Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Mar 19.
Published in final edited form as: Clin Exp Metastasis. 2016 Feb 12;33(4):325–337. doi: 10.1007/s10585-016-9781-2

Spheroid culture of LuCaP 136 patient-derived xenograft enables versatile preclinical models of prostate cancer

Maija P Valta 1,2, Hongjuan Zhao 1, Matthias Saar 1,3, Johanna Tuomela 4, Rosalie Nolley 1, Johannes Linxweiler 3, Jouko Sandholm 5, Jaakko Lehtimäki 4, Pirkko Härkönen 4, Ilsa Coleman 6, Peter S Nelson 6,7, Eva Corey 7, Donna M Peehl 1
PMCID: PMC12997442  NIHMSID: NIHMS1999016  PMID: 26873136

Abstract

LuCaP serially transplantable patient-derived xenografts (PDXs) are valuable preclinical models of locally advanced or metastatic prostate cancer. Using spheroid culture methodology, we recently established cell lines from several LuCaP PDXs. Here, we characterized in depth the features of xenografts derived from LuCaP 136 spheroid cultures and found faithful retention of the phenotype of the original PDX. In vitro culture enabled luciferase transfection into LuCaP 136 spheroids, facilitating in vivo imaging. We showed that LuCaP 136 spheroids formed intratibial, orthotopic, and subcutaneous tumors when re-introduced into mice. Intratibial tumors responded to castration and were highly osteosclerotic. LuCaP 136 is a realistic in vitro–in vivo preclinical model of a subtype of bone metastatic prostate cancer.

Keywords: Prostate cancer, Bone metastasis, Androgen-dependence, Osteosclerosis

Introduction

More than 90 % of patients with metastatic castration-resistant prostate cancer (PCa) have bone metastases [1]. These lesions are typically osteosclerotic, inducing new woven bone that has reduced mechanical strength, but osteolytic components are also commonly involved. Patients suffer from painful bone lesions, fractures and hypercalcemia [2], and death from PCa is often linked to bone metastases [3]. Current bone-targeted therapies reduce bone pain and delay skeletal events, but only one of these targeted treatments, radium-223, has been shown to increase survival of men with bone metastatic castration-resistant PCa [4]. Therefore current guidelines recommend radium-223 for patients with bone metastases and no visceral metastases [5]. Recent therapeutic advances including abiraterone, enzalutamide, docetaxel and sipuleucel-T have been shown to prolong survival of men with castration-resistant metastatic PCa and many of these men also suffer from bone metastases [6]. Despite extensive research efforts, bone metastasis remains a lethal consequence of advanced PCa.

The high failure rate for anti-cancer drug development is attributed in part to use of preclinical models that do not faithfully represent the human situation [7]. In particular, data obtained from cultured cancer cell lines often poorly predict human responses in clinical trials [7, 8]. Three classic cell lines, PC-3, DU 145 and LNCaP, are strongly over-represented in studies of PCa metastasis. These cells do not express wild-type androgen receptor (AR) [9] and produce mainly osteolytic metastases [10–12]. Models that accurately depict osteosclerotic bone metastases and capture the diverse phenotypes of advanced PCa are needed.

Among currently available preclinical models, the heterogeneity of PCa is perhaps best encompassed by LuCaP serially transplantable patient-derived xenografts (PDXs). Derived at the University of Washington from metastatic tissues obtained from a “rapid autopsy” program or from surgical specimens, these ~30 PDX lines include many of the molecular and cellular subtypes of PCa [13]. However, as a result of the expensive and labor-intensive nature of xenograft techniques, LuCaP PDXs have not reached their full potential due to historic inability to establish in vitro cell lines from these models. Recently, we successfully established, passaged, and cryopreserved spheroid cell cultures from several LuCaP PDXs [14]. Among them was LuCaP 147, a model of a subtype of PCa with a mutated AR, an SPOP mutation (recently recognized as a defining characteristic of a new molecular subtype of PCa [15]), and a hypermutator phenotype. LuCaP 147 cell cultures retained the characteristics of the original xenograft, including immunophenotype, genetic fidelity, gene expression profile and responsiveness to androgen [16]. Furthermore, cells formed tumors when re-introduced into mice, facilitating use of LuCaP 147 as an in vitro–in vivo preclinical model.

Here, we characterized in detail xenografts derived from the LuCaP 136 spheroid cell line. Initially derived from ascites, the original LuCaP 136 xenograft is characterized by expression of wild-type AR, loss of both PTEN alleles, and absence of the TMPRSS2:ERG fusion gene. Most importantly, the original PDX is osteosclerotic when implanted into tibiae of mice (unpublished data), thus representing the most common phenotype of PCa metastases. In the current study, we showed that LuCaP 136 spheroids formed tumors when implanted intratibially, orthotopically, or subcutaneously into immunocompromised mice. The geno- and immuno-phenotypes of the original xenograft, its derived cell culture, and the tumors formed by implanting the cells back into mice were similar. Lentiviral infection with the luciferase gene enabled bioluminescence imaging of implanted cells. Intratibial tumors were osteosclerotic like the original PDX, and LuCaP 136 cells metastasized from the bone to the lung. Tumor growth in bone was sensitive to castration, exhibiting the androgen-dependent phenotype of the original PDX. Altogether, we present a versatile preclinical model of a clinically relevant subtype of PCa that will facilitate drug discovery to treat metastatic PCa.

Materials and methods

Ethics statement

All animal studies were approved either by the Stanford Administrative Panel on Laboratory Animal Care, the local governmental animal care committee of the University of Saarland, or the State Provincial Offices of Finland (ESAVI 3257/04.10.07/2014) in accordance with the Directives 2012/707/EU, 2014/11/EU and of the European Parliament and of the Council for the Care and Use of Laboratory Animals.

Establishment, serial passage and cryopreservation of LuCaP 136 cell cultures

The establishment, propagation, freezing and thawing of LuCaP 136 spheroid cell cultures were previously described [14]. Cells were cultured in StemPro® hESC SFM (Invitrogen, Carlsblad, CA) supplemented with 10 nM of R1881 and 2 μM of Y-27632. The LuCaP 136 cell line was tested by short-tandem repeat analysis and proved to be unique and of human male origin [14]. Tests for mycoplasma and murine viruses were negative. Spheroid cultures between passages 3 and 13 were used in this study.

Lentiviral infection of luciferase into LuCaP 136 spheroids

LuCaP 136 cells were stably transfected with a pLenti CMV V5-LUC Blast vector (w567–1, Addgene, Cambridge, MA) to generate the LuCaP 136-luc cell line. Luciferase-expressing cells were selected in medium containing 1.25–1.5 μg/ml blasticidin, and luciferase expression was confirmed with the Dual-Luciferase Reporter Assay System (Promega, Madison, WI).

Tumor-forming capability of cultured LuCaP 136 cells and in vivo imaging

For subcutaneous implantation, intact LuCaP 136 spheroids were collected, suspended in cold Matrigel diluted 1:3 in HEPES-buffered saline (HBS), and injected subcutaneously (100 μl per site) with an estimated 2 × 106 cells into 8-week-old male SCID or nude mice (Charles River Laboratories, Germany). A 25-mg testosterone (Te) pellet with a release rate of 0.2 mg/day was inserted into a small incision made under the skin between the shoulder blades to bring androgen levels to those typical of adult human males [17].

For orthotopic implantation, 5 × 105 LuCaP 136 cells were injected into the left dorsal prostate lobe of each of 6 male SCID mice (10 μl per site). Tumor growth was monitored by ultrasound and magnetic resonance imaging (MRI) as previously described [18].

For intratibial experiments, intact LuCaP 136 or LuCaP 136-luc spheroids were collected and suspended in cold HBS on ice. Twelve-week-old male RAG2−/−γC−/− mice (bred in-house) or inbred Balb-c nude mice (Charles River Laboratories) were used. Te pellets were implanted as described above. Twenty μl of HBS containing 2.5–5X105 cells were injected into right tibiae as previously described [11]. To verify the correct position of the needle inside the tibial cavity, the tibia and the needle were X-rayed with Faxitron MX-20 DC-2 X-ray cabinet (30 kVp, 6 s, Faxitron X-ray Corp, Wheeling, IL). If the position was correct, the needle was removed, and another needle was inserted and 20 μl of cell suspension were slowly inoculated into the tibia.

To detect LuCaP 136-luc cell growth, non-invasive whole body bioluminescence imaging was applied (IVIS Spectrum, Perkin Elmer, Waltham, MA). Animals were injected intraperitoneally with 150 mg/kg D-Luciferin (Biosynth International, Inc., Itasca, IL) and anesthetized with 1–3 % isoflurane (Aerrane, Baxter, IL). Photon emission was measured over an integration time of 1–5 min. The mice were also X-rayed weekly for signs of tumor burden. The areas of osteosclerotic lesions revealed by Massońs trichrome staining, performed using standard methods, were quantified using Fiji-ImageJ ver. 1.50a [19]. Briefly, Masson’s trichrome colors were separated using the Brilliant Blue option in the Color Deconvolution function of Fiji-ImageJ. The bone channel (Channel 2) image was transformed into grayscale and then into a binary image. Then, a distance of 3.6 cm was measured from the growth plates of the tibiae. For analysis, a region-of-interest was manually drawn following the inner side of trabecular bone and the growth plate to the point of 3.6 cm, and bone share was measured. All images were evaluated blindly. In addition to this method the sclerotic changes were verified by hematoxylin and eosin (H&E) staining, X-ray, and micro-computed tomography (μCT). μCT was performed in vivo on a Siemens ImTek Micro-CAT II microtomograph (Siemens, Munich, Germany) at Stanford Center for Innovation in In-Vivo Imaging (voxel size 80 μm, X-ray voltage 80.0 kVp). Image analysis and 3D representations were conducted with GEHC Micro View Version MicroView Analysis +2.2 (GE Healthcare, Wauwatosa, WI).

Serum marker analysis was performed using commercial kits according to manufacturers’ instructions for human prostate-specific antigen (PSA) (R&D Systems, Minneapolis, MN) and Rat/Mouse PINP (IDS, Tyne & Wear, UK).

Androgen deprivation was accomplished by surgically excising the implanted Te pellet and/or surgical castration. Sham-operated mice served as intact, androgen-replete controls.

Immunohistochemistry (IHC)

Subcutaneous or orthotopic xenografts and organs were excised, fixed in 10 % buffered formalin overnight, paraffin-embedded and sectioned at 5-μm. Intratibial tumors were fixed in 10 % formalin for 2 days and incubated in 14 % EDTA for 15 days prior to paraffin-embedding and sectioning. IHC was performed as previously described [14] using antibodies listed in Table 1.

Table 1.

Antibodies used in IHC

Antibody Dilution Source
Keratin 5 1:50 Imgenex (San Diego, CA)
p63 1:50 Abcam (Cambridge, MA)
CD44 1:100 Abcam
Keratin 18 1:100 Santa Cruz Biotechnology (Santa Cruz, CA)
AR 1:100 BD Biosciences (Bedford, MA)
PCNA 1:100 Santa Cruz Biotechnology
PSA 1:100 Santa Cruz Biotechnology
PTEN 1:100 Santa Cruz Biotechnology
Ku70 1:200 Abcam
p63 and AMACAR (PIN Cocktail) Prediluted Biocaire (Concord, CA)
ERG 1:100 Epitomics (Abcam)
Osteocalcin 1:200 Millipore (Temecula, CA)
Ki67 1:200 Dako Cytomation (Hamburg, Germany)
Open in a new tab

Gene sequencing and gene expression profiling

For gene sequencing, genomic DNA was isolated from LuCaP 136 spheroid cultures at passage 2 using an AllPrep DNA/RNA Mini Kit (Qiagen, Germantown, MD). Exon 8 of p53 was selected for polymerase chain reaction (PCR) amplification and direct sequencing based on previous findings [20]. PCR primers for p53 were Fwd: 5′-CTTTGAGGTGCGTGTTTGTG-3′; Rev: 5′-GAAAACGGCATTTTGAGTGTT-3′. For gene expression profiling, total RNA were extracted from spheroids and xenografts preserved in Allprotect tissue reagent using an AllPrep DNA/RNA/Protein Mini Kit (Qiagen) according to the manufacturer’s directions. Probe labeling and hybridization were performed following the protocols suggested by Agilent and fluorescent array images were collected using the Agilent DNA microarray scanner G2505C. Control probes were removed, duplicate probes averaged, and spots flagged by Agilent Feature Extraction software as being foreground feature nonuniformity or population outliers were assigned a value of “NA”. Data were loess-normalized within arrays (normexp background correction with offset 50) and quantile-normalized between arrays in R using the Limma Bioconductor package [21]. Variations due to day of hybridization were removed by application of the ComBat function within the sva R package [22] to the log2 Cy3 signal intensities.

Statistical analysis

One- or two-way Anova (Graph Pad Prism 6.0C) was used. A p < 0.05 was considered significant. Pearson correlation analysis was performed to determine the correlations between transcriptomes of spheroids, derived subcutaneous and intratibial xenografts, and original LuCaP 136 subcutaneous PDXs using the cor.test function in R. A t test was used to establish whether each correlation coefficient was significantly different from zero. Scatter plots were generated using the log2 transformed expression levels of 16,791 genes.

Results

Tumor formation and phenotype after intratibial implantation of LuCaP 136 spheroids

LuCaP 136 spheroids were implanted into the tibiae of 5 male RAG2−/−γC−/− mice implanted with Te pellets. After 7 weeks, mice were sacrificed and tibiae were collected and processed. Engraftment was visualized in H&E-stained sections of all implanted sites and human origin was confirmed by Ku70 IHC staining (Fig. 1a). LuCaP 136 cells growing in bone were highly proliferative, as shown by IHC staining for PCNA, and the tumor cells expressed AR, AMACR, K18 and CD44 (Fig. 1a). PSA staining was low to moderate, and K5 staining was patchy (Fig. 1a). Cells were negative for p63, ERG and PTEN (Fig. 1a). This pattern of IHC staining of intratibial LuCaP 136 tumors was similar to that of the LuCaP 136 PDX from which spheroid cultures were originally derived (Fig. 1b). Sclerosis in bones bearing LuCaP 136 tumors was visualized by microscopy (osteocalcin IHC, Fig. 1a) and μCT (Fig. 1c).

Fig. 1.

Fig. 1

Open in a new tab

Histological and μCT features of LuCaP 136 cell-induced intratibial tumors. a Photomicrographs (×200) of intratibial LuCaP 136 tumors showing H&E-staining and IHC for various markers: Ku70 (human-specific antigen), PCNA (proliferating cell nuclear antigen), AR, AMACR (alpha-methylacyl-CoA racemase), K18 (keratin 18), CD44, PSA, K5 (keratin 5), p63, ERG, and PTEN. Tumor-induced bony spikes (arrows) outside the cortical bone were surrounded by osteocalcin-stained osteoblasts. Primary antibody was omitted in the negative control. b IHC staining of original LuCaP 136 xenograft (9200). Positive staining was observed for Ku70, AMACR, AR, K18, PSA, and PCNA. K5 was faint. c μCT image reconstructions of LuCaP 136-implanted tibia showing osteosclerotic lesions (tumor) and non-injected contralateral tibia (control)

Similar results were found in a second experiment in which an additional 5 mice were implanted, with engraftment and sclerotic growth occurring in all 5 animals. Normal organs were collected from the mice bearing intratibial LuCaP 136 tumors, and metastases were observed in the lungs of 4 of 5 (80 %) of these animals (Fig. 2a). IHC staining of Ku70 confirmed the human origin of these cells (Fig. 2a). The lung metastases also expressed K18 and AR (Fig. 2a), as well as low levels of PSA (Fig. 2a). Tumor invading and growing outside of the bone was also collected and expression of Ku70, K18, AR and PSA was confirmed (Fig. 2b).

Fig. 2.

Fig. 2

Open in a new tab

Metastatic cells to lung or tumor invading outside of bone marrow cavity. a IHC of human-specific Ku70 shows LuCap 136 tumor metastasis to lung and expression of K18, AR and PSA (×200). b IHC of Ku70, K18, AR and PSA of LuCaP 136 tumor growing outside of bone (×200)

In order to facilitate monitoring of tumor growth in bone, LuCaP 136 spheroids were infected with lentivirus carrying the luciferase gene and LuCaP 136-luc cells were selected for blasticidine resistance. Bioluminescence imaging of mice implanted with 5 × 105 LuCaP 136-luc cells in bone showed tumor formation (Fig. 3a), with a steady increase in signal from day 0 to day 18 (Fig. 3b, n = 9 mice).

Fig. 3.

Fig. 3

Open in a new tab

Bioluminescence imaging of intratibial LuCaP 136-luc tumors. a Luminescence data of LuCaP 136-luc intratibial tumors in individual nude mice overlaid on a photographic image. b Growth of LuCaP 136-luc intratibial tumors (n = 9) measured by average luminescence units (Radiance, p/sec/cm2/sr) over time. p, photons, sr, surface radiance. Data points represent mean ± SD

Response of intratibial LuCaP 136-luc tumors to androgen deprivation

LuCaP 136-luc spheroids were implanted in the tibiae of 20 male nude mice supplemented with Te pellets. Sixteen days after implantation, mice were randomized into 3 groups of 6–8 mice each. In one group, the mice were sham-operated. In another group, the Te pellets were removed. In the third group, Te pellets were removed and mice were castrated. Tumor volume over a period of 8 weeks was monitored by bioluminescence imaging. As shown in Fig. 4, castration in conjunction with excision of the Te pellet clearly suppressed intratibial growth of LuCaP 136-luc cells. Tumor volume was considerably decreased by 2 weeks after surgery, and remained low for another 4 weeks. In contrast, removal of Te pellets in the absence of castration did not alter tumor growth compared to growth in sham-operated controls (Fig. 4). Serum human PSA levels in all groups were low (beneath 0.3 ng/ml) at the end of the experiment (data not shown). This corresponds to our previous findings that LuCaP 136 tumors express low levels of PSA by IHC and that LuCaP 136 cells do not elevate PSA in response to androgen in vitro [14].

Fig. 4.

Fig. 4

Open in a new tab

LuCaP 136-luc growth is androgen-dependent as measured by bioluminescence. Growth of LuCaP 136-luc intratibial tumors in nude mice (n = 20) measured by average luminescence units (Radiance, p/sec/cm2/sr) over time. p, photons, sr, surface radiance. Data points represent mean ± SD. On day 0, a testosterone pellet was inserted and 5 × 105 LuCaP 136-luc cells were inoculated into right tibiae. On day 16 after cancer cell inoculation, mice were randomized into 3 groups. One group was castrated and the testosterone pellet was excised. The second group had testosterone pellets excised only without castration. The third group was sham-operated, and testosterone pellets were left intact. *p < 0.05, two-way ANOVA (luminescence signal comparison between sham-operated and castrated mice)

LuCaP 136 induces strong osteosclerosis in vivo in an androgen-dependent manner

At sacrifice, LuCaP 136-luc tumor-bearing tibiae in sham-operated mice carrying Te pellets and mice with Te pellet excision exhibited increased osteosclerosis compared with uninjected control tibiae as shown by x ray images and histology (Fig. 5a). Histomorphometric analysis of trabecular bone formation in tibiae stained with Massońs trichrome showed a statistically significant increase in bone formation in sham-operated mice carrying Te pellets and mice with Te pellet excision when tumor-bearing tibia was compared with the normal uninjected tibia of the same animal (Fig. 5b). There was no statistically significant increase in bone formation in the castrated mice (that had Te pellets excised) between cancer cell-injected and uninjected tibiae (Fig. 5a,b), which confirms that fast growing tumors in androgen-replete animals were associated with more evident osteosclerosis compared with tumors in androgen-depleted mice. Measurement for serum N-terminal propeptide of procollagen type 1a (PINP), a suggested bone turnover marker for PCa bone metastasis [23], showed a deeply decreasing trend during the experiment from average 359–36 ng/ml, but there were no statistically significant differences between the groups (data not shown). This decrease is likely to represent normal mouse physiology with younger mice having higher bone turnover compared with older mice [24], but no correlation with bone formation was found in this study.

Fig. 5.

Fig. 5

Open in a new tab

LuCaP 136-luc induced strong androgen-dependent osteosclerosis in vivo. a Representative images of small animal X-rays and Massońs trichrome stainings at the termination of the experiment showing strong osteosclerosis (white arrows in X-rays and black arrows in Masson’s trichrome staining) in LuCaP 136-luc tumor-bearing tibiae in sham operated mice and mice with Te pellet excision. Less sclerosis was found in castrated mice with Te pellet excision, when compared to contralateral uninjected tibiae. White bar (X-rays) and black bar (Masson’s trichrome staining) are 500 μm. b Histomorphometric analysis of relative amount (%) of trabecular bone area in Massońs trichrome stainings of LuCaP 136-luc-injected tibiae at the termination of the experiment. Tumor-bearing tibiae were compared with normal uninjected tibiae of the same animal. *p < 0.05, ***p < 0,001, one-way ANOVA. All mice had been inoculated with a testosterone pellet on day 0 of the experiment. “Sham-operated, Te pellet”, mice were bearing a testosterone pellet and sham-operated on day 16 of the experiment. “Te pellet excised”, testosterone pellet was excised on day 16. “Castrated, Te pellet excised”, mice were castrated and testosterone pellets were excised on day 16

Tumor formation and phenotype after orthotopic implantation of LuCaP 136 spheroids

LuCaP 136 spheroids were mixed with Matrigel and injected into the prostates of male SCID mice. The first ultrasound monitoring was performed 3 weeks after inoculation and was thereafter repeated at 5-, 6- and 7-weeks. In all 6 tumor-bearing animals (take rate 100 %), local tumor growth could be visualized by ultrasound. No pathologic symptoms such as weight loss of more than 20 % or urinary retention were observed. Ultrasound imaging allowed the tumors and the normal prostate tissue within the different lobes to be distinguished. Figure 6a illustrates the increase in tumor volume over time as monitored by ultrasound imaging. From the first ultrasound measurement at 3 weeks to the final measurement at 7 weeks, tumor volume increased ~5-fold. Histology and immunophenotype of the orthotopic LuCaP 136 tumors are shown in Fig. 6b. H&E-staining revealed solid tumor formation in the different lobes with invasion of the normal mouse prostate at the tumor margins, as seen by ultrasound. IHC staining of Ku70 delineated the human LuCaP 136 cells in the mouse prostate, with positive staining for AR, PSA, AMACR, and K18. Rare K5-positive cells were observed. Tumors were negative for ERG, as expected since the LuCaP 136 xenograft does not have the TMPRSS2:ERG gene fusion, and positive for Ki67, confirming high proliferation of the xenografts. To investigate whether orthotopic xenografts gave rise to metastases to the bone, tibiae were excised from mice bearing tumors and X-rayed prior to decalcification, but no metastatic lesions were found by this method.

Fig. 6.

Fig. 6

Open in a new tab

Orthotopic LuCaP 136 tumors from implanted spheroid cultures. a Tumor volumes at weeks 3, 5, 6 and 7 as assessed by ultrasound imaging after inoculation of 5 × 105 cells into the left dosal prostate lobe. A representative MRI image shows the orthotopic tumor 6 weeks after inoculation of spheroids (red dotted line). b H&E-staining and IHC of solid tumor masses within the prostate showed positive staining for AMACR, K18, AR, and PSA. Rare K5-positive cells were present. Staining for p63 and ERG was negative (positive ERG staining is present in endothelia). Proliferation and human identity were shown by Ki67 and Ku70 staining, respectively. Primary antibody was omitted in the negative control. Red staining was developed with the substrate alkaline phosphatase, and brown staining was developed with 3,3′-diaminobenzidine (X100). (Color figure online)

Tumor formation after subcutaneous implantation of LuCaP 136 spheroids

LuCaP 136 or LuCaP 136-luc spheroids were mixed with Matrigel and ~2 × 106 cells per site were implanted subcutaneously into male SCID or nude mice. Tumors formed in all mice (n = 5) and histology (Fig. 7a) was similar to that of the original xenograft (Fig. 1b), which is maintained by serial transplantation at the subcutaneous site. Doubling time was ~2 weeks, as determined by caliper measurements of tumors growing in two mice (Fig. 7b).

Fig. 7.

Fig. 7

Open in a new tab

Subcutaneous tumors from implanted LuCaP 136 spheroids. a H&E-staining of subcutaneous LuCaP 136 tumor excised 8 weeks after implantation of LuCaP 136 spheroids into mice (×200). b Volumes of two subcutaneous LuCaP 136 tumors over time as determined by caliper measurements [V = (W(2) × L)/2]

Fidelity of gene expression and P53 mutation in LuCaP 136 spheroids and xenografts

The original LuCaP 136 PDX has a STOP-gained nonsense mutation in p53 (R280 position) [20]. DNA sequencing revealed the same mutation in spheroid cultures (Fig. 8), showing that cultures retained the p53 mutant phenotype of the original PDX.

Fig. 8.

Fig. 8

Open in a new tab

The same p53 mutation as in the original LuCaP 136 xenograft was present in spheroid cultures of LuCaP 136

To determine whether in vitro culture changed the global transcription profile of LuCaP 136, we compared the transcriptome of spheroids to that of the original subcutaneous PDX from which the spheroids were derived. Pearson correlation analysis showed a correlation coefficient of 0.93 between expression levels of 16,791 genes in spheroids that had been passaged 11 times and those in the original subcutaneous PDX (Fig. 9a). In addition, the global gene expression profile of spheroids showed a close correlation with that of its derived subcutaneous tumors (correlation coefficient 0.94) (Fig. 9b). Notably, the global gene expression profile of spheroids showed an even higher correlation with that of its derived intratibial tumors (correlation coefficient 0.97) (Fig. 9c). Finally, the highest correlation (correlation coefficient 0.98) was observed between subcutaneous tumors derived from spheroids and the original subcutaneous PDX (Fig. 9d). These results demonstrated that spheroids retained the gene expression profile of the original PDX with high fidelity, and that reimplantation of the spheroids created subcutaneous tumors that were highly similar to the original PDX. The particularly high correlation of gene expression in spheroids and intratibial tumors is of interest, implying that LuCaP 136 spheroids provide a representative in vitro model of bone metastases.

Fig. 9.

Fig. 9

Open in a new tab

Pearson correlation analysis of transcriptomes of spheroids (n = 2), spheroid-derived subcutaneous (n = 2) or intratibial (n = 2) xenografts, and original LuCaP 136 subcutaneous PDXs (n = 3). a Spheroids versus original PDXs. b Spheroids versus subcutaneous tumors derived from spheroids. c Spheroids versus intratibial tumors derived from spheroids. d Subcutaneous tumors derived from spheroids versus original PDXs. Pearson’s correlation coefficient r, and p value for each comparison is shown on plots

Discussion

The LuCaP 136 serially transplantable PDX has several features that make it a representative and realistic model of a subtype of PCa. Like most metastatic prostate cancers, LuCaP 136 cells retain expression and function of AR and express AR target genes such as PSA, although expression is relatively low. LuCaP 136 cells also exhibit loss of PTEN, the most frequently deleted tumor suppressor gene in PCa [25]. The LuCaP 136 PDX also represents prostate cancers that have not undergone a fusion of the TMPRSS2 and ERG genes, which occurs in ~50 % of PCa [26].

Apparently resistant to monolayer culture, LuCaP 136 cells can be isolated from PDXs and grown and passaged in vitro as spheroids if they are never digested to single cells [14]. In our initial characterization of LuCaP 136 spheroids, we showed that the cultured cells were unique and of human origin, and retained the immunophenotype of the original PDX [14]. Cells mostly lack expression of p63 and K5, but express CD44, EpCAM, AR and K18.

In the current study, we tested the ability of LuCaP 136 spheroids to re-establish xenografts after implantation at intratibial, orthotopic, and subcutaneous sites. Regardless of site, we observed a 100 % take rate, with progressive increase in tumor volume over time. Histology of the spheroid-derived tumors was similar to that of the original PDX, as was the immunophenotype. Xenografts expressed AR, K18, PSA, and CD44. The antibody cocktail against p63 and AMACR, used clinically to identify PCa [27], appropriately stained xenografts (p63-negative and AMACR-positive). Since the subcutaneous site is desirable for its ease of monitoring tumor growth, the ability to re-establish subcutaneous LuCaP 136 xenografts from cell cultures will enable widespread use of this important preclinical model. Establishment of xenografts at the orthotopic site also has some potential advantages, such as greater propensity for metastasis compared to subcutaneous tumors [28]. In this study, we did not observe bony metastases from orthotopic LuCaP 136 tumors, but studies of longer duration might permit their development.

A particular emphasis of this study was evaluating intratibial xenografts of LuCaP 136 cells. One of the most important features of the LuCaP 136 PDX is its osteosclerotic (bone forming, osteoblastic) phenotype. The majority of human PCa metastases to bone are osteosclerotic, with osteolytic components sometimes present as well [29]. Few PCa models exhibit a completely osteosclerotic phenotype. Of commonly used PCa cell lines, both PC-3 and DU 145 are osteolytic [30]. VCaP and LNCaP cell lines produce mixed osteosclerotic and osteolytic lesions [30, 31]. MDA PCa 2B [32] and LNCaP-19 [33] are among the few osteosclerotic PCa cell lines. Serially transplanted, patient-derived PCa PDXs that are osteosclerotic include the AR-negative line MDA PCa 118b [34] and AR-positive MDA PCa 183 [35]. Six other LuCaP PDXs besides LuCaP 136 have an osteosclerotic phenotype, including LuCaP 23.1 [36].

Implantation of LuCaP 136 spheroids in the bone produced osteosclerotic lesions, recapitulating the phenotype of the original PDX. A benefit of culturing LuCaP 136 cells was the opportunity to introduce a luciferase gene by lentiviral infection, facilitating the monitoring of tumor growth in the bone by bioluminescence imaging. In conjunction with histomorphometric analysis of bone metastatic lesions, these methods provide powerful tools to quantitatively measure tumor burden in bone.

Notably, we observed that LuCaP 136 cells metastasized to the lungs from the intratibial tumor. This model therefore provides an opportunity to further study the mechanisms of metastasis from bone to visceral organs, the effects of microenvironment, and the efficacy of treatments on bone vs visceral metastases in the same animal. For example, in a similar scenario in which an intratibial osteosarcoma metastasized to lung, Brennecke et al. found that metastasis was prevented by anti-CXCR4 [37]. It has been observed that men with visceral PCa metastases (present in ~20 % of men participating in first-line clinical trials for metastatic castration-resistant PCa) have a poor prognosis, regardless of treatment strategy [38]. In particular, lung metastases were found in 46 % of men who died from PCa [38]. As discussed by Drake [38], an IHC study of visceral vs bone metastases showed more frequent expression of nuclear survivin in the former, and cytoplasmic survivin, BCL2 and MCL1 in the latter, suggesting different susceptibilities for apoptosis [39]. In another study, angiogenic profiles differed between metastases in soft tissues vs those in bone. Using the LuCaP 23.1 PDX, Morrissey et al. showed that blocking angiopoietin-2 inhibited growth of subcutaneous but not intratibial tumors [40]. Docetaxel similarly showed differential inhibition of subcutaneous vs intratibial LuCaP 23.1 PDXs in another study [41], underscoring the fact that drugs must be evaluated on tumors at different sites to mimic the human situation.

Since androgen deprivation is the standard of care for metastatic, hormone-naïve PCa, we aimed to demonstrate that LuCaP 136 tumors underwent typical regression when deprived of androgen. Indeed, intratibial LuCaP 136 xenografts regressed following castration of the male murine hosts and removal of the supplemental Te pellets and growth remained low for several weeks. Of note, removal of the Te pellets alone in the absence of castration did not diminish growth of the LuCaP 136 tumors. The original LuCaP PDXs were established and serially passaged in male mice that were not supplemented with testosterone (which brings circulating testosterone levels in mice to those equivalent of adult human males [42]). Therefore selection for growth in the presence of murine testosterone levels (equivalent to those in hypogonadal adult human males [42]) may occur. In comparison, castration of mice in conjunction with removal of pellets brings testosterone levels to those commensurate with abiraterone-treated human males [42]. As in humans, the response of LuCaP 136 tumors demonstrates the efficacy of further lowering androgen [6].

Other properties of LuCaP 136 cells make them an expedient model for future studies. The combination of PTEN loss and p53 mutation in LuCaP 136 cells is significant, given that aberrations in both tumor suppressor genes are common in advanced PCa and recent studies suggest that such cancers are vulnerable to PARP-PI3K inhibition [43]. Other new strategies to target PTEN-deficient PCa, such as inhibition of PIP5K1α [44], are also under development. The results of microarray profiling showed that cultured LuCaP 136 cells retained remarkable genetic fidelity to the PDX of origin. Gene expression of spheroids highly correlated with that of the original PDX, and the gene expression profile of subcutaneous tumors derived from spheroids was even more similar to that of the original (also subcutaneous) PDX, showing that spheroids recapitulated the original phenotype in vivo. Notably, gene expression of cultured spheroids was more similar to intratibial than subcutaneous tumors, suggesting that spheroids provide a representative model of bone metastases. Altogether, the LuCaP 136 in vitro–in vivo model provides a practicable system for testing experimental therapeutics, especially in the most clinically relevant setting of osteosclerotic metastases.

Acknowledgments

This work was supported by the Stanford Department of Urology [HZ, RN, DMP], funding for university level health research in Finland [MPV], a Ferdinand Eisenberger Grant of the German Society of Urology ID SaM1/FE-11 and HOMFOR (Saarland University, Germany) [MS], South-Western Cancer Foundation of Finland and K Albin Johansson Foundation, Finland [JT], Academy of Finland and Sigrid Jusélius Foundation, Finland [PH], The Richard Lucas Foundation and the PNW Prostate Cancer SPORE NIH P50 CA097186 [EC], NIH 2P50CA097186 [IC, PSN, EC], NIH 1R01CA165573 [IC, PSN, EC], NIH P01CA163227 [IC, PSN, EC], and the Prostate Cancer Foundation [EC, DMP]. Pharmatest Services LTD, Turku, Finland, is acknowledged for serum analyses and histological stainings.

Abbreviations

AR

Androgen receptor

H&E

Hematoxylin and eosin

HBS

HEPES-buffered saline

IHC

Immunohistochemistry

K5

Keratin 5

K18

Keratin 18

MRI

Magnetic resonance imaging

PCa

Prostate cancer

PDX

Patient-derived xenograft

PINP

N-terminal propeptide of procollagen type 1a

PSA

Prostate-specific antigen

Te

Testosterone

μCT

Micro-computed tomography

Footnotes

Compliance with ethical standards

Conflict of interest No potential conflicts of interest were disclosed.

References

  • 1.Weilbaecher KN, Guise TA, McCauley LK (2011) Cancer to bone: a fatal attraction. Nat Rev Cancer 11(6):411–425 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Groot MT et al. (2003) Costs of prostate cancer, metastatic to the bone, in the Netherlands. Eur Urol 43(3):226–232 [DOI] [PubMed] [Google Scholar]
  • 3.Lange PH, Vessella RL (1998) Mechanisms, hypotheses and questions regarding prostate cancer micrometastases to bone. Cancer Metastasis Rev 17(4):331–336 [DOI] [PubMed] [Google Scholar]
  • 4.Parker C et al. (2013) Alpha emitter radium-223 and survival in metastatic prostate cancer. N Engl J Med 369(3):213–223 [DOI] [PubMed] [Google Scholar]
  • 5.Mohler JL et al. (2014) Prostate cancer, version 2.2014. J Natl Compr Canc Netw 12(5):686–718 [DOI] [PubMed] [Google Scholar]
  • 6.Crawford ED et al. (2015) Treating patients with metastatic castration resistant prostate cancer: a comprehensive review of available therapies. J Urol 194(6):1537–1547 [DOI] [PubMed] [Google Scholar]
  • 7.Ocana A et al. (2011) Preclinical development of molecular-targeted agents for cancer. Nat Rev Clin Oncol 8(4):200–209 [DOI] [PubMed] [Google Scholar]
  • 8.Sausville EA, Burger AM (2006) Contributions of human tumor xenografts to anticancer drug development. Cancer Res 66(7):3351–3354 (discussion 4) [DOI] [PubMed] [Google Scholar]
  • 9.Sobel RE, Sadar MD (2005) Cell lines used in prostate cancer research: a compendium of old and new lines–part 1. J Urol 173(2):342–359 [DOI] [PubMed] [Google Scholar]
  • 10.Najy AJ et al. (2012) Cediranib inhibits both the intraosseous growth of PDGF D-positive prostate cancer cells and the associated bone reaction. Prostate 72(12):1328–1338 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Valta MP et al. (2008) FGF-8 is involved in bone metastasis of prostate cancer. Int J Cancer 123(1):22–31 [DOI] [PubMed] [Google Scholar]
  • 12.Bonfil RD et al. (2007) Prostate cancer-associated membrane type 1-matrix metalloproteinase: a pivotal role in bone response and intraosseous tumor growth. Am J Pathol 170(6):2100–2111 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.van Weerden WM, Bangma C, de Wit R (2009) Human xenograft models as useful tools to assess the potential of novel therapeutics in prostate cancer. Br J Cancer 100(1):13–18 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Young SR et al. (2013) Establishment and serial passage of cell cultures derived from LuCaP xenografts. Prostate 73(12):1251–1262 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Barbieri CE et al. (2012) Exome sequencing identifies recurrent SPOP, FOXA1 and MED12 mutations in prostate cancer. Nat Genet 44(6):685–689 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Saar M et al. (2014) Spheroid culture of LuCaP 147 as an authentic preclinical model of prostate cancer subtype with SPOP mutation and hypermutator phenotype. Cancer Lett 351(2):272–280 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Ibrahim L, Wright EA (1983) Effect of castration and testosterone propionate on mouse vibrissae. Br J Dermatol 108(3):321–326 [DOI] [PubMed] [Google Scholar]
  • 18.Saar M et al. (2012) Experimental orthotopic prostate tumor in nude mice: techniques for local cell inoculation and three-dimensional ultrasound monitoring. Urol Oncol 30(3):330–338 [DOI] [PubMed] [Google Scholar]
  • 19.Schneider CA, Rasband WS, Eliceiri KW (2012) NIH Image to ImageJ: 25 years of image analysis. Nat Methods 9(7):671–675 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Kumar A et al. (2011) Exome sequencing identifies a spectrum of mutation frequencies in advanced and lethal prostate cancers. Proc Natl Acad Sci USA 108(41):17087–17092 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Wettenhall JM, Smyth GK (2004) limmaGUI: a graphical user interface for linear modeling of microarray data. Bioinformatics 20(18):3705–3706 [DOI] [PubMed] [Google Scholar]
  • 22.Leek JT et al. (2012) The sva package for removing batch effects and other unwanted variation in high-throughput experiments. Bioinformatics 28(6):882–883 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Leeming DJ et al. (2011) Serum N-terminal propeptide of collagen type i is associated with the number of bone metastases in breast and prostate cancer and correlates to other bone related markers. Biomark Cancer 3:15–23 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Edwards JR, Mundy GR (2011) Advances in osteoclast biology: old findings and new insights from mouse models. Nat Rev Rheumatol 7(4):235–243 [DOI] [PubMed] [Google Scholar]
  • 25.Zhang X et al. (2015) Amplification and protein expression of androgen receptor gene in prostate cancer cells: fluorescence hybridization analysis. Oncol Lett 9(6):2617–2622 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Ayala G et al. (2015) Expression of ERG protein in prostate cancer: variability and biological correlates. Endocr Relat Cancer 22(3):277–287 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Adley BP, Yang XJ (2006) Application of alpha-methylacyl coenzyme A racemase immunohistochemistry in the diagnosis of prostate cancer: a review. Anal Quant Cytol Histol 28(1):1–13 [PubMed] [Google Scholar]
  • 28.Simmons JK et al. (2015) Animal models of bone metastasis. Vet Pathol 52(5):827–841 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Ganguly SS, Li X, Miranti CK (2014) The host microenvironment influences prostate cancer invasion, systemic spread, bone colonization, and osteoblastic metastasis. Front Oncol 4:364. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Nemeth JA et al. (1999) Severe combined immunodeficient-hu model of human prostate cancer metastasis to human bone. Cancer Res 59(8):1987–1993 [PubMed] [Google Scholar]
  • 31.Graham TJ et al. (2014) Preclinical evaluation of imaging biomarkers for prostate cancer bone metastasis and response to cabozantinib. J Natl Cancer Inst 106(4):dju033. [DOI] [PubMed] [Google Scholar]
  • 32.Yang J et al. (2001) Prostate cancer cells induce osteoblast differentiation through a Cbfa1-dependent pathway. Cancer Res 61(14):5652–5659 [PubMed] [Google Scholar]
  • 33.Hagberg Thulin M et al. (2014) Osteoblasts stimulate the osteogenic and metastatic progression of castration-resistant prostate cancer in a novel model for in vitro and in vivo studies. Clin Exp Metastasis 31(3):269–283 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Li ZG et al. (2008) Androgen receptor-negative human prostate cancer cells induce osteogenesis in mice through FGF9-mediated mechanisms. J Clin Investig 118(8):2697–2710 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Roychowdhury S et al. (2011) Personalized oncology through integrative high-throughput sequencing: a pilot study. Sci Transl Med 3(111):111ra21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Larson SR et al. (2013) Characterization of osteoblastic and osteolytic proteins in prostate cancer bone metastases. Prostate 73(9):932–940 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Brennecke P et al. (2014) CXCR4 antibody treatment suppresses metastatic spread to the lung of intratibial human osteosarcoma xenografts in mice. Clin Exp Metastasis 31(3):339–349 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Drake CG (2014) Visceral metastases and prostate cancer treatment: ‘die hard’, ‘tough neighborhoods’, or ‘evil humors’? Oncology (Williston Park) 28(11):974–980 [PMC free article] [PubMed] [Google Scholar]
  • 39.Akfirat C et al. (2013) Tumour cell survival mechanisms in lethal metastatic prostate cancer differ between bone and soft tissue metastases. J Pathol 230(3):291–297 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Morrissey C et al. (2010) Inhibition of angiopoietin-2 in LuCaP 23.1 prostate cancer tumors decreases tumor growth and viability. Prostate 70(16):1799–1808 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Brubaker KD et al. (2006) Administration of zoledronic acid enhances the effects of docetaxel on growth of prostate cancer in the bone environment. BMC Cancer 6:15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Michiel Sedelaar JP, Dalrymple SS, Isaacs JT (2013) Of mice and men–warning: intact versus castrated adult male mice as xenograft, hosts are equivalent to hypogonadal versus abiraterone treated aging human males, respectively. Prostate 73(12):1316–1325 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Gonzalez-Billalabeitia E et al. (2014) Vulnerabilities of PTEN-TP53-deficient prostate cancers to compound PARP-PI3K inhibition. Cancer Discov 4(8):896–904 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Semenas J et al. (2014) The role of PI3K/AKT-related PIP5K1alpha and the discovery of its selective inhibitor for treatment of advanced prostate cancer. Proc Natl Acad Sci USA 111(35):E3689–E3698 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES