutant genes in each carcinogen treated organoid cultures and were classified according to their pathway. Somatic mutations are presented according to the type of mutation (missense variant, intron variant, 3 or 5 prime UTR variant, frameshift variant, inframe deletion, or splice region variant) On the left, the total number of mutations of each gene within all three groups is shown with a bar plot, while the q-value of each significantly mutated gene is shown on the right.

#### **4. Discussion**

To improve the efficiency of tumorigenesis in GEMMs, multiple gene mutations are necessary. Given that the frequency of point mutations varies from less than 0.1 to greater than 50 mutations per megabase [36], GEMMs are a greatly oversimplified view of the numbers and types of mutations found in human cancers [37]. The use of GEMMs in combination with carcinogenesis increased the tumor spectrum or speeded the tumor formation observed in some GEMMs, such as the *p53*−/<sup>−</sup> mouse model in combination with exposure to carcinogens or radiation [21,38]. In the current study, P*pp2r1a*−/<sup>−</sup> mouse model in combination with exposure to different carcinogens induced tumorigenesis through similar signaling pathways, while GEMM or carcinogen exposure alone did not induce tumorigenesis. These data suggest that the combination of GEMMs with exposure to carcinogen or other environmental agents is a logical approach in studying tumorigenesis and cancer progression.

PP2A is a tumor suppressor that regulates many oncogenic pathways. In fact, decreased PP2A activity has been reported as a common event in colorectal cancer [39]. DMBA and MNU are important environmental carcinogens. PhIP, a food-borne carcinogen produced while cooking meat and fish, models human colon cancer in rodents [40]. Here, we demonstrated that DMBA, MNU, or PhIP each induced intestinal organoid transformation when combined with PP2A deficiency in Lgr5<sup>+</sup> intestinal stem cells but not in differentiated villus cells, suggesting that PP2A plays a role in suppressing colorectal tumorigenesis induced by chemical carcinogen exposure. Our results provide further experimental evidence to demonstrate that the cell of origin of intestinal tumor is crypt stem cells instead of differentiated villus cells. We did not focus on the molecular mechanism underlying the differential tumorigenicity pathways between stem cells and differentiated cells, which requires further investigation.

A recent study shows that unless the R-spondin and Wnt ligands are both present, the default fate of Lgr5<sup>+</sup> crypt stem cells is differentiation [41]. However, gain-of-function studies reveal that R-spondin and Wnt ligands have qualitatively distinct, non-interchangeable roles in crypt stem cells. Wnt proteins confer a basal competency unto the Lgr5<sup>+</sup> crypt stem cell by inducing R-spondin receptor expression that enables R-spondin-driven Lgr5<sup>+</sup> crypt stem cell self-renewal. In the current study, using a combination of carcinogen treatment and PP2A deficiency, intestinal organoids derived from Lgr5-cre; Ppp2r1aflox/flox, but not Villin-Cre; Ppp2r1aflox/flox, underwent oncogenic transformation and exhibited CSC phenotypes that were dependent on the presence of R-spondin 1 in the culture media. These studies together suggest the important roles of R-spondin signaling in stem cell self-renewal and preventing differentiation. The discrepancy of R-spondin 1 dependence between tumorigenicity induced by the current method and previous GEMM relied on *Apc* depletion or *CTNNB1* activation [4] is supported by the mutual exclusion of *R-spondin* fusion and *Apc* or *CTNNB1* mutation identified in human intestinal tumors [20]. The Lgr5/R-spondin 1 complex degrades Rnf43 and Znrf3, two transmembrane E3 ligases that remove Wnt receptors from the stem cell surface. Consistently, simultaneous deficient of these two E3 ligases in the intestinal epithelium induced the formation of unusual adenomas consisting entirely of Lgr5<sup>+</sup> stem cells and their niche [42].

Simultaneous carcinogen treatment and conditional deletion of PP2A in villus cells did not induce transformation or increased proliferation or dysplasia of intestine organoids. The current study did not aim to study the underlying mechanism that differentiates Lgr5<sup>+</sup> crypt stem cells from differentiated villus cells in terms of vulnerability to carcinogenesis. Based on our current results, both Wnt and Rspondin/Lgr5 signaling pathways play important roles in nuclear-beta catenin localization, which is important for stem cell self-renewal and may also initiate tumorigenesis once dysregulated.

#### **5. Conclusions**

In summary, we demonstrated that carcinogen-induced cancer arises from Lgr5<sup>+</sup> crypt stem cells in *Ppp2r1a*−/<sup>−</sup> mice. In addition, combing carcinogenesis with GEMM recapitulated the developmental process of environmentally induced human tumor, while increasing the rate and percentage of tumorigenesis. Interestingly, exposure to different carcinogens, such as DMBA, MNU, or PhIP, when combined with PP2A deficiency in Lgr5<sup>+</sup> intestinal stem cells induced tumorigenesis that was dependent on the activation of pathways including Wnt, PI3K, and RAS-MAPK signalings, the common altered pathways revealed by TCGA human intestinal tumor project [14]. Together, these data suggest that PP2A functions as a tumor suppressor in intestine carcinogenesis. This organoid platform provides experimental evidence as to its usefulness in detection of key oncogenes and suppressor genes as early molecular epidemiological biomarkers of carcinogenesis, and is useful in human cancer prevention practice as well.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4409/9/1/90/s1, Figure S1: Screening strategy for phosphatase hub gene in TCGA-COAD data set, Figure S2: DMBA and PP2A deficiency did not induce dysplasia and oncogenic transformation in *Villin-Cre; Ppp2r1aflox*/*flox* intestinal organoid culture, Figure S3: Serine/threonine phosphorylation protein array screening in individual organoid groups, Figure S4: Combination of MNU or PhIP and TAM induces dysplasia and oncogenic transformation in *Lgr5-EGFP-CreERT2; Ppp2r1aflox*/*flox* intestinal organoid culture, Figure S5: Ingenuity pathway analysis (IPA) to identify significant canonical pathways and interactome in three carcinogens induced oncogenic transformation in *Lgr5-EGFP-CreERT2; Ppp2r1aflox*/*flox* intestinal organoids.

**Author Contributions:** Y.-T.Y. designed and performed experiments, analyzed the data and wrote the paper. M.C. contributed to Lgr5-cre and Vilin-cre PP2A knockout mice analysis and helped tissue processing. Y.-C.L. performed the RNA-seq analysis. D.-P.C. performed the WES analysis. C.-M.C. reviewed and edited the paper. M.-C.H. reviewed and edited the paper. S.-C.H. wrote the proposal, designed and supervised the study, analyzed the data and wrote the paper. All authors have read and agreed to the published version of the manuscript.

**Funding:** Grants supported by Minister of Science and Technology (MOST 106-2321-B-039-003) and Integrative Stem Cell Center, China Medical University Hospital. This work was also financially supported by the "Drug Development Center, China Medical University" from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in Taiwan. The funding sources had no involvement in study design, in the collection, analysis and interpretation of data, in the writing of the report, and in the decision to submit the article for publication.

**Acknowledgments:** We thank Bin Tean Teh (Cancer Science Institute of Singapore, National University of Singapore, Centre for Life Sciences, Singapore) for discussion of our data.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Mouse Embryonic Stem Cell-Derived Ureteric Bud Progenitors Induce Nephrogenesis**

**Zenglai Tan 1,\*, Aleksandra Rak-Raszewska 1, Ilya Skovorodkin <sup>1</sup> and Seppo J. Vainio 1,2,\***


Received: 18 December 2019; Accepted: 27 January 2020; Published: 31 January 2020

**Abstract:** Generation of kidney organoids from pluripotent stem cells (PSCs) is regarded as a potentially powerful way to study kidney development, disease, and regeneration. Direct differentiation of PSCs towards renal lineages is well studied; however, most of the studies relate to generation of nephron progenitor population from PSCs. Until now, differentiation of PSCs into ureteric bud (UB) progenitor cells has had limited success. Here, we describe a simple, efficient, and reproducible protocol to direct differentiation of mouse embryonic stem cells (mESCs) into UB progenitor cells. The mESC-derived UB cells were able to induce nephrogenesis when co-cultured with primary metanephric mesenchyme (pMM). In generated kidney organoids, the embryonic pMM developed nephron structures, and the mESC-derived UB cells formed numerous collecting ducts connected with the nephron tubules. Altogether, our study established an uncomplicated and reproducible platform to generate ureteric bud progenitors from mouse embryonic stem cells.

**Keywords:** mouse embryonic stem cell; differentiation protocol; ureteric bud progenitor cells; 3D kidney organoids

#### **1. Introduction**

Pluripotent stem cells (PSCs) possess great potential of differentiating into multiple cell types that are widely used for studies in developmental biology and regenerative medicine [1]. Kidney organoids derived from PSCs have been shown to be able to mimic the in vivo kidney structure development and function in vitro [2–4]. Renal organoids in a four-dimensional (4D) (3D plus time) culture system self-organize into highly complex tissue-specific morphology that is sufficient to model tissue development, disease, and injury [5–8]. A combination of genome editing and stem cell technologies allows for generation of personalized kidney organoids, which provide powerful tools for kidney disease treatment, drug toxicity screening, and tissue regeneration [9,10].

Recently, multiple protocols enabling generation of renal lineages from mouse and human PSCs have been published. We and several other groups reported induction of nephron progenitors, which have the potential to develop into epithelial nephron-like structures [2–4,8,11–18]. Other groups have shown the derivation of ureteric bud (UB) progenitors [19]. However, they did not show nephron progenitor induction and also lacked the connection between collecting ducts and nephrons [2,19]. A newly published study has shown generation of UB structures from PSCs, which possessed UB-like branching morphogenesis when aggregated with the primary metanephric mesenchyme (pMM) to form chimeric kidney organoids [8]. However, the protocol is technically complex, which limits its application. Analysis of these reports suggests that we still need further studies to develop simple, reproducible, and stable protocols for UB progenitor generation.

Here, we report a simple protocol to direct differentiation of mouse embryonic stem cells (mESCs) into UB progenitor cells. The newly generated UB progenitor cells have a potential to develop into ureteric bud structures and have a capacity to induce nephrogenesis when co-cultured with dissociated pMM. In reconstructed kidney organoids, the pMM developed into nephron structures and the UB progenitor cells formed collecting ducts, which also connected with the nephron tubules. The chimeric kidney organoids also display the presence of endothelial cells forming a vascular network. In conclusion, our study established an uncomplicated and reproducible method for generation of UB progenitors from PSCs that can be used for tubulogenesis induction.

#### **2. Materials and Methods**

Animal care and experimental procedures in this study were in accordance with Finnish national legislation on the use of laboratory animals, the European Convention for the protection of vertebrate animal used for experimental and other scientific purposes (ETS 123), and the EU Directive 86/609/EEC. Animal experimentation was also authorized by the Finnish National Animal Experiment Board (ELLA) as being compliant with the EU guidelines for animal research and welfare.

#### *2.1. Mouse ES Cell Line Generation and Maintenance*

mESC line Sv129S6 was obtained from the Biocenter Oulu Transgenic core facility. The Sv129S6 mESC line was derived from Taconic's W4/129S6 inbred mouse strain and has been tested to have a normal karyotype [19]. The mESC and mESC-GFP (Green Fluorescent Protein, pcDNA3.1 transfected) line was described previously [20]. It was maintained on mitotically inactivated mouse embryonic fibroblasts (MEFs) in mESC medium: Dulbecco's Modified Eagle's Medium (DMEM; Life Technologies, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, Waltham, MA, USA ), 1% (*v*/*v*) nonessential amino acids (Life Technologies), 100 mM 2-mercaptoethanol (Nacalai Tesque, Kyoto, Japan), and 1000 U/mL leukemia inhibitory factor (LIF, Millipore, Espoo, Finland).

#### *2.2. Directed Di*ff*erentiation of mESCs to UB Progenitors*

Mouse ESCs were cultured as described above in MEF-coated 6-well plates in mESCs medium up to 70%–90% confluency. The cells were passaged on 1% geltrex-coated 24-well plates at 30,000 cells/cm<sup>2</sup> in 500 μL 50% ReproFF2 (ReproCELL, Glasgow, UK)—50% CM (conditioned MEF medium) supplied with 10 ng/mL fibroblast growth factor (FGF) 2 (PeproTech Nordic, Stockholm, Sweden) and 10 ng/mL Activin A (Cell Guidance Systems, Cambridge, UK). After 2 days, the cells reached 60–80% of confluency, and the medium was changed to differentiation medium: Advanced RPMI (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 8 μM CHIR99021 (Reagents Direct, Encinitas, CA, USA) and 4–8 ng/mL Noggin (R&D systems, Minneapolis, MN, USA) for 2 days, followed by treatment of the cells with Activin A (10 ng/mL) for 2 days and then 3 days with FGF9 (40 ng/mL) (PeproTech Nordic, Stockholm, Sweeden) in Advanced RPMI medium (ThermoFisher Scientific).

#### *2.3. Gene Expression Analysis*

A RNeasy kit (Qiagen, Germantown, MD, USA) was used according to the manufacturer's recommendations to extract total RNA. cDNA synthesis (First Strand cDNA Synthesis Kit, ThermoFisher Scientific, Waltham, MA, USA) was performed using standard protocols. qPCR was run using a CFX96 BioRad thermocycler. Brilliant III SYBR® Green QPCR Master Mix (Agilent Technologies, Santa Clara, CA, USA) was used according to the manufacturer's instructions. GAPDH probe served as a control to normalize the data. The data was analyzed using the 2−ΔΔCT method. Gene expression measurements were performed in triplicates on three independent experiments. The primer sequences are given in Supplementary Table S1.

#### *2.4. Chimera 3D Kidney Organoids Formation*

At day 9 of mESCs differentiation, which represents the UB progenitor cells stage, cells were dissociated into single cell suspension using TrypLE select (Life Technologies). After three washes in phosphate buffer saline (PBS), cells were reconstituted in organ culture medium (DMEM supplemented with 10% fetal bovine serum, 1% penicillin–streptomycin). For generation of kidney organoids, we mixed dissected and dissociated pMM cells (from CD-1 pregnant females, at E11.5), as described before [21], with differentiated UB progenitors at a ratio of 3:1, respectively. Briefly, kidney rudiments were dissected out of E11.5 embryos and treated with trypsin/pancreatin solution for 30–40 s to separate the UB from the pMM [22]. The MM cells were dissociated using 2.0 mg/mL Collagenase IV solution (Wortington, Lakewood, NJ, USA) in 0.1% bovine serum albumin (BSA) in 1×PBS. 10 min incubation was interrupted by pipetting and continued until cells were separated. The reaction was stopped by washing three times in complete organ culture medium [22]. The pMM- and mESC-derived UB cells were mixed in a 3:1 ratio and centrifuged for 4 min at 1380<sup>×</sup> *g* to form a pellet (5 <sup>×</sup> 104 cells) in Lo-Binding Eppendorf tubes. Following centrifugation, we carefully transferred the differentiated UB and pMM pellets to filter into Trowel culture to aggregate as an organoid. The organ culture medium was changed every 3–4 days.

For generation of whole kidney organoids, we dissected mouse kidney rudiments at E11.5 from CD-1 pregnant females. Kidney rudiments were dissociated into single cell suspension as described previously [19]. After dissociation, the embryonic kidney cells (7 <sup>×</sup> 104) were mixed with undifferentiated mESC or differentiated mESCs-derived UB progenitors (1 <sup>×</sup> 104) to make the pellet. We then continued the procedure as described above.

#### *2.5. Whole-Mount Immunohistochemistry*

Kidney organoids were washed 2 times with PBS and fixed with 100% cold methanol (–20 ◦C) for 30 min at room temperature (RT) or with 4% paraformaldehyde in PBS (organoid with GFP or dye) for 30 min at RT in the dark. After fixation, the organoids were washed at least three times in PBS and blocked in 0.1% Triton-X100 (Sigma, Lyon, France), 1% BSA, and 10% goat serum/0.02M glycine-PBS for 1–3 h at room temperature. Incubation of the organoids with primary antibodies was performed in a blocking buffer overnight at 4 ◦C. The samples were washed 6 times with PBS and incubated with secondary antibodies Alexa Fluor 405, 488, 568, 546, or 647 (1:1000, Life technologies) and fluorescein anti-LTL (Lotus Tetragonolobus Lectin, 1:350, #FL-1321, Vector Laboratories, Burlingame, CA, USA) overnight at 4 ◦C and counter-stained with Hoechst (Thermo Fisher Scientific). The primary antibodies used in stainings were: Wt1 (1:100, #05-753, Millipore), Pax2 (1:200, #PRB-276P, Covance, Cambridge, MA, USA), Troma1 (1:200, DSHB, Iowa City, IA, USA), Gata3 (1:20, #AF2605-SP, R&D Systems), E-cad (1:300, #610181, BD Biosciences, Franklin Lakes, NJ, USA), Synaptopodin (SYNPO) (1:4, #ABIN112223, antibody on line.com, Aachen, Germany), Umod (1:25, #LS-C150268, LSBio, Seattle, WA, USA), CD31 (1:100, #550274 BD Biosciences), Laminin (1:200, #L9393, Sigma), and Cleaved Caspase-3 (1:200, #9661s, Cell Signaling Technology, Leiden, Netherlands). Stained organoids were mounted with Shandon™ Immu-Mount™ (Thermo Scientific™). A Zeiss LSM780 microscope and Zeiss Axiolab (Zeiss, Oberkochen, Germany) were used for image capture and analysis.

#### *2.6. Nephrotoxicity Assay*

3D kidney organoids were cultured in organ culture medium supplemented with gentamicin at 5 mg/mL (#G1264, Sigma) for 48 h, or with cisplatin at 5, 20, or 50 μM (#P4394 Sigma) for 24 h after day 8 of organ culture. Organoids were then fixed with 100% cold methanol for 30 min for whole-mount immunohistochemistry. The Notch inhibitor, *N*-*S*-phenyl-glycine-*t*-butyl ester (DAPT, #D5942, Sigma), was used (10μM) to investigate toxicity towards proximal tubule development.

#### *2.7. Statistical Evaluation*

All data are presented as mean ± standard deviation (SD) and represent a minimum of three independent experiments. Student's two-tailed t-test was used for statistical evaluation. *p*-value < 0.05 was considered significant.

#### **3. Results**

#### *3.1. Direct Di*ff*erentiation of mESCs into UB Progenitor Cells*

During development, both the nephron and ureteric bud progenitor cells are derived from the intermediate mesoderm (IM). To establish a protocol and direct differentiation of mESCs into UB lineage, we first differentiated mESCs into IM (Figure 1A). We treated the mESCs with FGF2 and activin A to differentiate mESCs into epiblast in monolayer culture (Supplementary Figure S1A). The differentiated cells showed expression of epiblast markers such as Fgf5 and T (Brachyury) (Supplementary Figure S1B). We then used glycogen synthase kinase-3β ( GSK-3β) inhibitor, CHIR99021 (CHIR), together with a low concentration of bone morphogenetic protein (BMP) signaling inhibitor, Noggin (suppresses number of cells differentiated to lateral plate mesoderm) [2], to activate differentiation of mouse epiblast cells into primitive streak (PS). The Noggin-treated cells expressed PS markers Cdx2, T, Tbx6, and Mixl1 (Supplementary Figure S1B). Activin A has been previously used for specification of the mesodermal cells towards intermediate mesoderm [19]. Therefore, we followed for two days with activin A treatment, which differentiated the cells to the IM stage and expressed Osr1, Lhx1, and Pax2 (Supplementary Figure S1B).

Previous studies have demonstrated that FGF9 is able to induce renal lineage differentiation from the IM population [2]. Therefore, we treated these cells with a moderate concentration of FGF9 for an additional three days, directing them to differentiate into UB progenitor cells with expression of UB markers. These cells expressed UB tip markers: Ret, Wnt11, and Sox9, as well as other markers of UB: Lhx1, Ecad, Hnf1b, Wnt7b, Wnt9b, Calb1, Emx2, Gata3, Hoxb7, and Tacstd2 (Figure 1B and Supplementary Figure S1C). In addition, expression of stromal cell marker Foxd1 nephron progenitor cell markers, Six2 and Eya1 (Figure 1B), or other epithelial segment markers, were observed at day nine of differentiation (Supplementary Figure S1D). Immunofluorescence staining further revealed that the use of a moderate concentration of FGF9 induced the cells to express Pax2, E-cadherin (Ecad), and Gata3 (Figure 1C–F), which may suggest that these differentiated cells represent putative UB progenitor cells.

**Figure 1.** Differentiation of mouse embryonic stem cells (mESCs) to ureteric bud (UB) progenitor cells. (**A**) Schematic of the differentiation protocol of mESCs into UB progenitor cells. AA: Activin A; F2: FGF2; C: CHIR99021; N: Noggin; F9: FGF9. (**B**) Graphs of qPCR results showing the gene expression (fold change) of ureteric bud markers relative to mESC. No expression of stroma (Foxd1) and nephron progenitor cell markers (Six2 and Eya1) was observed at day 9 of differentiation. (*n* = 3). (**C**–**E**) Immunocytochemistry of Pax2, Ecad, and Gata3 in mESCs on day 9 of differentiation. Scale bars, 50 μm. (**F**) Quantification of the number of cells expressing Pax2, Ecad, and Gata3 at day 9 of differentiation. *n* = 3 samples per marker (3 randomly chosen areas in 3 independent experiments).

#### *3.2. Generation of Kidney Organoids by mESC-Derived UB Progenitor Cells and Dissociated Primary MM Population*

We and other groups previously reported that dissociation of mouse pMM into single cells maintains the nephron progenitor stemness. The dissociated MM population develops into nephrons when induced by the inducer such as the embryonic UB or spinal cord cells [8,21,23–27]. To establish the potential and function of the mESC-derived UB progenitor cells, we aggregated these cells with mouse E11.5-dissociated pMM cells to generate a kidney organoid. The cell aggregates were cultured for up to 11 days in a traditional Trowell organ culture system, during which they spontaneously formed kidney organoids with complex structures (Figure 2A,B). On day three, we observed Troma1+ UB structures and the formation of renal vesicles adjacent to the UB (Figure 2C–D).

**Figure 2.** Generation of renal organoids by mESC-derived UB progenitors and primary metanephric mesenchyme (pMM) cells. (**A**) Schematic of generation of kidney organoids from mESC-derived UB progenitors with mouse E11.5-dissociated pMM. (**B**) Global bright field images of self-organizing kidney organoids in a Trowel organ culture system. Scale bars, 500 μm. (**C**) Immunofluorescence of

kidney organoids at day 3 show formation of a renal vesicle next to the Troma1+ structure generated by mESC-derived UB progenitor cells. (**D**) Confocal image at day 3 showing Pax2+ renal vesicle (yellow dotted line) surrounded by Troma+Pax2+ ureteric epithelium (white dotted line). (**E**–**M**) Immunofluorescence of kidney organoids at day 8 or 11. (**E**) Glomeruli are marked with Wt1, proximal tubules are marked with LTL, and distal tubules are marked with Pax2+LTL–. The arrowhead shows the connection location of glomeruli with proximal tubule. (**F**) Immunostaining of distal tubule marked with Pax2 and proximal tubule marked with LTL, with all nuclei stained with DAPI. The arrow shows the proximal tubules (Pax2+LTL+) connected with distal tubules (Pax2+LTL–). (**G**) Glomeruli (Synpo+) adjacent to proximal tubules (LTL+) (the arrowhead marks the place of connection). (**H**) Immunostaining of proximal tubule with LTL and distal tubule with Ecad. Confocal image shows proximal tubules (LTL+) connected with distal tubules (Ecad+LTL–) (marked with an arrow). (**I**) Confocal image of Loops of Henle marked by Umod and Ecad on day 11. (**J**) Confocal images of Troma1+Gata3+Ecad+ collecting duct structure. (**K-K'**) A "T" shaped UB structure in the kidney organoid. (**L**) mESC-derived UB progenitor cells-generated collecting ducts (Troma1+Ecad+) connecting with nephron's distal tubules (Troma1–Ecad+) (marked with an arrow). (**M**) Kidney organoids developed collecting duct trunk structures. Scale bars, (**B**) 500 μm, (**E**) 50 μm, (**C**–**D**, **F**–**M**) 20 μm.

Whole-mount immunostaining of day eight chimeric organoids showed development of nephrons with positive staining of glomerular markers Wilms tumor 1 (Wt1+) and Synaptopodin (Synpo+), proximal tubule marker Lotus Tetragonolobus Lectin (LTL+), and distal tubules markers Pax2+LTL– and Ecad+ (Figure 2E–H, Supplementary Figure S2A). Moreover, we found numerous Synpo+ and Wt1+ glomeruli adjacent to LTL+ proximal tubules, which connected with Pax2+LTL–/Ecad+ distal tubules (Figure 2E–H marked with arrowheads and arrows, Supplementary Figure S2A), indicating a proper nephron structure with glomerulus/proximal tubule/distal tubule organization. On day 11, the organoids also displayed Henle's loop with the expression of uromodulin (Umod+) and Ecad+ (Figure 2I).

We also tested whether the mESC-derived UB progenitor cells have in vivo UB capacity to form the collecting duct system. On day eight, we observed Troma1+Gata3+Ecad+Pax2+LTL– collecting duct structures in the kidney organoids by immunocytochemistry (Figure 2J–M, Supplementary Figure S2B). We also found some Troma1+ "T" shaped UB structures (Figure 2K, K'), indicating that the mESC-derived UB progenitors behave in a manner similar to UB cells in vivo. Importantly, we observed that the collecting ducts (Troma1+Ecad+) connected with the distal tubules (Ecad+Troma1–) of the nephron structure (Figure 2L, Supplementary Figure S2C), generating an interconnection between collecting ducts and nephrons, which is essential for urine drainage. The collecting ducts displayed the morphology of branches and long collecting duct trunks (Figure 2M). Altogether, the *in vitro* reconstructed organoids developed kidney structures that are similar to in vivo kidney, although it is unclear if the generated UB structures form a proper network.

We have also attempted to induce vascularization of developing glomeruli in these renal organoids. Previous studies demonstrated that rho-associated protein kinases (ROCK) are downstream effectors of vascular endothelial growth factor (VEGF), and negatively regulate the process of angiogenesis [28]. Therefore, we used the ROCK inhibitor to enhance angiogenesis in the renal organoids. However, this treatment did not increase the endothelial network area, and CD31+ cells could not be found in the developing glomerular tuft (Supplementary Figure S3).

#### *3.3. Characterization of Kidney Organoids*

To rule out the possibility that the differentiated cells (mESC-derived UB progenitors) could give rise to nephrons when induced by the embryonic UB, we aggregated day nine differentiated cells with E11.5-dissociated UB cells and grew them in organ culture. Since UB survival and development depend on the presence of metanephric mesenchyme, in the organ cultures, the purified UB failed to branch in the presence of UB-differentiated cells (Supplementary Figure S4A). Next, we wanted to verify that nephron structures in the organoids were generated via the MM population induced by the mESC-derived UB progenitors, and not by contamination of the MM with the primary UB cells. To assess this possibility, we cultured E11.5 MM tissue in isolation. The tissue already underwent apoptosis at the second day of culture and died at day three (Supplementary Figure S4B). This result also confirmed that without a suitable inducer, the MM cells will not undergo nephrogenesis and cannot survive for a long time under the organ culture conditions in vitro. These data suggest that we did differentiate mESC towards UB progenitors, and that they can further develop into collecting ducts when co-cultured with pMM.

To further confirm that the nephron structures and collecting ducts were derived from pMMand mESC-derived UB progenitors respectively, we used pMM population isolated from mTmG mice (td-tomato+) to aggregate with differentiated mESCs—unlabeled (Figure 3A). The aggregates formed well-developed nephron structures (WT1+ glomeruli, LTL+ proximal tubules, Ecad+Troma1– distal tubules), which originated from MM cells (mTmG+) (Figure 3B) and Troma1-labeled collecting duct, which was mTmG– and therefore originated from the mESC-derived UB progenitors (Figure 3C–E). We also generated an mESC line with stable expression of the GFP. We differentiated these cells into UB progenitors and used them to induce pMM cells. Immunofluorescence analysis confirmed that the Troma1+Ecad+ collecting ducts were derived from GFP+ mESCs (Figure 3F, Supplementary Figure S4C). Altogether, these data demonstrate that the nephrogenesis occurring in generated kidney organoids is specifically derived from an interaction between the pMM and mESC-derived UB progenitors.

#### *3.4. mESC-Derived UB Progenitor Cells Integrated into the 3D Ureteric Bud Structures*

In order to identify the mESC-derived UB progenitors from wild-type UB cells in chimeric organoids, and to ensure that they will only generate UB cells, we generated an organoid with an mESC-GFP line. We dissociated a whole E11.5 embryonic kidney rudiment (MM and UB) and mixed it with undifferentiated mESC or mESC-derived UB progenitors generating chimeric kidney organoids [19,21,27] (Figure 4A).

**Figure 3.** UB structures in kidney organoids are specifically derived from differentiated mESCs. (**A**) Schematic of generation of kidney organoids from mESC-derived UB progenitors with mouse E11.5-dissociated mTmG (td-tomato+) pMM. (**B**) Confocal images of three-dimensional (3D) kidney organoids derived from E11.5 mTmG+ pMM and mESC-derived UB progenitors. Nephron structures such as WT1+ glomeruli and LTL+ proximal tubules were derived from mTmG+ pMM. Scale bars, 50 μm. (**C**) Confocal images of kidney organoids showing ureteric bud structures (Ecad+Troma1+) being generated by mESC-derived UB progenitor cells. Scale bars, 20 μm. (**D**) Confocal images showed Troma1+Ecad+ collecting duct derived from mTmg– cells (mESC-derived UB progenitors) and connected with Troma1–Ecad+ renal tubules (connection marked with an arrow). Scare bars, 20 μm. (**E**) The Troma1+ UB structures derived from mTmG– cells (mESC-derived UB progenitors). Scale bar: 20 μm. (**F**) Troma1+Ecad+ collecting ducts are derived from GFP+ mESCs-derived UB progenitors. Scale bar: 20 μm.

**Figure 4.** Mouse ESC-derived UB progenitor cells form anembryonic UB in 3D organ culture in vitro. (**A**) Schematic of kidney organoid generation from mESC-derived UB progenitors with mouse E11.5-dissociated kidney rudiments. (**B**) Immunofluorescence analysis demonstrating random

localization of undifferentiated mESCs in organ co-cultures. Scale bars, 20 μm. (**C**) Immunofluorescence analysis demonstrating mESC-derived UB progenitors integrated into the UB structures and enhanced chimeric ureteric bud formation. Scale bars, 20 μm. (**D**) Confocal image showing localization of mESC-derived UB progenitor cells (GFP+) in Troma1+ UB structures of the chimeric organoid and not in the Six2+ nephron progenitor's region. Scale bars: 20 μm.

The undifferentiated mESCs aggregated with dissociated embryonic kidney rudiments randomly localized within the organoids and had a negative effect on nephrogenesis. The undifferentiated mESC had formed groups within the renal structures; however, there were only a few structures, and very few Troma1+ ureteric buds could be observed (Figure 4B). When we aggregated the mESCs-derived UB progenitors with the dissociated embryonic kidney, we noticed much more robust tubulogenesis and numerous chimeric ureteric bud structures (Figure 4C). The UB progenitor cells did not interfere with renal development, but efficiently and specifically integrated into the mouse ureteric buds, as indicated by Troma1 and GFP+ staining (Figure 4C and D). Moreover, the immunostaining of day three organoids with nephron progenitor's marker, Six2, presented integration of mESC-derived UB progenitors (GFP+) into the UB structure (Figure 4D). Together, these data demonstrate that mESC-derived UB progenitors do integrate well into ureteric bud structures in chimeric renal organoids and are able not only to form UB de novo, but also generate chimeric structures.

#### *3.5. Kidney Organoids as Models to Study Kidney Development and Drug Toxicity*

After transfer of the 3D chimeric organoids to organ culture, they self-organized and generated a well-structured kidney organoid with proper nephron and collecting duct structures. As patterning of the nephron into its different segments begins at the renal vesicle stage during development [29,30], we postulated that developmental patterning could be changed by chemical modulation of these endogenous signals. A previous report revealed that Notch signaling is required for proximal tubule fate acquisition in a mammalian nephron [31]. We therefore treated the organoids (pMM cells mixed with mESC-derived UB progenitor cells in a ratio of 3:1, Figure 2A) with a Notch signaling inhibitor, *N*-*S*-phenyl-glycine-*t*-butyl ester (DAPT). DAPT leads to suppression of proximal tubules' formation in the human nephron organoid culture [4,31,32]. We added DAPT to the culture medium from day 3 to 11 of 3D organoid culture. Immunofluorescence analysis of the cells demonstrated that with the DAPT treatment, formation of glomerulus and ureteric bud ducts was normal, but development of proximal tubules was severely suppressed (Supplementary Figure S5A).

In order to use stem cell-derived kidney organoids for disease modelling and drug screening, they need to present functional maturation of the nephrons within the organoids. To test whether these organoids could be used to study kidney injury and toxicity in vitro, we focused on drug-induced nephrotoxicity, which has been shown as an important cause of acute kidney injury in hospitalized patients [33]. We treated the chimeric kidney organoids with gentamicin, a commonly used antibiotic with well-established proximal tubular toxicity, after nine days of organ culture for 48 h [34]. We also treated organoids with another nephrotoxicant, cisplatin, from day 10 for 24 h. Cisplatin induces caspase-mediated acute apoptosis of proximal tubular cells in the kidney [35]. Whole-mount immunostaining of control organoids with caspase 3 showed random apoptotic interstitial cells; however, both gentamicin and cisplatin induced acute apoptosis in LTL+ proximal tubules (Supplementary Figure S5B). The percentage of apoptotic proximal tubular cells induced by cisplatin increased in a dose-dependent manner: the 5 and 20 μM cisplatin doses mainly affected the proximal tubule compartment, 32% and 62% respectively, while 50 μM cisplatin led to apoptosis of almost all of the proximal tubule cells (≈96%), but was also toxic to other cell types presenting a global type of nephrotoxicity at this concentration (Supplementary Figure S5C).

To summarize, our work shows that mESC-derived UB progenitors induce nephrogenesis in pMM cells, and furthermore, chimeric renal organoids generated from these progenitors show an expected response to toxic chemicals and drugs. We also demonstrated that the kidney organoid system can be used to test nephrotoxicity of drugs and other chemicals in vitro.

#### **4. Discussion**

Kidney development starts from an interaction between two precursor tissues of the kidney, UB and MM. A major part of the MM cell population comprises the nephron progenitor cells (NPCs) which will differentiate into nephrons, and the ureteric bud will form the collecting duct tree. Protocols for differentiation of mouse and human pluripotent stem cells to renal progenitor cells, and further to self-organized kidney organoids containing nephrons, have been well established [2–4,13,36], but current methods of differentiation of pluripotent stem cells, specifically to ureteric bud progenitor cells, need further development [8,19].

Previous studies on generation of ESC-derived UB have shown derivation of a UB-like population by selective induction with metanephric mesenchyme cells [19] or through CHIR99021 treatment for nephron differentiation of both ureteric bud and nephron structures [2]. However, the ESC-derived UB-like cells did not show nephron progenitor induction [19], and therefore, an inter-nephron connection with collecting ducts was lacking [2]. Another group recently published a protocol for derivation of UB structures from PSCs, and generation of kidney organoid composed of mESCs-derived UB aggregated with pMM, or mESCs-derived UB combined with mESC-derived NPC and primary stromal progenitor cells (SPs). Although successful, their protocol requires a knock-in of markers in the PSCs, which later involves sorting of cells for the specific marker [8]. More recently, Mae and colleagues reported a protocol for generation of branching ureteric bud tissues from human pluripotent stem cells (hPSC) with a series of growth factors, but there was no evidence of nephrogenesis [37] (Supplementary Table S2). While these published protocols produce functional kidney organoids, they are technically complex. Here, we reported an establishment of a simple (directed with growth factors), efficient (>90% of Pax2+, Ecad+ and >70% Gata3+ cells), and reproducible differentiation protocol of mESC to ureteric bud progenitor cells. These mESC-derived UB progenitor cells induced pMM cells to undergo nephrogenesis leading to development of well-structured nephrons. These nephrons consisted of glomeruli, proximal tubules, loops of Henle, and distal tubules, and were connected with collecting duct structures generated by mESC-derived UB cells. Moreover, these mESC-derived UB progenitor cells formed a UB de novo when combined with pMM cells, and they generated chimeric structures when combined with kidney rudiment cells (pUB and pMM) (Supplementary Table S2).

While our culturing conditions produced well-functioning kidney organoids, further studies are needed to fine-tune the culturing conditions. Previous studies showed that ROCK kinases are VEGF downstream effectors, which negatively regulate the process of angiogenesis [28]. We added ROCK inhibitor (Y27) to culture expecting an increase in angiogenesis. However, we failed to see a difference between control and Y27-treated samples. Better results could be obtained by supplementing the medium with VEGF, which was already published by Freedman and co-workers, although they still did not observe endothelial cells entering the glomerular tuft in their system [9]. Similarly, development of a vascular glomeruli in organotypic kidney cultures and renal organoids was reported earlier by our research team [38]. It seems that in the absence of the blood flow, the endothelial cells are not able to properly interact with developing nephrons [39] and formation of glomerular vasculature does not proceed further than migration of endothelial cells into the vascular cleft region of an S-shaped stage nephron. Sufficient vascularization of the organoids may be achieved by treatment with angiogenic factors, co-culture with blood vessel organoids [40], by providing flow to the system [15], or with a combination of the aforementioned treatments.

Even though organoids generated by mESC-derived UB progenitor cells did not have a proper vascular network, the developed nephrons did respond to toxicological tests as expected. We and others [4] have shown that the use of gentamicin and cisplatin induces apoptosis in proximal tubular cells. Thus, these organoids present a functional platform to test drug-induced nephrotoxicity.

In summary, we have developed an easy and reproducible protocol for generation of UB progenitors from mESCs. This work generates a strong foundation for in vitro kidney studies, including disease modelling and drug discovery approaches, which are difficult to perform, and require animal models and/or primary cells which may not faithfully recapitulate all features of developmental or disease processes. Given the rapid progress in the field, we hope that in the near future, researchers will be able to generate fully functional nephrons in kidney organoids where the UB and MM parts are derived from PSC. Using these cells will enable generation of not only well-structured nephrons, but also the collecting duct tree. This is the first step for generating high-throughput gene discovery models and advancing tissue engineering for producing organs for transplanting. However, these organoids need to be successfully vascularized and grown to appropriate size. The studies presented here produce new insights into renal pathophysiology and open new avenues for developing new treatment options.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4409/9/2/329/s1. Table S1: Primers (5 -3 ) used in the study. Table S2: Summary of available protocols of differentiation of mouse and human PSC to UB linages. Figure S1: Differentiations of mESC into ureteric bud progenitor cells. Figure S2: Differentiated mESC induce embryonic MM to nephrogenesis. Figure S3: Vascularization of the kidney organoids. Figure S4: Characterization of kidney organoids. Figure S5: Kidney organoids model kidney development and injury.

**Author Contributions:** Conceptualization, Z.T., A.R.-R., I.S., and S.J.V.; Funding acquisition, S.J.V.; Investigation, Z.T.; Visualization, Z.T.; Writing—original draft, Z.T.; Writing—review and editing, A.R.-R. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported financially by H2020 Marie Skłodowska-Curie Actions Innovative Training Network "RENALTRACT" (ID 642937), Academy of Finland (ID 315030, Centre of Excellence, ID 251314), Sigrid Jusélius Foundation, Cancer Research Foundation, and Finnish Cultural Foundation (personal grant to A.R.-R).

**Acknowledgments:** We thank Paula Haipus, Hannele Härkman, Johanna Kekolahti-Liias, and Sanna Kauppinen for technical assistance; Biocenter Oulu Transgenic core facility for the mouse ESCs; Dr. Susanna Kosamo for critical reading of the manuscript and English language corrections; Dr. Jingdong Shan, Dr. Florence Naillat, MSc. Abhishek Sharma, Dr. Fariba Jian Motamedi, and Prof. Andreas Schedl for the discussion.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Review* **Traditional and Advanced Cell Cultures in Hematopoietic Stem Cell Studies**

#### **Antonio Carlos Ribeiro-Filho 1, Débora Levy 2, Jorge Luis Maria Ruiz 3, Marluce da Cunha Mantovani <sup>1</sup> and Sérgio Paulo Bydlowski 1,2,4,\***


Received: 13 August 2019; Accepted: 4 December 2019; Published: 12 December 2019

**Abstract:** Hematopoiesis is the main function of bone marrow. Human hematopoietic stem and progenitor cells reside in the bone marrow microenvironment, making it a hotspot for the development of hematopoietic diseases. Numerous alterations that correspond to disease progression have been identified in the bone marrow stem cell niche. Complex interactions between the bone marrow microenvironment and hematopoietic stem cells determine the balance between the proliferation, differentiation and homeostasis of the stem cell compartment. Changes in this tightly regulated network can provoke malignant transformation. However, our understanding of human hematopoiesis and the associated niche biology remains limited due to accessibility to human material and the limits of in vitro culture models. Traditional culture systems for human hematopoietic studies lack microenvironment niches, spatial marrow gradients, and dense cellularity, rendering them incapable of effectively translating marrow physiology ex vivo. This review will discuss the importance of 2D and 3D culture as a physiologically relevant system for understanding normal and abnormal hematopoiesis.

**Keywords:** hematopoiesis; hematopoietic stem cells; stem cell culture; 2D culture; 3D culture

#### **1. Introduction**

Blood is a connective tissue made up of approximately 34% cells and 66% plasma, transporting nutrients, gases and molecules in general to the whole body. Hematopoiesis is the process by which blood cells are formed, replenishing the blood system over the life of an individual. The hematopoietic process is a highly hierarchical phenomenon, in which hematopoietic stem cell (HSCs) differentiation and proliferation are of vital importance. Each cell within the hematopoietic hierarchy can be distinguished based on specific surface markers, which contain epitopes that are recognized by antibodies [1–3]. Figure 1 shows the main markers in human hematopoietic hierarchy. The hematopoietic process in humans starts in the yolk sac (mesoblastic phase). Then, it is transferred to the liver and spleen. Finally, the bone marrow becomes the main organ responsible for hematopoiesis. In the bone marrow, HSCs have the capacity of unlimited self-renewal, producing progeny that is the same as the original cell. They are generally in the G0 phase of the cell cycle and have the capacity to differentiate into specialized cells.

**Figure 1.** Hierarchy of human hematopoiesis. LT-HSC: Long Term-Hematopoietic Stem Cell; ST-HSC: Short Term-Hematopoietic Stem Cell; MPP: Multipotent Progenitor; OPP: Oligopotent Progenitor; LRP: Lineage Restricted Progenitor; MEC: Mature Effector Cell. The markers of the most important lineages are listed: Common Lymphoid Progenitor (CLP); Common Myeloid Progenitor (CMP); Megakaryocyte-Erythrocyte Progenitor (MEP); Granulocyte-Macrophage Progenitor (GMP). Restricted lineage progenitor cells: Megakaryocyte Progenitor (MkP); Erythrocytic Progenitor (EryP); Granulocytic Progenitor (GrP); Monocyte Progenitor (MncP); Dendritic Progenitor Cell (Pro DC); Progenitor Cell-T (Pro-T); Progenitor Cell-B (Pro-B); Progenitor Cell-Nk (Pro-Nk).

#### **2. Hematopoietic Stem Cells**

The medullary microenvironment participates in the quiescence, self-renewal, proliferation, maturation and apoptosis of HSCs and contains several cells (i.e., mesenchymal stem cells, endothelial cells, fibroblasts, osteoblasts, reticular cells, adipocytes). These cells are sources of cytokines, growth factors, glycoproteins and glycosaminoglycans, among other regulators. Different combinations of these molecules lead to the formation of specific microenvironments within the medullary cavity, known as niches [4]. Histologically defined microenvironments are subdivided into four regions: endosteal, subendosteal, central, and perisinusoidal. Granulocytes and monocytes are found in all regions of the bone marrow, whereas erythroblasts proliferate preferentially in the central region [5]. Concerning the dynamics of the lymphoid lineage, B lymphocyte precursors are found in the subendosteal region, gradually decreasing toward the central region, whereas mature B cells are found throughout the bone marrow [6]. HSCs are located in the endosteal region, also known as the osteoblastic niche, but studies suggest that HSCs may migrate to the perisinusoidal region or vascular niche and remain quiescent or differentiate depending on the needs of the organism [2,5,7–9]. In fact, studies with new markers for HSCs and niche cells, new image techniques, including labeling protocols, have shown that most HSCs reside adjacent to sinusoidal vessels, leading to the proposed existence of a perivascular niche for HSCs [10]. It is assumed that in the bone marrow, there are at least two different niches: the endosteal niche, which would harbor quiescent HSCs, and the perivascular niche, which would harbor cycling HSCs [11]. Although most studies have been done on non-humans, researchers suppose that the data reflect what happens in humans. It has previously been proposed that HSCs are maintained in the endosteal (osteoblastic) niche; however, the available evidence does not seem to support this model. Nevertheless, the endosteal niche seems to support the maintenance of a subset of lymphoid progenitors [10]. Approximately 80% of dividing and non-dividing HSCs have been described to be associated with sinusoidal vessels, with another 10% of HSCs being adjacent

to arterioles, and almost another 10% in transition zone vessels [10]. A small percentage of HSCs are located in the endosteum. In fact, a quantitative model of cellular components that could define these microenvironments, and the preferential location of HSCs in the bone marrow are still lacking [12]. Obstacles to recognizing the HSCs in the bone marrow include the low frequencies at which HSCs are found in the bone marrow and the cellular complexity of the bone marrow microenvironment.

Cell signaling in the HSC niche is a complex process and passes through an extensive signaling network balancing self-renewal and differentiation [13]. This signaling involves several substances, such as growth factors and cytokines that are secreted by both the medullary stroma cells and hematopoietic stem cells and are important signaling factors for hematopoiesis, proliferation, and differentiation [14].

The Jak/STAT, Ras/Erk, and PI3K/Akt signaling pathways have been described as important inducers of erythropoiesis transduction by activation of the erythropoietin receptor, and these intracellular pathways are responsible for the survival, proliferation, and differentiation of normal erythropoietic progenitors [15].

Other intracellular signaling pathways that are important for the control of hematopoiesis have been described, such as the Notch, Wingless-type (Wnt), and Hedgehog pathways [16], which have been associated with self-renewal and maintenance of HSCs. Notch proteins are highly conserved receptors on the surface of the cell membrane that regulate the development of stem cells, and mutations in this receptor may cause leukemia [17] and breast cancer [18]. Activation of the Notch pathway is necessary to keep HSCs undifferentiated. This pathway is more active in HSCs and less active in differentiated cells. Inhibition of the Notch pathway potentiates the differentiation of HSCs and loss of the bone marrow reconstitution capacity of sub-lethally irradiated animals; thus, Notch has been used as a marker of undifferentiated HSCs [19]. Wnt protein regulates several phenomena during fetal development, and this protein has been related to the self-renewal of stem cells [20]. Hedgehog (Hh) protein has been described as regulating embryonic and adult stem cell activity. In mammals, three genes are known to be responsible for this protein—Sonic Hedgehog (SHh), Indian Hedgehog (IHh), and Desert Hedgehog (DHh) [21,22].

Soluble factors are also closely associated with the maintenance and regulation of the undifferentiated state of HSCs in the bone marrow of adults, in addition to regulating the proliferation and differentiation of this population. Stromal-cell-derived factor-1 (SDF-1/CXCL12) and its CXCR4 linker are activated to recruit endothelial progenitor cells (EPCs) and regulate HSCs [2,13,23]. Other soluble factors act to promote the maintenance of HSCs in their niche; for example, the stem cell factor (SCF/Kit-ligand) and its c-Kit receptor (CD117) are both required by HSCs for their maintenance. SCF is an important soluble cytokine for hematopoiesis, and THE c-Kit receptor is expressed on the HSCs surface; altered forms of this receptor have been associated with several types of cancer [24,25]. Thrombopoietin (TPO) and its MPL ligand are also important soluble factors necessary for the maintenance of HSCs in their niche. TPO is a primary physiological regulator responsible for the stimulation of platelet production, a primary dominant factor and megakaryocytopoiesis stimulator. In addition, recent in vitro studies have shown that TPO alone or in combination with growth factors, such as a c-Kit ligand, IL-3, or even Flt-3, stimulates the proliferation of hematopoietic progenitor cells [26,27]. Many other factors also modulate the function of HSCs but are not necessarily required, such as angiogenin, angiopoietin-1, G-CSF, IL-6, and TGFβ, among others [25].

#### **3. Stem Cell Culture Methods**

The complex microenvironment and cellular interactions are difficult to reproduce in vitro. Some improved techniques can help researchers mimic the bone marrow architecture for hematopoiesis studies and cell production in regenerative medicine applications.

Cell culture is the growth of cells from animal or plant in a favorable, artificial, and controlled environment. Today, cell culture is the basis of biology techniques and essential for regenerative medicine procedures [28]. These cells can originate directly from the tissue or after enzymatic or mechanically dissociated tissue or can be derived from an already established cell line.

Traditional cell culture, also known as 2D cell culture, is a very well-established method and easy-to-use culture model (Figure 2A,C). This technique depends on adherence to a flat surface, typically a culture flask, dish or polystyrene glass, to provide mechanical support for its growth in monolayers and access to nutrients and growth factors present in the culture media. Cells can also grow in suspension, as those derived from blood or bone marrow [29]. This method may present some advantages, such as low cost and performance of the functional assays, but this type of culture may also present some disadvantages, such as not mimicking the natural structures of the tissue and not being able to efficiently mimic cell–cell and cell–environment extracellular interactions [29].

**Figure 2.** Two-dimensional (2D) and three-dimensional (3D) cell culture. (**A**) Schematic model of 2D cell culture. Standard model of 2D cell culture. Cells are cultured as a single layer in a culture flask. (**B**) Schematic model of 3D cell culture. 3D model of cell culture giving the notion of height, width and depth; cells are surrounded by the medium. (**C**) Example of 2D culture. HEK 293 cells in 2D cell culture; the adherence to a flat surface provides mechanical support for growth in monolayers. Scale bar 400 μm. (**D**) Example of 3D culture. HEK 293 cells 3D cell culture in agarose allowing the cells to grow or interact with their surroundings in all three dimensions. Scale bar 1000 μm.

Primary cell culture has been one of the greatest tools of cellular biology for evaluating cellular aspects such as chemoresistance, karyotypes, cellular parameters, metabolism and in vitro modeling of physiological and pathological models [30]. The isolated cells resemble their tissue of origin. Primary cell culture is complex and requires specific care, including storage, thawing-freezing procedures, or choice of enzymatic treatment, but the main problem is how to keep the primary cell in culture long enough to be established for experimental tests [31–33]. Primary culture can be representative of the cell types of the tissue from which they were isolated; although these cells are difficult to maintain, they best mimic a pathological condition or physiological function.

Immortalized cells or established cell lineages are cells that can grow in vitro indefinitely due to natural or induced transformation (e.g., embryonic stem cells or viral transformation) [34]. Immortalized cells have several advantages over primary cells, such as profitability, easy manipulation, unlimited supply of material, and lack of ethical concerns associated with the use of animal and human tissues [35]. Immortalized cells also provide a uniform and homogeneous population of cells [36]. Immortalized cell lines emerged as a solution to some problems that appeared with primary cell cultures, such as misidentification of the cell line used and genotypic and phenotypic instability. Care should also be taken to avoid the use of aged cell lines that do not maintain the original physiological characteristics. The immortalized strains constitute a simple and representative model system for functional studies and therapeutic targets. As each cell line may have unique characteristics, specific studies should take these characteristics into account [37].

The co-culture technique is used for numerous applications, including the study of natural or synthetic interactions between distinct populations of cells. Co-culture methods are of great importance in research, as they are used to observe cell–cell interactions, how cells are organized, how they participate in the development of diseases such as cancers, in which different types of cells are involved, including the microenvironment [38].

Both self-renewal and differentiation are required abilities in any cell of the hematopoietic lineage [39]. However, the expansion of these cells in vitro has been challenging for the scientific community, as regulation of these cells depends on several mechanisms of intercellular communication resulting from the secretion of local- and systemic-acting factors [40]. HSCs, which can also be obtained and expanded from umbilical cord blood, should respond to and integrate events in their microenvironment to ensure sustained production of all hematopoietic lineages [41].

Although HSCs have been extensively analyzed and characterized, their ex vivo expansion remains a problem [42–44]. It has been shown that the cell culture of HSCs is viable for at least 169 days but very much depends on the quality of the HSCs. There are currently several approaches to ex vivo expansion. Most first require the isolation of CD34<sup>+</sup> or CD133<sup>+</sup> cells from frozen or fresh hematopoietic tissue and incubation in culture medium supplemented with cytokines, granulocyte colony stimulating factor (G-CSF), stem cell factor (SCF), and thrombopoietin (TPO) [45].

Therefore, the cell culture of HSCs must take into account the microenvironment in which these cells are inserted so that it can reproduce the whole framework of the hematopoietic structure. The hematopoietic niche contains several types of cells, and mimicking this microenvironment in vitro requires a stromal layer, which controls multiple cellular parameters, including quiescence, self-renewal, differentiation, apoptosis and migration. Under the artificial culture conditions, HSCs undergo differentiation and apoptosis [46]. A support layer is required so HSCs can survive and proliferate.

Co-culture of HSCs is one of the most frequently used models for understanding how the highly specific bone marrow niche interacts ex vivo with hematopoietic cells, promoting their differentiation and expansion. The technique can focus on the importance of non-contact culture systems on the successful maintenance of hematopoietic cells [47] and on the use of the stroma as a cell layer to provide support for HSC culturing, as mentioned above. Several studies have shown that contact between HSCs and stromal cells is important for maintaining HSC function [48,49]. The stromal cells from bone marrow include osteoblasts, macrophages, endothelial cells, and mesenchymal cells [50,51]. Mesenchymal cells are multipotent progenitor cells that can differentiate into mesenchymal cells, such as osteoblasts, adipocytes, and chondrocytes, and also have the potential for differentiation into cells such as neurons and lung cells [52]. These multipotent cells have interactions with hematopoietic cells forming the framework of the HSC niche, supporting the development of hematopoiesis and acting as immunological regulators [53,54]. In addition, recent studies have indicated that other molecules, such as N6-methyladenosine (M6A), by modulating the expression of a group of YTHDF2 genes at the mRNA level, are important regulators of HSC self-renewal. Some authors have suggested that although extensive efforts have led to multiple methods for in vivo expansion of HSCs, it is not

possible for some single molecules or pathways to be manipulated simultaneously due to a large number of essential targets for self-renewal of stem cells [55].

In this regard, there are still important challenges to overcome, including the development of more efficient methods for the maintenance of HSCs in vitro, and methods of ex vivo expansion for therapeutic development in regenerative medicine; then, it will be possible to have a better understanding of the hematopoietic niche and its intrinsic and extrinsic regulators from a physiological and pathophysiologic point of view. The ability to genetically reprogram HSCs for clinical therapeutic use [56] still needs to be improved.

#### **4. 3D Hematopoietic Stem Cell Culture**

Cell culture in 2D format is simple and provides excellent material for studying homogeneous populations [57]. However, it does not consider several other important parameters of cell physiology, such as cell–cell communication or communication between the cell and microenvironment or adjacent molecules [58]. Therefore, the major problem with 2D cell culture is its limits [59]. One of the main criticisms of this study format is that animal physiology cannot be mimicked using only one layer of cells, which certainly does not correspond to the original tissue considering the complexity of the cellular microenvironment in the tissue of origin [60].

As an alternative to the technical limitations of 2D cell culture, 3D cell culture allows a better simulation of the in vivo structural complexity, replicating several characteristics present in tissues (Figure 2B,D), not only the interaction of cells with their microenvironment, but also morphology, differentiation, polarity, proliferation rate, gene expression profiles, and cell heterogeneity [61–64]. In fact, 3D culture has proven to be a realer model for translating research results into in vivo applications.

Current 3D cell culture models and methods include spheroids, organoids, microcarrier cultures, organs-on-chips, and 3D bioprinting [65,66]. Organotypic explant culture methods are used mostly when a technical/specific requirement of the tissue is to be studied and mimicked [67]. However, although the complexity of the 3D system is evident, some criteria must be considered and cannot be disregarded, such as the choice of material for the scaffold and hydrogels and cell type and culture methods, which vary considerably according to the tissue studied [66].

Organoid culture is most commonly used to describe constructs derived from pluripotent stem cells (embryonic or induced cells) or adult stem cells from various organs, including the hematopoietic tissue [68]. Organoid culture is classified as either tissue organoids (i.e., organotypic) or stem cells depending on how the cell layers are formed [69]. Organotypic tissue refers to free stromal cells without parenchyma, and its application is mainly with epithelial cells because of their high intrinsic ability to self-organize. With this method, it is possible to study HSCs, and Christopher et al. [70] were able to produce mature T cells from stem cells and hematopoietic progenitors in a thymus organoid culture.

Spheroid culture models simulate the microenvironment conditions of a living cell. Compared to the classic 2D model, the spheroid culture model emphasizes the interactions between cells and their relationship to the extracellular matrix (ECM) [71,72]. Tissues are not composed of a homogeneous population of cells; they are complex structures formed by several different components, with an intricate relationship such as vessels, nerves, and stroma, which should be considered in any kind of tissue engineering. This complex interaction between cells and the extracellular compartment requires a 3D environment to best represent these interactions [73,74].

Spheroid cell culture can be used in scaffold and scaffold-free models. The scaffold methods comprise hydrogel support, decellularized extracellular matrix, and resistant polymeric material support. Hydrogel consists of polymeric groups with a water-swollen hydrophilic structure [75]. Depending on their nature, these compounds are classified into different categories, including ECM protein–containing hydrogels (ECMPs), natural hydrogels, or synthetic hydrogels, each one with its own properties [76,77]. The decellularization technique consists of removing cells from the native tissue by chemical treatment, preserving the ECM. The tissue can then be replenished with cultured cells and

grafted in vivo or used for ex vivo cell differentiation. In contrast to artificial scaffolds, decellularized scaffolds preserve the vascular structure, tridimensional niches, and chemical composition of bone marrow. Extracellular matrix components are directly related to several important factors for cell survival support, including cell behavior, signaling for survival, and proliferation, among others. Resistant polymeric material supports are structures similar to fibers or sponges. Cells maintained in this type of support exhibit a physiological behavior close to that of native tissue (such as those of cartilaginous tissue). Regarding the polymeric support, the most commonly used material is polystyrene, but biodegradable supports such as polycaprolactone have also been employed.

The development of a bioprinting system for the bone marrow microenvironment is important since it cannot be mimicked by methods such as organoids. Advanced 3D bone marrow models could serve for several different studies, including hematopoietic regulation, BM, MSCs, HSC interaction and expansion, interactions of hematological cancer cells, and evolution of several hematological diseases. In addition, these models could be used as a platform for expanding HSCs for transplantation. However, it is difficult to obtain a reliable BM model because of several critical technical challenges, from currently available 3D printing techniques to the possibility of precisely mimicking the different BM niches [78].

Relatively few studies have been done using hematopoietic stem cells in these methods. Decellularization of cartilaginous tissues has been tested for ex vivo culturing of hematopoietic cells [79]. A co-culture model of hematopoietic stem/progenitor cell (HSPC) spheroids using polydimethylsiloxane (PDMS) has shown that the effectiveness of three-dimensional culture for HSPC expansion for clinical use is still a strategy that needs further improvement [80].

A three-dimensional collagen-based culture using HSPC, bone-marrow-derived MSCs or the umbilical cord (UC) to mimic the main compartments of the bone marrow hematopoietic niche has been proposed [81]. Data analysis generated the following compartments: (I) HSPC in suspension above collagen and (II) migratory HSPC in collagen fiber matrix. The different sites were representative of the distinct microenvironments that make up the bone marrow and have a significant impact on the fate of HSPC. The authors suggested that this 3D culture system using collagen and BM-MSC allowed HSPC expansion and provided a potential platform for advanced study of niches and hematopoiesis and their regulatory mechanisms.

A bone marrow composed of two compartments, solid and liquid, that act harmoniously, has been proposed [82]. A bone marrow structure was created using a macroporous PEG hydrogel that resembled the macroporous 3D architecture of the trabecular bone, site of the red bone marrow, and, therefore, where the HSPC niches are located. This bone marrow analog was found to be suitable for HSPC culture and for enhancing HSPC expansion compared to conventional 2D cell culture. The developed model of a perfused 3D bone marrow analogue mimicked the HSCs niche under steady-state or activated cell conditions that favor the maintenance or differentiation, respectively, of HSCs and allowed drug testing [82] It was concluded that the system reflected the behavior of HSPC in the niche under physiological conditions.

Some studies have shown that the interaction of bone marrow stromal cells with leukemic cells increases the resistance of these leukemic cells. Numerous scaffolds have been created to provide a minimal structure based on the 3D leukemic microenvironment [73]. As 3D culture could result in resistance to drugs, it can be a good screening tool for drug evaluation prior to the administration of chemotherapic drugs [74,83]. A classic example of scaffolds used to recreate the leukemic niche is polycaprolactone, which is an aliphatic and biodegradable polyester with good biocompatibility [84,85].

Recently, a scaffold of degradable zwitterionic hydrogel was tested for human HSPC expansion [86]. A 73-fold increase of long-term hematopoietic stem cells was observed (LT-HSC). The viscoelasticity and smoothness of the highly hydrated zwitterionic hydrogels seems to be important for the creation of cell niches, by their unique mechanical and antifouling properties.

The organs-on-chips technique was developed to study the mechanical and physiological response of a tissue, combining concepts of tissue engineering and microfluidics [87]. A bioelectronic device based on a conductive polymer scaffold was integrated with an electrochemical transistor configuration that allows 3D cell growth and the real-time monitoring of cell adhesion and growth.

This technique consists of manipulating small amounts (10−<sup>9</sup> to 10−<sup>18</sup> L) of fluids using small openings with micrometer dimensions. This methodology has been widely reported because it offers several advantages, such as using small amounts of samples and reagents, performing separations and detections with high resolution, and a short time for analysis [61]. Such techniques allow the precise control of fluids and particles in a given cell culture. This culture method enables the control of such nanoliter-scale fluids as described above and further enables and facilitates the simultaneous manipulation of cultured cells from a single cell [62]. Through precise manipulation of the components of the microfluidic culture medium, it is possible to transport nutrients, hormones, and oxygen growth factors to facilitate homeostasis and recreate mechanical signals that are absent in traditional culture [88].

This cell culture system offers several advantages for basic research applications, including the precision of micromanufacturing, which allows the presentation of a controllable and reproducible microenvironment [64]. Another advantage of this model is that it provides complete control over the conditions of the cell culture, including dynamic cell control, nutrient addition, removal of metabolites, stimulation with drugs and proteins, and simultaneous image and chip format [62,89,90].

The organs-on-chips technique was described in cancer studies (cancer-on-chip), where it can replicate the microenvironment to achieve robust and reliable results [91,92]. However, up to now, it has been difficult to culture HSCs in an organs-on-chips model. A device with a central cavity that successfully mimics the bone and bone marrow has been described. This device is made from polydimethylsiloxane (PDMS) with bone inducers inserted. The central chamber of this device is composed of porous PDMS membranes, and cytokines are added in the microfluids [93]. Although several methods have been used, results using HSCs are generally poor [94].

#### **5. Final Remarks**

Bone marrow HSCs are the stem cell most used in medical practice. Finding effective ways to mimic the bone marrow HSCs and microenvironment, the hematopoietic niche, in vitro is a challenge. In addition to HSCs, several other bone marrow cell types, including megakaryocytes, macrophages, monocytes and endothelial cells, directly or indirectly regulate HSCs and niche function. Two-dimensional cell cultures have been widely used due to their low maintenance cost and easy learning. However, 2D culture has several limitations. As a result, 3D cell culture techniques have been developed. Although the current cost is relatively high, the fact that a 3D culture has characteristics much closer to the tissue being studied is a great advantage. There are several methods currently employed or in development, such as spheroids, organoids, 3D bioprinting and organs-on-chip, and these are facing the challenge of culturing HSCs while maintaining their properties. This achievement will allow ex vivo HSC expansion, bringing new medical perspectives for HSCs transplantation, drug testing, and personalized treatment.

**Author Contributions:** Conceptualization, A.C.R.-F.; D.L. and S.P.B.; writing—original draft preparation, A.C.R.-F.; D.L.; J.L.M.R. and M.d.C.M.; writing—review and editing, A.C.R.-F.; D.L.; J.L.M.R.; M.d.C.M. and S.P.B.; supervision, S.P.B.; funding acquisition, S.P.B.

**Funding:** This research was funded by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Instituto Nacional de Ciência e Tecnologia–Fluidos Complexos (INCT-FCx), Instituto Nacional de Ciência e Tecnologia—Medicina Regenerativa (INCT-Regenera), all from Brazil.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Cells* Editorial Office E-mail: cells@mdpi.com www.mdpi.com/journal/cells

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18