A molecular complex of bovine milk protein and oleic acid selectively kills cancer cells in vitro and inhibits tumour growth in an orthotopic rat bladder tumour model

Authors

  • Zhengwen Xiao,

    1. Department of Surgery, University of Alberta, Edmonton, Alberta, Canada
    2. Department of Oncology, Faculty of Medicine, University of Alberta, Edmonton, Alberta, Canada
    Search for more papers by this author
  • Allan Mak,

    1. Department of Surgery, University of Alberta, Edmonton, Alberta, Canada
    2. Department of Oncology, Faculty of Medicine, University of Alberta, Edmonton, Alberta, Canada
    Search for more papers by this author
  • Karen Koch,

    1. NatImmune A/S, Copenhagen, Denmark
    Search for more papers by this author
  • Ronald B. Moore

    Corresponding author
    1. Department of Oncology, Faculty of Medicine, University of Alberta, Edmonton, Alberta, Canada
    • Department of Surgery, University of Alberta, Edmonton, Alberta, Canada
    Search for more papers by this author

Correspondence: Professor Ronald B. Moore, Department of Surgery, University of Alberta, 2D2 Walter Mackenzie Health Sciences Centre, 8440-112 Street, Edmonton, Alberta, Canada T6G 2B7.

e-mail: ron.moore@albertahealthservices.ca

Abstract

What's known on the subject? and What does the study add?

  • Novel intravesical therapies are needed for superficial bladder cancer that reduce the risk of infection associated with Bacillus Calmette–Guérin (BCG) and further destabilization of the urothelium associated with cytotoxic chemotherapy. Experimental therapies to date have included photodynamic therapy, oncolytic viruses, gene therapy (antisense oligonucleotides and silencing RNA), cytokine therapy, death receptor agonists (tumour-necrosis-factor-related apoptosis-inducing ligand and anti-DR5 monoclonal antibody), naturally occurring substances (curcumin and deguelin) and human α-lactalbumin made lethal to tumour cells (HAMLET). HAMLET, a natural occurring product in milk, induces apoptosis in urothelial cancer cells but has limitations in clinical application because of its human source. A previous study in patients with bladder cancer has demonstrated that intravesical HAMLET (daily for 5 days before tumour resection) caused selective apoptotic tumour cell death. BAMLET, the bovine equivalent of HAMLET, is a complex of bovine α-lactalbumin and oleic acid (bLAC) that has been shown in vitro to accumulate in the endolysosomal compartment of tumour cells and induce leakage of lysosomal cathepsins into the cytosol followed by activation of the pro-apoptotic protein Bax.
  • This is the first in vivo study to show that BAMLET (bLAC) induces apoptosis in urothelial cancer cells and controls the growth of high risk urothelial cancer in a syngeneic rat orthotopic model. This same bladder cancer model system has been used to test other novel therapies, including BCG, and therefore provides a relative comparison of its effectiveness with other intravesical therapies.

Objective

  • To investigate the efficacy of a complex of bovine α-lactalbumin and oleic acid (bLAC) to kill urothelial cancer cells in vitro and inhibit tumour growth and progression in a high risk bladder tumour model.

Materials and Methods

  • The cytotoxicity of bLAC to a large panel of urothelial cell cancer (UCC) cells was tested by the MTT (3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide) assay, using bLA, the folded α-lactalbumin without oleic acid, as a control.
  • The mechanism of bLAC-inducing cell death was evaluated by annexin V staining, TUNEL (terminal deoxynucleotidyl transferase mediated nick end labelling) assay and sub-G1 DNA analysis.
  • The selective bLAC cytotoxicity was examined using multicellular spheroids consisting of UCC and non-transformed fibroblasts.
  • Rats bearing orthotopic tumour received intravesical instillations (twice weekly, for 3 weeks) of bLAC, bLA, BCG or saline, starting 6 days after UCC (AY-27) cell inoculation. Animals were monitored for survival, toxicity and tumour growth control.

Results

  • A dose-dependent bLAC-inducing apoptotic-like cell death was shown in UCC cells tested, including cells refractory to classic apoptosis-inducing agents, whereas bLA showed little cytotoxicity.
  • bLAC selectively destroyed cancer cells in spheroids.
  • Intravesical bLAC therapy demonstrated marked reduction in tumour growth/progression and significantly prolonged animal survival vs saline instillations (P = 0.004, log-rank test) and showed comparable efficacy with BCG (standard) therapy.

Conclusion

  • The findings identify bLAC as a new candidate for UCC therapy and suggests that topical administration of bLAC alone or with BCG to prevent progression of bladder cancer warrants further exploration.
Abbreviations
NMIBC

non-muscle-invasive bladder cancer

HAMLET

human α-lactalbumin made lethal to tumour cells

bLAC

bovine α-lactalbumin complex with oleic acid

TRAIL

tumour-necrosis-factor-related apoptosis-inducing ligand

FBS

fetal bovine serum

MTT

3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide

TUNEL

terminal deoxynucleotidyl-transferase-mediated nick end labelling

LTS

long-term survivor

UCC

urothelial cell cancer

mTOR

mammalian target of rapamycin

Introduction

Bladder cancers are common genitourinary malignant tumours with 70%–80% of the cases initially diagnosed as superficial non-muscle-invasive bladder cancers (NMIBCs). Superficial bladder tumours are treated by transurethral resection of the tumours, followed by intravesical chemotherapy or immunotherapy to prevent tumour recurrence and progression [1]. BCG remains the most effective intravesical agent against recurrence and the standard of care for high risk (high grade, Tis) NMIBC [2]. However, intravesical BCG is associated with accumulative toxicity. Only 36% of patients were able to tolerate 2-year monthly maintenance instillations and only 16% of patients in the Southwest Oncology Group trial were able to complete the 3-year treatment schedule [3, 4]. Ideally, an antitumour agent should selectively kill cancer cells with minimum toxicity toward normal cells. This is particularly true for bladder cancer where there is an existing carcinogenic field effect [5]. Currently there are only a few such agents for pre-clinical or clinical trials. Among them is a human milk protein–lipid complex termed HAMLET (human α-lactalbumin made lethal to tumour cells) which is able to effectively kill cancer cells whilst sparing the non-transformed cells both in vitro and in vivo [6-10]. HAMLET has been suggested to have different intracellular trafficking in cancer cells than that of normal cells and to induce an apoptotic-like cytotoxicity which is independent of Bcl-2, p53 and caspases [8, 11]. Therefore HAMLET is able to kill tumour cells that have acquired resistance to classic pro-apoptotic agents.

HAMLET is a molecular complex of human α-lactalbumin and oleic acid. It is formed when the protein unfolds upon release of the tightly bound Ca2+ ion; the fatty acid then stabilizes the unfolded conformation [6, 7]. Interestingly, only the protein–lipid complex (not the native folded α-lactalbumin) showed cytotoxicity to cancer cells in vitro [7]. Amino acid sequences of human and bovine α-lactalbumins show 71% homology which allows the formation of bovine α-lactalbumin complex with oleic acid (bLAC) using the same method used to prepare HAMLET [12]. Akin to HAMLET, bLAC has been reported to be cytotoxic toward numerous cancer cells by mechanisms that are different from classic apoptosis [13]. It is of great significance to be able to prepare the protein–lipid complex using bovine milk because it will allow large-scale production, circumventing the problem of limited availability of human milk.

In this study we investigated the therapeutic efficacy of bLAC on bladder cancer both in vitro and for the first time in an orthotopic tumour model. The findings show that bLAC selectively kills cancer cells in vitro and inhibits tumour growth in a previously well characterized rat model without noticeable toxicity.

Materials and Methods

Agents

The preparation of bLAC has been reported previously [13]. It was produced from bovine α-lactalbumin mixed with oleic acid and the mixture was then applied on an anion exchange column. The bLAC used in the present study was provided by NatImmune A/S (Copenhagen, Denmark). The bLAC preparation was concentrated to 2, 6 or 10 mg/mL and aliquoted into 1.2 mL/vial by the manufacturer. The native bovine α-lactalbumin without oleic acid, bLA (10 mg/mL, 1.2 mL/vial), was also provided by NatImmune A/S as a parallel control for bLAC. Tice strain BCG (OncoTICE®, Organon Canada Ltd, Canada) was provided by the Cross Cancer Institute (Edmonton, Alberta, Canada). BCG aliquots were reconstituted in sterile saline (0.9% NaCl) before use.

Cells and Cytotoxicity Assay

Cell lines used for cytotoxicity assays included human bladder urothelial cancer cell lines UMUC-3, 6, 9 and 14 (kindly provided by H. Barton Grossman, MD Anderson Cancer Center, University of Texas, USA); RT4, T24, 253J (kindly provided by Martin Gleave, Prostate Center, University of British Columbia, Vancouver, Canada); MGH-U3 (kindly provided by Yves Fradet, Laval University, Quebec, Canada); HTB-9 (American Tissue Culture Collection, ATCC), RT112 (ATCC) and HT-1376 (ATCC, CRL-1472); rat bladder urothelial cancer cell line AY-27 (originally provided by Steven H. Selman, University of Toledo Medical Center, OH, USA, and reconfirmed by us); and a human fibroblast cell line F2P6 (kindly provided by Aziz Ghahary, Wound Healing Research, Department of Surgery, University of Alberta, Canada). HT-1376 cells were reported to be resistant to apoptosis after treatment by TRAIL (tumour-necrosis-factor-related apoptosis-inducing ligand) [14]. Cells were grown in a humidified incubator under standard conditions of 37 °C and 5% CO2 in RPMI-1640 or Dulbecco's modified Eagle's medium (Gibco®, Invitrogen; Carlsbad, CA, USA), supplemented with 10% heat-inactivated fetal bovine serum (FBS) except where noted, 100 U/mL penicillin, 100 μg/mL streptomycin and 0.25 μg/mL amphotericin-B (Invitrogen).

Exponentially growing cells (1 × 104) in 96-well plates were incubated with appropriate medium overnight. On day 2, cells were incubated with serum-free medium containing escalated concentrations of bLAC, bLA or no drug (sham control) for 2 h. Therapeutic medium was replaced with fresh medium with 10% FBS and cells were incubated for another 20 h. Cell viability was estimated using MTT assay (Invitrogen; Carlsbad, CA, USA) with 50 μL of 50 μg/mL MTT (3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide, which is reduced to purple formazan by mitochondrial enzymes in living cells) and the resultant formazan crystals dissolved in 150 μL dimethyl sulfoxide. The absorbance of each well was measured with a microplate reader at 570 nm and percentage reduction in cell viability was calculated as [1 − (absorbance of treated cells)/(absorbance of control cells)] × 100. Each assay was repeated at least three times.

Apoptosis Detection by Microscopy

Apoptotic cells were detected by membrane change using the annexin V staining kit (Alexa Fluor® 488 Annexin-V/Propidium iodide, Molecular Probes™) used according to the manufacturer's protocol. Cells (4 × 104/well) were seeded in two-chamber Lab-Tek slides 24 h before exposure to bLAC (0.4 mg/mL) or bLA (0.8 mg/mL) in serum-free medium for 2 h at 37 °C. Fluorescence of the cell membrane and nucleus was observed by Zeiss confocal microscopy (LM510) using appropriate filters. Random representative images of three separate experiments were analysed and captured for presentation.

The terminal deoxynucleotidyl-transferase-mediated nick end labelling (TUNEL) assay was also used to detect apoptotic nuclei with the TdT-FragEL DNA fragmentation detection kit (Calbiochem®) according to the manufacturer's instructions. Cells were plated on chamber slides and cultured to 80% confluence before incubation with bLAC or bLA in medium without FBS for 2 h. DNA fragments were detected by streptavidin horseradish peroxidase conjugate and H2O2 plus 3,3′-diaminobenzidine. Cells were counterstained with methyl green. Stained cells were imaged with a Zeiss Axioplan upright digital microscope equipped with a cooled charge coupled device (CCD) camera and analyzed with the MetaMorph® software.

Cell Cycle DNA Analysis by Flow Cytometry

Cells were seeded in 24-well plates and cultured in medium with 10% FBS overnight before exposure to bLAC in serum-free medium for 2 h. Cells were then incubated in fresh medium with 10% FBS for an additional 20 h, harvested and suspended in cold PBS/anhydrous ethanol. Cell preparations (5 × 105/mL) were kept at 4 °C overnight. Cells were pelleted by centrifugation and washed with cold PBS. To the cell pellet, 100 μL RNaseA/propidium iodide (RNaseA 250 μg/mL, propidium iodide 20 μg/mL in PBS) was added. Cells were kept at 22 °C for 30 min in the dark and cell cycle cellular DNA quantity analysed by flow cytometry (ModFit DNA software, Becton Dickinson, Canada). The percentage of DNA content in sub-G1 representing the percentage of apoptotic cells.

Selective Cytotoxicity Assay in Co-Cultured Spheroids

Multicellular spheroids were produced as previously reported [15]. Briefly, spheroids of bladder cancer cells and normal fibroblasts were grown in 96-well plates coated with 1% agarose gel, in appropriate medium under standard conditions (5% CO2, 37 °C). When two small spheroids were combined (co-cultured), the cancer cells (AY-27) grew around the fibroblasts, mimicking a tiny tumour for in vitro topical application studies. To demonstrate the bladder cancer and fibroblast cellular orientation within the co-cultured spheroids anti-cytokeratin 13 (clone CAM, 5.2, BD Canada, San Jose, CA, USA) and anti-vimentin (DAKO) antibodies respectively were used, on formalin-fixed spheroid sections. When co-cultured spheroids reached sizes of 300–400 μm in diameter, they were transferred to new wells and cultured with bLAC (0.8 mg/mL) in serum-free medium for 4 h, followed by incubation in fresh medium with 10% FBS for 20 h. To stain the live and dead cells of the spheroids, Syto-16 (5 μM, Molecular Probes) and propidium iodide (5 μM) in 100 μL medium were added to each well and incubated for 1 h at 37 °C in the dark. Multiple images of the spheroids were obtained using the Zeiss Axioplan upright digital imaging microscope equipped with a CCD camera. Images were merged with MetaMorph™ software and Adobe Photoshop. Representative images were captured for analysis and presentation.

Animal Tumour Model and Intravesical Therapy

The orthotopic rat bladder tumour model was established as previously reported [16]. All animal procedures were carried out in accordance with guidelines regulated by the Canadian Council on Animal Care and approved by the University of Alberta Animal Care Committee. Female Fischer F344 rats (Charles River Laboratories, Quebec, Canada) were bred in the university facilities. Animals weighing ≈150 g were anaesthetized with inhalation of 2% isoflurane in oxygen. The bladders were catheterized with an 18 gauge angiocatheter (BD Insyte™, Sandy, Utah, USA) and bladder mucosa preconditioned with 0.1 M HCl, neutralized with 0.1 M KOH and then flushed with sterile PBS (pH 7.4) three times. Single cell suspensions of AY-27 cells (3 × 106) in 0.5 mL of serum-free medium were then instilled via the catheter and left indwelling for 1 h. The rat position was changed from side to side to facilitate full bladder wall exposure. The catheter was then removed and the rats were allowed to void spontaneously. Nearly 100% tumour engraftment has been achieved in these syngeneic immunocompetent rats when ≥2 × 106 tumour cells are inoculated [16-19]. By cystoscopic and histological examination the majority of the established tumours have the tiny papillary configuration associated with carcinoma in situ by 6 days post-inoculation and progress to invasive disease if untreated [18, 19].

Six days after tumour cell inoculation, animals were treated with intravesical instillations of escalated doses of bLAC (2 mg/mL, 6 mg/mL or 10 mg/mL), bLA (10 mg/mL, control), BCG (standard control, based on previous studies) [17, 20] or saline (Table 1). Treatments were given twice weekly for 3 weeks. Instilled volume was 0.5 mL per treatment. To absolutely ensure accurate drug exposure, a purse-string suture was placed in the skin around the urethral meatus for 90 min to keep the solution in the bladder while the animals were under anaesthesia. Once the animals completed the treatment regimes, they were observed for ≥90 days post-implant except for those that needed to be euthanized earlier for signs and symptoms of tumour progression (body weight loss ≥10%, haematuria and urinary retention). All animals were euthanized using pentobarbital (Euthanyl®, Bimeda-MTC Animal Health Inc.) and subjected to necropsy for tumour growth and invasion. The bladders were excised and photographed. Samples were embedded in paraffin, cut and stained with haematoxylin and eosin for histological examinations. Other organs that appeared grossly abnormal were also excised and histologically examined.

Table 1. Fisher rats receiving intravesical treatments starting 6 days after inoculation of 3 million AY-27 rat UCC cells
Treatment groupsDrug concentrationTreatment scheduleNo. of ratsNo. of LTSsaNo. (LTSs) tumour-freeP vs saline
  1. Instilled volume was 0.5 mL per treatment.
  2. aLong-term survivor (LTS) means animals survival >90 days post-implant.
Control0.9% NaClTwice/week, 3 weeks1121Not applicable
bLA10 mg/mLTwice/week, 3 weeks12440.06
bLAC-1010 mg/mLTwice/week, 3 weeks121080.004
bLAC-66 mg/mLTwice/week, 3 weeks11760.01
bLAC-22 mg/mLTwice/week, 3 weeks11650.07
BCG (Tice®)2 × 107 CFU/mLTwice/week, 3 weeks12870.01

Statistical Analysis

All data on cell death are presented as the mean ± sd of at least three independent experiments. The results were analysed using the unpaired two-tailed t test. Animals terminated earlier than 90 days post-implant were allocated to censored cases, while those surviving ≥90 days were counted as long-term survivors (LTSs). Survival curves from different groups of animals were plotted by the Kaplan–Meier method. The log-rank (Mantel–Cox) test was used to compare survival time distributions vs saline control (GraphPad Prism®, version 5.05, San Diego, CA, USA). P < 0.05 was considered significant.

Results

bLAC Induces Apoptotic-Like Cell Death in Bladder Cancer Cells

To examine the cytotoxic effect of bLAC and bLA on bladder cancer cells, several human urothelial cell cancer (UCC) cell lines and a rat UCC cell line (AY-27) were tested by the MTT assay. All the human UCC cell lines tested showed a dose-dependent sensitivity to bLAC killing with a range of LC50 values from 0.15 to 0.8 mg/mL (Fig. 1A). These results were consistent with HAMLET cytotoxicity to cancer cell lines from prior studies with LC50 values ranging from 0.2 to 0.4 mg/mL [13]. Interestingly, the TRAIL-resistant cell lines HT-1376 and UC14P1000 (which was selected from UMUC-14 after exposure to TRAIL [14]) also showed similar sensitivity to that of TRAIL-sensitive cell lines; TRAIL being one of the mechanistic pathways for BCG.

Figure 1.

In vitro cytotoxicity of bLAC to urothelial cell cancer (UCC) of the bladder. (A) Using the MTT assay (see Materials and Methods for details), all human UCCs tested show a dose-dependent bLAC-induced reduction in cell viability, with LC50 values ranging from 0.15 to 0.5 mg/mL in 11 out of 12 cell lines. UMUC-9 cells are slightly less sensitive (LC50, 0.8 mg/mL). The values are means ± sd of three separate experiments. (B) To compare the cytotoxicity of bLAC and bLA to UCCs, two human UCC cell lines with differing sensitivity and a rat UCC cell line were tested by the MTT assay. Again, bLAC reduces UCC viability in a dose-dependent fashion, while bLA shows little effect. Values are means ± sd of triplicate experiments. There is a statistical difference between bLAC and bLA for each cell line's response to doses of ≥0.2 mg/mL (*P < 0.05, **P < 0.01; unpaired t test). (CE) Representative images of HTB-9 cells treated with bLAC, 0.4 mg/mL, for 2 h (C,E) or bLA, 0.8 mg/mL, for 2 h (D) show that only bLAC induces cell apoptosis. Using the annexin V staining kit, annexin V (green) stains the cell membrane of apoptotic cells (early apoptosis) and propidium iodide (pinkish red) stains the nuclei of apoptotic cells; live cells appear blue (C). The TUNEL assay detects apoptotic cell nuclei in brown (E) and live cells in green (D,E). Scale bars, 200 μm.

To compare the cytotoxic activity of bLAC and bLA on UCC cells, human cancer cells with various sensitivities (253J, HT-1376) and the rat AY-27 cells were treated with graded doses of bLAC or bLA (first 2 h in serum-free medium followed by 20 h in normal medium). Cell viability was determined by the MTT assay. Again, bLAC induced dose-dependent cell death in all UCC cells, while bLA did not show marked cytotoxicity on the tested cells (Fig. 1B). The differences were significant by a two-tailed unpaired t test between bLAC and bLA for each cell line at drug dose ≥0.2 mg/mL (Fig. 1B).

Apoptotic-like cell death induced by bLAC was determined qualitatively by annexin V staining and TUNEL assays in HTB-9 cells (Fig. 1C–E), and quantitatively by flow cytometry analysing the sub-G1 DNA content of UCC cells with varying sensitivity (Fig. 2). Annexin V (green) has a high affinity to phospholipid phosphatidylserine, which is translocated from the inner to outer leaflet of the cell membrane in apoptotic cells following bLAC exposure. The nuclei of these cells were stained by propidium iodide (red), and live cells were stained blue (Fig. 1C). The TUNEL assay detected apoptotic cells undergoing DNA fragmentation following bLAC treatment (brown, Fig. 1E). No apoptosis was observed in cells exposed to bLA (green, Fig. 1D). Similar results from other UCC cells (RT4, UMUC-3, 6, 9, 14) were observed (data not shown).

Figure 2.

Flow cytometric analysis of sub-G1 nuclear DNA content indicates a dose-dependent apoptosis induced by bLAC in human UCC cells. UCC cells were left untreated (control) or treated with bLAC for 2 h in serum-free medium followed by 20 h incubation in medium with 10% FBS. Representative histograms of DNA content from UCC cells show a higher percentage of apoptosis (sub-G1) in more sensitive cells (253J) than in less sensitive cells. By doubling the bLAC doses in those less sensitive cells, a dose-dependent increase of sub-G1 DNA percentage is observed.

Since both annexin V staining and TUNEL assays could not include the detached cells which were washed away during staining, flow cytometry was performed to quantitatively detect bLAC-induced apoptosis. The percentage of sub-G1 DNA content represents the percentage of apoptotic cells [21]. Figure 2 depicts the sensitive 253J and less sensitive MGH-U3 showing marked differential apoptosis after exposure to 0.4 mg/mL bLAC for 2 h. As much as 46% apoptosis was induced in 253J, but only 3.4% apoptosis was observed in MGH-U3. Interestingly, in those less sensitive cell lines, a dose-dependent cell apoptosis was induced by increasing the bLAC dose.

bLAC Selectively Kills Cancer Cells in Co-Cultured Spheroids

Along with the bLAC cytotoxicity studies in the panel of UCC cells, we investigated whether bLAC could selectively kill cancer cells while sparing normal cells. To this end, multicellular spheroids of AY-27 cells and normal fibroblasts were treated with bLAC (0.8 mg/mL) in serum-free medium for 4 h. Therapeutic medium was then replaced with fresh medium with 10% FBS. Spheroids were incubated for an additional 20 h, followed by staining with live cell nucleic probe Syto-16 (green) and dead cell stain propidium iodide (red) for the live/dead assay (Fig. 3). As shown in Fig. 3A,B, both AY-27 cells (peripheral layers) and fibroblasts (centre part) were live (green) when the co-cultured spheroid was incubated in medium without bLAC (control). However, bLAC selectively killed AY-27 cells (red) while sparing the normal fibroblasts (green) when two side-by-side multicellular spheroids (Fig. 3C,D) or co-cultured spheroids (Fig. 3E,F) were treated with bLAC.

Figure 3.

Vital staining shows that bLAC selectively kills cancer cells in multicellular spheroids. (A,B) A co-cultured spheroid composed of AY-27 cells (outer layers) and fibroblasts (centre) was cultured in medium without bLAC (control) for 24 h (first 4 h without serum). (C,D) Two separate side-by-side spheroids from AY-27 cells (left) and fibroblasts (right) were treated with bLAC (0.8 mg/mL) for 4 h, followed by an additional 20 h incubation in fresh medium. (E,F) A co-cultured spheroid composed of AY-27 cells and fibroblasts (with AY-27 cells oriented around the fibroblasts) was incubated with bLAC (0.8 mg/mL) for 4 h, followed by an additional 20 h incubation in fresh growth medium. Spheroids were then stained with 5 μM live cell nucleic acid probe Syto-16 (green) and 5 μM dead cell stain propidium iodide (red) for live/dead assay. White light images of the spheroids are shown in A, C and E, and the respective fluorescent images are shown in B, D and F. (G,H) Formalin-fixed IHC 5 μm sections of co-cultured spheroids stained with either anti-vimentin (fibroblast specific) (G) or anti-cytokeratin 13 (epithelial cell specific) (H), demonstrating peripheral orientation of urothelial cells. Scale bars, 100 μm.

Intravesical bLAC Inhibits Tumour Growth In Vivo

Based on the in vitro findings of bLAC cytotoxicity, we then tested the antitumour efficacy and toxicity of bLAC in an orthotopic rat bladder cancer model. The tumour model is a well established rat UCC of the bladder model which mimics early stage human UCC of the bladder [16-19]. BCG, the most commonly used intravesical agent in the clinic [2], was used as a standard control. The dose of BCG was determined based on previous animal studies [17, 20]. In addition, the native bLA and saline were chosen as control agents. Intravesical therapies were started 6 days after tumour cell inoculation (see Table 1). At this time period >90% of the rats were reported to have early stage bladder cancers confirmed by histology and cystoscopy [16, 18, 19]. Figure 4A shows that marked survival benefits were achieved for animals treated with bLAC instillations. A dose-dependent response for tumour-bearing animals to bLAC therapies was observed. For animals receiving bLAC concentrations greater than 6 mg/mL, significant LTSs and tumour growth inhibition (tumour-free LTSs) were attained compared with rats receiving saline instillations (P = 0.01 or 0.004, for 6 mg/mL or 10 mg/mL, respectively; log-rank test), suggesting that bLAC was as effective as BCG therapy (P = 0.01, Table 1 and Fig. 4A). There were noticeable differences in tumour growth between bLAC and saline-treated rat bladders (Fig. 4B–E). Saline-treated bladders showed extensive papillary tumours in the bladder lumen and thickened bladder walls due to tumour invasion (Fig. 4B,C), whereas bLAC-treated bladders appeared to be normal (Fig. 4D,E). Figure 5 depicts early stage bladder tumours at day 6 post-inoculation before treatment was started. In contrast to our hypothesis based on in vitro cytotoxicity data, bLA (10 mg/mL, P = 0.06 vs saline) also showed similar benefit to low dose bLAC (2 mg/mL, P = 0.07) in this immunocompetent rat model. No evident side-effects other than procedure-related signs were found with bLAC therapies with regard to rat body-weight change and general well-being.

Figure 4.

Intravesical instillations of bLAC inhibit bladder tumour growth and prolong animal survival. (A) Kaplan–Meier survival curves of rats bearing AY-27 bladder cancer treated with instillations of bLAC (2, 6, 10 mg/mL), bLA (10 mg/mL), BCG (2 × 107 CFU/mL) or saline commenced at day 6 post-implantation. See Table 1 for details of treatment and statistical analysis. Animals receiving bLAC 6 mg/mL or greater, or BCG instillations, all attain significant survival benefits compared with controls receiving saline (P = 0.01). A marginal survival benefit is shown for animals receiving bLA (10 mg/mL) or lower dose bLAC (2 mg/mL) instillations (P = 0.06, 0.07, respectively, vs control). (B) Gross photographs of rat bladders (sagittally opened from anterior midline) treated with saline instillations. Animals were terminated before 90 days post-implantation of AY-27 cells. Extensive papillary tumours cover the bladder lumen with thickened bladder wall due to tumour invasion. (C) A representative tissue section (haematoxylin and eosin staining) of the bladder from (B) shows extensive invasion of cancer cells in the muscular layers. Scale bar, 200 μm. (D) Gross photographs of rat bladders (sagittally opened from anterior midline) treated with bLAC (10 mg/mL) instillations. Animals were sacrifice >90 days post-inoculation of AY-27 cells. Bladders are smaller in size with thinner smooth wall compared with saline-treated ones. (E) A representative tissue section (haematoxylin and eosin staining) of the bladder from (D) depicts a virtually normal bladder wall. Scale bar, 200 μm.

Figure 5.

A representative tissue section (haematoxylin and eosin staining) of the bladder wall at day 6 post-inoculation of AY-27 cells depicts tiny papillary tumours associated with flat carcinoma in situ. Scale bar, 200 μm.

Discussion

An ideal antitumour agent should selectively kill cancerous cells while sparing normal cells or tissues. Among the current list of limited available options, milk protein products appear to be promising. Human α-lactalbumin is a protein in human milk. HAMLET is produced in vitro by removal of Ca2+ and by adding oleic acid, which locks the protein in the active partially unfolded conformation [7]. It has been hypothesized that HAMLET might act as a natural protectant (tumour inhibitor) in breast-fed children. Studies show that HAMLET selectively kills tumour cells in vitro as well as in in vivo tumours [7-10]. The mechanisms of HAMLET-induced cell death and the resistance of normal cells are not fully understood. Early studies show that HAMLET has many targets in tumour cells but not in differentiated cells. HAMLET translocates from the cell surface by endocytosis to the nuclei, where it binds histones and disrupts the function of chromatin [22]. HAMLET also affects the mitochondria and causes an apoptotic response with cytochrome c release and DNA fragmentation [6, 23]. However, tumour cell death does not depend on classical apoptosis, and caspase inhibitors or over-expression of Bcl-2 do not influence cell sensitivity to HAMLET [11]. Recently, HAMLET has been shown to trigger mitochondrial damage and inhibition of mammalian target of rapamycin (mTOR) activity, leading to macroautophagy in tumour cells [24], which plays a role in cell death as a backup system for apoptosis [25, 26]. Thus, combination therapy of HAMLET with mTOR inhibitors may have a role in eliminating transformed UCC.

The bovine counterpart of HAMLET, bLAC, consists of partially unfolded bovine α-lactalbumin and oleic acid. Prior comparative studies of bLAC and HAMLET showed similar efficacy and specificity toward malignant cells [13]. The selective antitumour activity of bLAC has been confirmed in our study using a large panel of UCC cells and a well characterized orthotopic rat bladder cancer model. The mechanism of action of bLAC has been proposed to be slightly different from that of HAMLET. Instead of accumulating in the mitochondria, bLAC accumulates in the endolysosomal compartment where it induces destabilization of the lysosomal membrane causing leakage of lysosomal content (e.g. cathepsins and other lysosomal hydrolases) into the cytoplasm, which ultimately leads to cell death [13].

Similar to HAMLET, bLAC causes a dose-dependent apoptotic-like cell death in all UCC cells tested in the present study. Some of these cells have been shown to be refractory to TRAIL, which is a classical apoptosis-inducing agent of the tumour necrosis factor family and a major effector of intravesical BCG therapy for bladder cancer [14, 27, 28]. bLAC selectively kills cancer cells in co-cultured multicellular spheroids consisting of UCC cells and fibroblasts (Fig. 3). The orthotopic rat model used in this study represents a relatively aggressive bladder tumour model with rapid progression in which only therapies with robust antitumour efficacy can inhibit tumour growth [17, 19, 20]. Our in vivo findings clearly demonstrate intravesical bLAC can inhibit tumour growth and prolong tumour-bearing animals' survival vs saline control. The antitumour efficacy of bLAC is dose-dependent and comparable with that of BCG therapy. A previous study in patients with bladder cancer suggested that intravesical HAMLET (daily for 5 days before tumour resection) caused direct tumour cell death, whereas the native α-lactalbumin did not [9]. This antitumour activity was substantiated in a later animal study with a similar treatment regimen but without long-term follow-up [29]. Whether HAMLET or its native α-lactalbumin can elicit an immunological reaction in the bladder wall (like BCG) is unclear, especially if the patients were to receive longer treatments. In the present animal model, bLA (10 mg/mL) did show a similar antitumour effect to low dose bLAC. The explanation of this observation is unknown. It is possible that these proteins (bLAC or bLA) may interact immunologically or otherwise with the complex tumour microenvironment following repeat instillations, an interaction that would not be observed in vitro. How this interaction changes the agent (conformation) or the host response needs further investigation.

In conclusion, findings from ourselves and others have shown that protein products from the milk of human or cow have unique antitumour activity both in vitro and in vivo. Since many cancers have inherent or acquired resistance to drugs which cause classic apoptosis, and there is a shortage of selective cancer therapeutics, the nature of bLAC makes it a promising antitumour agent used alone or in combination for topical therapies. The lack of mutagenic or infective properties makes it an ideal agent for intravesical therapy of NMIBC. Further clinical studies are needed for bLAC alone or in combination with intravesical BCG for BCG naïve or refractory NMIBC. Once clinical efficacy is established in high risk NMIBC, then post-transurethral resection installation therapy should be explored for prevention of recurrence from implantation.

Acknowledgements

The authors thank Mads Axelsen of NatImmune A/S for providing bLAC, bLA and other agents for this study. We also appreciate the contribution of X.J. Sun and G. Barron from the imaging facilities of the Cross Cancer Institute.

This work was financed in part by NatImmune A/S (KK, RBM), Alberta Cancer Research Institute grant #23798 (RBM), Alberta Innovates and Health Solutions (RBM) and the National Cancer Institute of Canada grant #21531 (RBM).

Conflict of Interest

The sponsors had no role in data collection and analysis, interpretation of the results, the preparation of the paper or the decision to submit the paper for publication. Karen Koch is an Employee of Sponsor.

Ancillary