Cancer is a second leading cause of death globally and one of the biggest challenges facing biomedical scientists. Despite the considerable progress made in targeted cancer treatment, the heterogeneity of solid tumor greatly limits precision oncology therapy. Dual targeting strategies applying targeting moieties recognized different receptors in tumor population can be a solution of this problem.
Tumor targeting with naked antibodies and antibody-drug conjugates has become an established strategy for cancer-related therapy in clinics, particularly if conventional therapies have failed [1
]. However, antibodies have practical limitations due to their poor expression yield and aggregation tendency, at least for some constructs [3
]. A solution might come from the use of alternative non-IgG binding scaffolds [5
]. Due to their small size, high affinity to target and robust production scaffold proteins represent an attractive alternative to immunoglobulin proteins.
Previously designed ankyrin repeat proteins (DARPins), a novel class of non-IgG scaffolds based on naturally occurring ankyrin repeats [7
] have been shown to bind to protein targets with specificity and affinity exceeding those of antibodies [7
]. DARPins are small (13−20 kDa), highly soluble in water, very stable and lack cysteine residues. A number of DARPin molecules that specifically bind to different tumor-associated antigens, such as human epidermal growth factor receptor 2 (HER2) or the epithelial cell adhesion molecule (EpCAM) overexpressed in breast and ovarian cancer cells, have been developed [8
The transmembrane protein HER2 is one of the most well-studied tumor markers; its overexpression represents a hallmark of many types of tumors associated with an increased risk of metastasis and resistance to chemotherapy [13
]. The epithelial cell adhesion molecule (EpCAM) has also emerged as a promising structure for targeted therapy of solid tumors. EpCAM is expressed at low levels on basolateral cell surfaces of some normal epithelia [14
]. In contrast, high levels of homogenously distributed EpCAM are detectable on cells of epithelial tumors [15
], and its overexpression represents an independent prognostic marker for reduced survival in patients with breast and ovarian cancer [17
As a tumor-targeting moiety for delivery of cytotoxic agents (e.g. toxic proteins), DARPins are widely used in modern research [19
]. Toxins of bacterial origin are widely used as cytotoxic component in anticancer therapy [21
]. In this work we use ribonuclease Barnase (Bn) from Bacillus amyloliquefaciens
] and Pseudomonas
exotoxin A secreted by gram-negative bacteria Pseudomonas aeruginosa
with removed or inactivated human B-cell recognition sites (LoPE) [27
]. The cytotoxic action of Bn is based on its high ribonuclease activity, LoPE irreversibly inhibits eukaryotic elongation factor eEF2 that leads to protein biosynthesis blocking in a cell.
Previously we reported a novel method for preparation of HER2-targeted liposomes (80–90 nm in diameter), each containing thousands of encapsulated protein molecules [28
]. Here we use this approach for the preparation of liposomes loaded with RNAse Bn for anti-HER2 cancer therapy. We have shown that simultaneous treatment of animals with anti-EpCAM fusion toxin based on LoPE and HER2-specific liposomes loaded with Bn leads to concurrent elimination of primary tumor and metastasis.
Targeting therapy of solid tumors represents a great challenge because of heterogeneity of tumor-associated antigen expression. Combination therapy, targeting two (or more) different receptors on a tumor cell has gained considerable attention in the field of oncology in recent years, with numerous studies demonstrating its significant advantage over monotherapies [34
High levels of the HER2 expression observed in a variety of carcinomas and clinical success of the HER2-directed monoclonal antibody Herceptin make the HER2 gene product a promising therapeutic target [38
]. EpCAM is also a promising therapeutic target, as numerous studies indicate its role as a cell surface marker for various types of carcinoma. A clinical study has shown that coexpression of both HER2 and EpCAM at high levels correlates with poor prognosis in breast cancer patients [39
], suggesting that a drug targeting both these moieties would be highly beneficial to these high-risk patients.
In the present study we compared antigen-monospecific therapy with dual-specific therapy based on DARPin-fusion toxic complexes targeted two different tumor-associated antigens, HER2 and EpCAM. For this purposes we developed HER2-specific liposomes loaded with a large amount of Bn and created EpCAM-specific fusion protein based on DARPin and LoPE (Figure 1
The effectiveness of the developed targeted agents for solid tumors as well as metastasis treatment was evaluated using mouse model with xenograft tumors derived from BT-474 cell line. It was recently shown that ductal carcinoma cell line BT-474 has a great potential as cancer model with spontaneous metastasis [40
]. It was shown BT-474-based tumors can form metastases in lymph nodes, lungs or spleen with various frequencies. Here we observed the formation of metastasis in inguinal lymph node after s.c. injection of BT-474 cells in all experimental animals that do not receive any specific treatment, which makes this tumor model a very convenient tool for tumor and metastasis grow dynamics study.
Antigen specificity and cell toxicity of these protein complexes were thoroughly proved in vitro (Figure 2
). Using whole-body fluorescence imaging we demonstrated that DARP-Lip(Bn) and EC1-LoPE displayed highly efficient tumor targeting (Figure 3
). Fluorescence histochemistry of the frozen sections of tissue samples revealed intracellular localization of both target agents in the tumor with the distinct colocalization in the tumor area (Figure 4
). We have shown that monotherapy with EpCAM-specific LoPE or HER2-specific liposomes loaded with Bn significantly reduced primary tumors but did not affect distant metastases (Figure 5
). On the other hand, dual-specific therapy led to elimination of both primary tumors and metastases.
Dual targeting strategy is widely used in contemporary oncology researches. It includes dual targeting of different epitopes on one receptor or ligand, dual targeting of a receptor and a ligand, dual retargeting of toxins (reviewed in [44
]). Results from all these studies demonstrate that in general dual targeting strategies outperforms monotreatment.
The results obtained in this work confirm the advantages of dual targeting therapy, and we can state that protein therapeutics for dual targeting suggests a new way in cancer treatment and will enter clinical study in the near future.
4. Materials and Methods
Unless otherwise stated, reagents and chemicals were obtained from Sigma-Aldrich (St. Louis, MO, USA) and were used without further purification.
The following proteins were used in experiments. DARPin_9-29 were produced in E. coli
and purified as described in details in our previous work [28
]. Wild-type Bn was extracted and purified from culture media following the method of Hartley [26
] with slight modifications.
Recombinant EC1-LoPE was produced in Escherichia coli
BL21(DE3) strain cells transformed with the plasmids pET22-EC1-LoPE. The last plasmid was obtained by replacement of Darpin_ 9-29
gene in the plasmid pDARP-LoPE [47
] with the gene for Darpin-EC1
. The Darpin-EC1 nucleotide sequence was deduced from its amino acid sequence published by Stefan et al. [12
], taking into account the codon usage in highly expressed E. coli
]. The EC1
gene was assembled from chemically synthesized overlapped oligonucleotides of 50 bp length by PCR and placed into the plasmid pDARP-LoPE between restriction sites Nde
I and EcoR
I. Expression and purification procedures for EC1-LoPE were essentially the same as was described earlier for the protein DARPin-LoPE [47
]. Briefly, E. coli
was grown in autoinduction ZYM-5052 medium prepared according to Studier [49
] containing 100 µg/ml kanamycin at 25 °C. The cells were harvested by centrifugation at 10,000 × g at 4 °C for 20 min, and resuspended in lysis buffer (200 mM Tris-HCl, 500 mM sucrose, 1 mM EDTA (pH 8.0), 1 mM PMSF and 60 µg/ml lysozyme). The suspension was diluted 2-fold with distilled water and incubated at room temperature for 30 min. Cells were broken on ice using a Vibra Cell ultrasonic liquid processor VCX130 (Sonics & Materials, Inc., Newtown, CT, USA). The cellular debris was pelleted at 70,000 × g at 4 °C for 30 min. After addition of imidazole (30 mM) and NaCl (500 mM), the supernatant was filtered through a 0.22 µm membrane and applied onto a HisTrap HP 1 ml column (GE Healthcare, Little Chalfont, UK) equilibrated with 20 mM NaPi (pH 7.5), 500 mM NaCl and 30 mM imidazole. The bound proteins were eluted with a linear 30–500 mM imidazole gradient. The fraction containing EC1-LoPE was combined, diluted 5-fold with 25 mM Tris-Cl (pH 8.0), and loaded onto a MonoQ 10/100 GL column (GE Healthcare, Little Chalfont, UK) equilibrated with the same buffer. The bound proteins were eluted with a linear 0–1 M NaCl gradient. The fractions were analyzed by 15% reducing SDS-PAGE. Protein concentration was determined by UV spectroscopy using ε280
= 48,220 M−1
4.2. Preparation of Bn-Encapsulated Liposomes
The phospholipid suspension was prepared by swelling 0.2 g of L-α-phosphatidylcholine granules (Soy 40%, 341602, Avanti Polar Lipids, Alabaster, AL, USA) in 10 mL of distilled water for 16-20 h at ambient temperature. The suspension was further filtered through a 0.8 µm syringe filter and stored at 4 °C until used.
An aliquot of 0.5 mL of the phospholipid suspension was mixed with 0.5 mL of Bn (4 mg/mL) in 20 mM K-Pi (pH 7.5). The protein-lipid suspension was frozen and thawed 5 times and subsequently extruded 19 times through a polycarbonate membrane with 100 nm pore size at ambient temperature using an Avanti Mini Extruder (Avanti Polar Lipids, Alabaster, AL, USA). The extrusion yields unilamelar liposomes (SUVs) with an average diameter of 80–90 nm. The liposomes were purified from the excess of Bn on a Sepharose CL-2B column (10 × 35 mm) equilibrated with 10 mM K-Pi (pH 7.5) containing 0.2 M NaCl. At high ionic strength the protein dissociates from the outer surface of the liposome membrane. Chromatography on a Sepharose CL-2B column enabled to completely separate the liposomes from the protein not associated with the vesicles. The liposome fraction, collected in the void volume, was stored at 4 °C until used.
4.3. Covalent Coupling of DARPin_9-29 to Liposomes
Covalent attachment of DARPin_9-29 to the outer surface of the liposome membrane was described in details in [28
]. In short, liposomes containing Bn were prepared as described above and incubated with 6 mM 2-iminothiolane (Traut’s reagent) at ambient temperature for 1 h in 100 mM K-Pi (pH 8.0) containing 1 mM EDTA. Incubation with Traut’s reagent leads to the attachment of a residue terminated by SH-group to primary amino groups of the amine-containing phospholipids composing the liposome membrane. The phospholipid mixture used here (soy PC, 40%, from Avanti Polar Lipids, Alabaster, AL, USA) consisted of about 16% phosphatidylethanolamine, making it possible to introduce many SH-groups to the surface of a liposome. The incubation mixture was passed through a NAP-5 desalting column (GE Healthcare, Little Chalfont, UK) equilibrated in 100 mM K-Pi (pH 7.5). The void volume fraction (~0.5 mL) was collected and kept on ice.
DARPin_9-29 (2–4 mg/mL) was incubated with a 10-fold molar excess (1–2 mM) of sulfo-EMCS (N-ε-maleimidocaproyl-oxysulfosuccinimide ester) for 45 min under ambient conditions. This heterobifunctional amine-to-sulfhydryl linker, containing the succinimide (which specifically reacts with primary amino groups) and the maleimide (which reacts with thiols) moieties at opposite ends of the molecule, covalently binds to DARPin_9-29 through one of primary amino groups on the surface of the protein. The protein lacks SH-groups, thus leaving the maleimide group of the linker untouched and available for further reaction with the Traut’s reagent-treated liposomes. The DARPin modification should occur simultaneously with treatment of the liposomes with Traut’s reagent in order to avoid oxidation of SH-groups introduced to the bilayer. The incubation mixture was passed through a NAP-5 desalting column (GE Healthcare, Little Chalfont, UK) equilibrated in 100 mM K-Pi (pH 7.5). The sulfo-EMCS-treated DARPin was mixed with equal volume of the Traut’s reagent-treated liposomes and incubated for 2 h under ambient conditions. The non-bound DARPin was separated from the liposomes on a Sepharose CL-2B column (10 × 35 mm) equilibrated with 20 mM K-Pi (pH 7.5). The void volume fraction was collected and stored at 4 °C until used.
4.4. Cryo-Electron Microscopy
Cryo-electron microscopy (cryo-EM) was used for direct visualization of liposomes. Prior to cryo-EM study 3 μl of liposome suspension were applied to a glow discharged Lacey carbon EM grid, which was then blotted for 2.5 s at 4 °C and vitrified by a rapid plunging into liquid ethane pre-cooled with liquid nitrogen using Vitrobot Mark IV (ThermoFisher Scientific, Waltham, MA, USA). The obtained samples were studied using Titan Krios 60-300 TEM/STEM (ThermoFisher Scientific, Waltham, MA, USA) cryo-EM, equipped with XFEG electron source, TEM direct electron detector Falcon II (ThermoFisher Scientific, Waltham, MA, USA) and Cs image corrector (CEOS, Heidelberg, Germany) at accelerating voltage of 300 kV. To minimize radiation damage during image acquisition low-dose mode in EPU software (ThermoFisher Scientific, Waltham, MA, USA) was used.
4.5. Labeling with Fluorescent Dyes
Fluorescent dyes were obtained from Lumiprobe (Lumiprobe, Hunt Valley, MD, USA). For the sake of visualization, DARPin, EC1-LoPE or proteoliposomes complexes (DARP-Lip(Bn)) lacking visible fluorescence were covalently labeled with Cyanine 3.5 (N-hydroxysuccinimide ester) dye. For the whole-body fluorescence image Bn or EC1-LoPE were conjugated with Cyanine 5.5. For the fluorescence histological analysis before incorporation into liposomes Bn was conjugated with Cy3 and EC1-LoPE—with Cy5.5.
For the labeling with cyanine dyes, a few grains of the dye were dissolved in water and the concentration of the solution was determined spectophotometrically. Eight-fold molar excess of the dye solution was used for conjugation with protein or proteoliposomes complexes. The reaction was carried out in 0.1 M phosphate buffer (pH 8.5) for 1 h at room temperature to get the dye-conjugate (protein or proteoliposome). The dye-labeled conjugate was separated from the unbound dye on NAP-5 desalting column (GE Healthcare, Little Chalfont, UK) equilibrated in 0.1 M phosphate buffer (pH 7.5). The void volume fraction was collected and used in flow cytometry or bioimaging experiments.
4.6. Cell Culture
The human breast ductal carcinoma BT-474 (HTB-20™; ATCC) and MDA-MB-231 (human breast adenocarcinoma, ATCC no. HTB-26) were maintained in DMEM medium (HyClone, Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS, HyClone) and 2 mM L-glutamine (PanEko, Moscow, Russia). Cells were cultured in humidified atmosphere with 5% CO2 at 37 °C.
4.7. Cell Cytotoxicity
The cytotoxicity of DARP-Lip(Bn) and EC1-LoPE were investigated using standard MTT assay [50
]. Cells were seeded on a 96-well plate at 104
cells per well in 100 μL of DMEM supplemented with 10% FBS and cultured overnight. Proteoliposomes DARP-Lip(Bn) or EC1-LoPE at different concentration were added to wells in 100 μL of DMEM and cells were incubated for 72 h. Then the medium was removed and MTT solution (0.5 g/L in 100 μL of DMEM) was added to the cells. Samples were incubated for 2 h at 37 °C, MTT solution was then removed and 100 μL of DMSO was added to the wells, the plate was gently shaken until the formazan crystals dissolved completely. The optical density of each well was measured using an Infinite 1000 Pro microplate reader (Tecan, Mannedorf, Zurich, Switzerland) at a wavelength of λ = 570 nm.
4.8. RNAse Activity
Ribonuclease activity of Bn was tested on yeast RNA by the acid-insoluble RNA precipitation assay as described in [30
]. Protein solution in 40 μL buffer (0.125 M Tris-HCl, pH 8.5) was mixed with 160 μL of yeast RNA at 2 g/L and incubated at 37 °C for 15 min. The RNAse reaction was stopped by adding 200 μL of 6% HClO3
and incubating at 4 °C for 15 min. Undigested RNA was separated by centrifugation at 16,000 × g for 10 min at 4 °C and absorption (OD260
) of supernatant was measured. According to [30
], this value is proportional to the RNAse activity of the tested protein.
4.9. Flow Cytometry
Cells (2 × 105) of each cells line were treated with 50 nM protein or 0.5 nM proteoliposome dye-labeled conjugate for 10 min at 37 °C in complete medium. After that the cells were washed 3 times with PBS and analyzed with flow cytometer NovoCyte 3000 (Bioscience, San Diedo, CA, USA). Cy3.5 fluorescence was excited with a laser at 561 nm and the emission of the dye was detected using a 615/20 bandpass filter. Autofluorescence level was determined on protein-untreated cells. A minimum of 50 000 events was collected for each sample and the data were processed with NovoExpress software.
4.10. Measurements of Proteoliposomes Size and Zeta-Potential
For size and zeta-potential experiments, proteoliposomes were diluted 1:1 with PBS. All the measurements were performed using a Zetasizer Nano ZS (Malvern Panalytical, Malvern, UK) at room temperature. Experiments were made in triplicate and the results are given as means ± standard deviation.
4.11. Tumor-Bearing Mice
Six- to eight-week-old female BALB/c nu/nu athymic mice (22–25 g) were purchased from the SPF (specified pathogen-free) licensed nursery of the Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry of the Russian Academy of Sciences. Mice were housed in specific-pathogen-free facilities with a 12-h light-dark cycle, fed with sterilized enriched laboratory food and supplied with water and food ad libitum. All experimental procedures were approved by the Animal Care and Use Committee of the Institute.
For tumor generation mice were inoculated subcutaneously with 2 × 106
cells BT-474 stably expressing NanoLuc luciferase
]. Inoculation of cancer cells was performed in 30% Matrigel (Corning, NY, USA) in culture medium. Tumor size was measured with Vernier calipers across two diameters every other day and the tumor volume (V) was calculated according to the formula: V = length × width2
4.12. In Vivo Tumor Imaging
To demonstrate the tumor-targeting capabilities of DARP-Lip(Bn) and EC1-LoPE in living animals, tumor-bearing athymic BALB/c Nude mice were administered intraperitoneal (i.p.) with proteoliposomes (10 nM) or fusion toxin (40 µg) conjugated with Cy5.5. The isoflurane inhalation for anesthesia was performed using the RAS-4 Rodent Anesthesia System (Perkin Elmer, Waltham, MA, USA). Living animal were than visualized using the IVIS Spectrum CT imaging system (Perkin Elmer, Waltham, MA, USA) with the following excitation filters: 640, 675 and 710 nm and emission filters: 680, 700, 720, 740, 760 and 780 nm, followed by Living Image software processing and autofluorescence subtraction using a Living Image Spectral Unmixing Tool. As a control of uniform settings for all acquired images, a tube containing 10 nM DARP-Lip(Bn-Cy5.5) or EC1-LoPE-Cy5.5 was imaged along with the animals.
4.13. Histological Analysis
To investigate the microdistribution of DARP-Lip(Bn) and EC1-LoPE in tumors, before loading into liposomes Bn was conjugated with Cy3 and EC1-LoPE was conjugated with Cy5.5. Tumor-bearing mice were injected i.p. with EC1-LoPE-Cy5.5. DARP-Lip(Bn-Cy3) were injected i.p. 19 h later. 6 h after liposomes injection, the mice were euthanized and the tumors with the surrounding healthy tissues were excised. The fresh derived tumors were mounted on a platform with Neg-50™ frozen section medium (Richard-Allan Scientific™, Canton, MI, USA) and rapidly frozen at -26 °C. Sections of frozen tissues were performed on a ThermoFisher device using the standard method. Frozen cross sections, 15 μm thick, were placed on glass slides, air-dried and cell nuclei were additionally stained with 2 nM Hoechst 33,342 (Invitrogen, Carlsbad, CA, USA) for 10 min at 25 °C. Then the sections were washed twice with PBS, air-dried, and mounted under a cover glass in glycerol.
Sections were analyzed immediately after preparing under an LSM-710-NLO confocal microscope (Carl Zeiss, Oberkochen, Germany) using 514 nm laser for excitation Cy 3.0 and 633 nm laser for Cy 5.5. Fluorescence was detected in the range of 550–630 nm for Cy3.0 and in the range 650–740 nm for Cy 5.5. Hoechst was exited at 740 nm using femtosecond laser Chameleon Ultra II (80 MHz, 140 fs, tunable in 690–1060 nm, Coherent Inc., Santa Clara, CA, USA) and the emission was detected at 400–550 nm. To acquire the images, a 20× Plan-Apochromat (NA 0.8) objective was used.
4.14. In Vivo Tumor Therapy
Twelve days after the inoculation, when tumor volumes reached ~100 mm3, twenty animals were randomly split into 4 groups. Treatment was carried out every other day according to the following schedules: 1) i.p. injection of 100 μL of 10 nM (~70 μg Bn) DARP-Lip(Bn), 10 doses; 2) i.p. injection of 40 μg EC1-LoPE (i.p.), 10 doses; 3) i.p. injection of 100 μL of 10 nM DARP-Lip(Bn) and 40 μg EC1-LoPE (i.p.), 10 doses; 4) i.p. injection of 100 μL PBS (control group). The tumor sizes and body weights of mice were controlled every other day. Mice were euthanized when tumors reached a volume of ∼1000 mm3. To calculate tumor growth inhibition coefficient (TGI), the following formula was used: TGI (%) = [(Vcontrol-Vtreatment) × 100%]/Vcontrol, where V is the tumor volume at a selected time point.
4.15. Bioluminescent Imaging in Vivo
In vivo bioluminescence imaging was performed on IVIS Spectrum CT system (Perkin Elmer, Waltham, MA, USA). The BT-474/NanoLuc bearing mouse was placed in an anesthesia chamber and isoflurane inhalation was performed using the RAS-4 Rodent Anesthesia System (Perkin Elmer, Waltham, MA, USA). Furimazine was administered i.p. at a dose of 7 μg per animal, the mouse was then moved to the IVIS Spectrum CT imaging chamber, and data acquisition started one minute after the introduction of the substrate. All bioluminescence data were normalized to the acquisition conditions and are displayed in radiance (photons/s/cm2/str).
The data (MTT-test and tumor volume measurements) were statistically processed using one-way analysis of variance (ANOVA).