Recombinant nanobody against MUC1 tandem repeats inhibits growth, invasion, metastasis, and vascularization of spontaneous mouse mammary tumors.

Alteration in glycosylation pattern of MUC1 mucin tandem repeats during carcinomas has been shown to negatively affect adhesive properties of malignant cells and enhance tumor invasiveness and metastasis. In addition, MUC1 overexpression is closely interrelated with angiogenesis, making it a great target for immunotherapy. Alongside, easier interaction of nanobodies (single-domain antibodies) with their antigens, compared to conventional antibodies, is usually associated with superior desirable results. Herein, we evaluated the preclinical efficacy of a recombinant nanobody against MUC1 tandem repeats in suppressing tumor growth, angiogenesis, invasion, and metastasis. Expressed nanobody demonstrated specificity only toward MUC1-overexpressing cancer cells and could internalize in cancer cell lines. The IC50 values (the concentration at which the nanobody exerted half of its maximal inhibitory effect) of the anti-MUC1 nanobody against MUC1-positive human cancer cell lines ranged from 1.2 to 14.3 nm. Similar concentrations could also effectively induce apoptosis in MUC1-positive cancer cells but not in normal cells or MUC1-negative human cancer cells. Immunohistochemical staining of spontaneously developed mouse breast tumors prior to in vivo studies confirmed cross-reactivity of nanobody with mouse MUC1 despite large structural dissimilarities between mouse and human MUC1 tandem repeats. In vivo, a dose of 3 µg nanobody per gram of body weight in tumor-bearing mice could attenuate tumor progression and suppress excessive circulating levels of IL-1a, IL-2, IL-10, IL-12, and IL-17A pro-inflammatory cytokines. Also, a significant decline in expression of Ki-67, MMP9, and VEGFR2 biomarkers, as well as vasculogenesis, was evident in immunohistochemically stained tumor sections of anti-MUC1 nanobody-treated mice. In conclusion, the anti-MUC1 tandem repeat nanobody of the present study could effectively overcome tumor growth, invasion, and metastasis.

antibody drug conjugates

anti‐histidine

basic fibroblast growth factor

complementary DNA

complementarity‐determining regions

distilled water

Dulbecco's modified Eagle's medium

dimethylsulfoxide

ethylenediaminetetraacetic acid

enzyme‐linked immunosorbent assay

estrogen receptor

antigen‐binding fragment

fluorescence‐activated cell sorting

fetal bovine serum

Food and Drug Administration

fluorescein isothiocyanate

variable fragment

gastrointestinal

granulocyte‐macrophage colony‐stimulating factor

hematoxylin and eosin staining

human epidermal growth factor receptor 2

horseradish peroxidase

inhibitory concentration 50

immunohistochemistry

interleukin

immobilized metal affinity chromatography

isopropyl β‐D‐1‐thiogalactopyranoside

kinase insert domain receptor

lysogeny broth

monoclonal antibodies

matrix metallopeptidase 9

3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5‐diphenyltetrazolium bromide

mucin1

microvessels density

nuclear factor kappa‐light‐chain‐enhancer of activated B

nickel‐nitrilotriacetic acid

optical density

phosphate buffered saline

platelet endothelial cell adhesion molecule‐1

propidium iodide

progesterone receptor

prostate‐specific membrane antigen

polyvinylidene fluoride

real‐time polymerase chain reaction

single‐chain variable fragment

sodium dodecyl sulfate polyacrylamide gel electrophoresis

spontaneous mouse mammary tumor

Tris/EDTA Sucrose

tumor infiltrated lymphocytes

triple negative breast cancer

tumor necrosis factor alpha

vascular endothelial growth factor

vascular endothelial growth factor receptor2

single variable domain on a heavy chain

Introduction

At the turn of the third millennium, cancer is ranked among the top three leading causes of death and an important hurdle to increase life expectancy in each nation worldwide. Based on the World Health Organization’s (WHO) evaluation in 2019, cancer is ranked as the first or second cause of mortality in two‐third of the world’s countries and the third or the fourth in the rest [1]. More importantly, the highly complex identity of the disease has turned the process of cancer therapy to a time‐consuming struggle with relatively low succession rate. So far, surgical resection of the affected tissue and radiation therapy of the remaining biomass in sequence with administration of a systemic chemotherapy regimen form the routine strategy for cancer therapy. Currently available antineoplastic agents consist of DNA intercalating agents (e.g., cisplatin and doxorubicin), DNA alkylating agents (e.g., chlorambucil and busulfan), antimetabolites (e.g., 5‐FU and methotrexate), microtubule‐destabilizing agents (e.g., taxans), hormonal therapies (e.g., Triptorelin), and molecular targeting agents (e.g., sorafenib and pazopanib) [2]. Nevertheless, severe nonselective toxicity profile of these agents, as well as of a limited clinical efficacy owing to a rapid process of resistance acquisition, highly restricts beneficial outcomes associated with chemotherapeutic agents’ application in clinic and result in failure of cancer treatment. Tackling these drawbacks, discovering new targets and establishing novel treatment modalities with higher selectivity toward malignant cells seems crucial [3, 4, 5, 6].

Generally, in tumor cells, growth promotory signals and underlying pathways are highly overexpressed while growth inhibitory signals are either attenuated or completely inhibited. In addition, these cells can evade apoptosis, replicate limitlessly, induce angiogenesis, and develop invasion and metastasis. Assuming genetic alteration as the source for these abnormalities, detecting aberrantly expressed biomarkers, specific to cancer cells may be the key for development of cancer specific therapies. The ideal tumor specific antigen for targeted immunotherapy should be abundantly expressed in the tumor cells, but scarcely observed or not exist in normal tissues. In addition, targeted antigen should play an important role in tumor progression, angiogenesis, invasion, or metastasis. One such invaluable antigen is MUC1 mucin, a high molecular weight O‐glycosylated protein, constitutively expressed by ductal epithelial cells of several organs such as breast, lung, pancreas, and gastrointestinal tract. Structurally, MUC1 is made from two main domains, each with specific functional properties [7]. Tandem repeats of extracellular section of MUC1 are highly O‐glycosylated during normal physiological condition. However, this unique pattern becomes strongly altered during carcinoma, making tandem repeats accessible to the immune system and antibodies. Thus, many researchers consider tandem repeats of tumor‐associated MUC1 (tMUC1) as an ideal cancer antigen [8].

Monoclonal antibodies (mAbs) are a group of novel therapies capable of selectively targeting cancer antigens and disrupting underlying pathways [9]. During the last few decades, numerous mAbs have been raised against a range of cancer‐specific antigens and some of them have succeeded in acquiring United States of America Food and Drug Administration (FDA) approval for the treatment of specific malignancies. For instance, anti‐HER2 mAb, trastuzumab (Herceptin®) has received FDA approval for treatment of metastatic breast cancer overexpressing HER2 antigen [10] or antivascular endothelial growth factor (VEGF) antigen mAb, and bevacizumab (Avastin®) has received FDA approval in treatment of metastatic renal cell carcinoma in combination with interferon gamma administration [11]. However, large molecular size of antibodies restricts their effective penetration to the tumor site [12, 13]. Furthermore, the unique morphology of mAb’s recognition site, comprising of two variable domains, noncovalently attached via hydrophobic bonds, makes development and engineering of these biological therapeutics very difficult. Thus, a new format of mAb with smaller size, less complex structure, improved stability and better in vivo pharmacodynamics is highly recommended [14]. Antigen‐binding fragment (Fab), variable fragment (Fv) and single‐chain variable fragment (scFv), either derived from natural sources or prepared synthetically are some of these newly developed antibody formats. Nevertheless, suboptimal effectiveness and low affinity toward antigen have mostly limited their application in clinic [15]. Additionally, newly developed synthetic molecules and protein scaffolds including affibodies, DARPins, and minibodies are still in preclinical experimental settings and no clinical data exist so far, addressing their superiority compared to conventional mAbs or potency to induce immunological responses [14].

Discovery of the camelid heavy‐chain antibodies lacking the light chains of conventional antibodies in the early years of twenty‐first century was the breakthrough in antibody‐bioengineering field [16]. Filling only 1/10 of the conventional antibodies volume, the heavy‐chain variable domain of these ‘nonconventional’ antibodies (VHH or Nanobody®, named by its developing company, Ablynx), acts as a completely functional binding fragment per se. Nanobodies retain full functionality against their target molecules and bind them with similar affinity compared to conventional mAb. Trivial size of nanobodies and high affinity and specificity toward antigen make them ideal antibody format for application in clinic [17]. Considering that nanobodies do not require post‐translational modifications and are encoded by merely a single gene, they can be easily recombinantly expressed in micro‐organisms and cost‐effectively manufactured utilizing microbial fermentation [18]. In addition, contrary to conventional mAbs, they are highly water soluble and less inclined for aggregation in aqueous solutions. This, along with their small size, allows them easy access to the antigen’s buried epitopes in membrane [19, 20].

Different studies have demonstrated that high levels of tMUC1 expression are associated with higher risk of invasiveness and metastasis of all types of breast, as well as ovarian, prostate, gastric, liver, and pancreatic cancers [21]. This is partly due to the fact that tMUC1 interferes with cell–cell and cell–extracellular matrix interactions, facilitating detachment of malignant cells from their primary site [8]. Angiogenesis, the process of blood vessels sprouting from pre‐existing ones, also plays pivotal role in tumor progression, invasion, and metastasis. Unfortunately, rapid development of resistance to currently existing anti‐angiogenic therapies has mostly restricted application of these agents in clinic, and thus, development of new agents with anti‐angiogenic characteristic is of great importance [22, 23]. Interestingly, multiple studies have now revealed that tMUC1 overexpression in different types of neoplasms is also associated with increased expression of numerous angiogenic factors and aggressive behavior of tumor [24]. These findings encouraged us to target tandem repeats of tMUC1 with recombinant nanobodies. Here, we reported facile expression of a recombinant single‐domain antibody against Tandem Repeat Region of MUC1in Escherichia coli and evaluated its preclinical efficacy for suppressing tumor growth, angiogenesis, invasion, and metastasis.

Materials and methods

Bacteria strains and cell culture

Cloning and gene expression were performed using E. coli strain NovaBlue GigaSingles™ (Catalog Number: 71227, Novagen, Madison, WI, USA) as initial cloning host and BL21 (DE3) (Catalog number: 69450, Novagen) as expression host. pET‐32 LIC vector (Catalog number: 69076, Novagen) was used for cloning and gene expression studies.

Human breast cancer cells (MCF‐7, T47D, and MDA‐MB231) and PC3 prostatic cancer cells were purchased from Pasteur Institute of Iran (IPI, Iran). Other human cancer cells (SW742, HEPG2, and A549) and normal MCF‐10A cell lines were obtained from Iranian Biological Resource Center (IBRC, Iran). Human skin fibroblast cell line was a kind gift from Dr. Behnam Sadeghi.

Whole cancer cell lines used in present study were cultured in a high Glucose Dulbecco's modified Eagle's medium (DMEM; Catalog number: 12430047, Gibco, Amarillo, TX, USA) supplemented with 10% fetal bovine serum (FBS) (Catalog number: 26140079, Gibco), 2 mm L‐glutamine (Catalog number: 25030081, Gibco), 1 mm sodium pyruvate (Catalog number: 11360070, Gibco), 1% nonessential amino acid (Catalog number: 11140050, Gibco), and 1% penicillin‐streptomycin (Catalog number: 15070063, Gibco) in gamma radiated sterile polystyrene flasks and maintained in a 5% CO2 humidified atmosphere at 37 °C. Moreover, MCF‐10A cell line and primary dermal fibroblasts were cultured in DMEM/F12 Ham’s mixture medium (Catalog number: 11320033, Gibco) according to the supplier’s recommended growth and culturing instructions.

Recombinant anti‐MUC1 nanobody modeling and protein–protein docking

The detailed process of construction and screening of anti‐MUC1 single‐domain antibody libraries has been described elsewhere [25]. The 3D structure of the designed nanobody was modeled using ABodyBuilder©, an established program for translating nanobody sequence in to structural prototype for subsequent antibody‐antigen docking studies [26]. In the next step, the InterEvDock2 Web‐based program was applied to determine the binding sites of the designed antibody to the SAPDTRPAPG sequence of MUC1 variable number tandem repeat (VNTR) regions [27]. In this server, the structure of protein–protein complexes are first determined by FRODOCK rigid‐body docking program and then, based on the InterEvScore [28], FRODOCK [29], and SOAP‐PP potentials [30], 10 most probable models are chosen from the large pool of created decoys [29]. Noteworthy, InterEvScore is a scoring system for docking proteins based on a combination of two‐ and three‐body statistical potentials, together with scores of interface contacts deduced from multiple sequence alignments (MSAs). Applying this method for integrating evolutionary information is practically superior to those merely accounting for conserved positions [28].

Quantification of MUC1 expression in cells by Real‐time Polymerase Chain Reaction (RT‐PCR)

Evaluation of the MUC1 mRNA transcription in T47D (breast triple positive cancer cell line), MCF7 (breast ER+/PR+ cancer cell line), MDA‐MB‐231 (breast triple negative cancer cell line), SW742 (human colon cancer cell line), A549 (human lung cancer cell line), PC3 (human prostatic cancer cell line), HepG2 (human liver cancer cell line), and MCF‐10A (normal breast) cell lines, as well as the normal human skin fibroblast cells, was assessed by real‐time PCR technique. In brief, total RNA of at least 106 cells from each cell line was extracted using GenElute™ total RNA purification kit (Catalog number: RNB100, Sigma, St. Louis, MO, USA). 2–5 μg of total RNA from each cell line was then retro‐transcripted into complementary DNA (cDNA) using RevertAid® First Strand cDNA Synthesis Kit (Catalog number: K1622, Thermo Fisher Scientifics, Waltham, MA, USA) and used as template for RT‐PCR. Quantitative analysis of MUC1 expression was carried out according to a previously established protocol [31]. In brief, a set of primers (forward primer, 5' ‐agacgtcagcgtgagtgatg‐3'; reverse primer, 5' ‐gacagccaaggcaatgagat‐3') was designed for MUC1 and blasted to confirm their specificity. Preliminary PCRs with β‐actin (housekeeping gene)‐specific primers (forward primer, 5’‐cagcagatgtggatcagcaag‐3’ and reverse primer, 5’ ‐gcatttgcggtggacgat‐3’) were carried out to normalize the procedure (Table S1). RT‐PCR was then performed using a RealQ Plus Master Mix Green (Catalog number: A324402, Amplicon, Brighton, UK) and a LightCycler® 96 Instrument (Catalog number: 05815916001, Roche, Mannheim, Germany). The total volume of the reaction mixture was 20 μL, containing 10 μL of 2× SYBR Green master mix, 0.8 μL of each primer (10 μm), and 2 μL cDNA (0.01 μg·μL−1). All samples were analyzed twice, and in cases of over 10% variation, a third run was applied. MUC1 transcription level in each cell line was expressed compared to that of normal skin fibroblast primary cell, considering β‐actin as the house keeping gene and using the formula . Cells were considered MUC1 overexpressing if the value of −∆∆CT was higher than zero, MUC1 expressing if the value ranged between 0 and −10, and MUC1 negative, if the value was higher than −10.

Designation of anti‐MUC1 nanobody expression cassette

After identification of the coding sequence of nanobody, a stII signal peptide coding sequence was introduced to the 3' end of the anti‐MUC1 nanobody coding sequence. This results in localization of nanobody in periplasmic space of E. coli which can be simply isolated in its native form through disrupting outer membrane of bacteria. A six constitutive histidine tag coding sequence was introduced at the 5' end of the anti‐MUC1 nanobody coding sequence for accelerating purification by Nickel‐Nitrilotriacetic acid (Ni‐NTA) reagent kit (Catalog number: ACR5000NT, Amicon®, Merck KGaA, Darmstadt, Germany; Fig. 1B). Full‐length sequence was then codon optimized for expression in E. coli and synthetized by Cinnagen Inc. (Tehran, Iran).

Fig. 1

Anti‐MUC1 nanobody was successfully cloned, expressed, and purified in E. coli and predicted to interact with MUC1 VNTRs through specific amino acids located in its CDRs. (A) The coding sequence for anti‐MUC1 nanobody was retrieved from GenBank, and its structural prototype was determined using ABodyBuilder©. CDR regions are enclosed in green boxes. (B, C) Modeling data obtained by InterEvDock2 (http://bioserv.rpbs.univ‐paris‐diderot.fr/services/InterEvDock2/), representing possible interacting motifs of anti‐MUC1 nanobody with SAPDTRPAPG sequence of MUC1 VNTRs. Arg54, Tyr57, Tyr58, Tyr101, Lys63, Thr55, and Thr67 residues of anti‐MUC1 nanobody play the most prominent role in formation of correct interaction between the nanobody and selected MUC1 region. Note that the Thr56 residue of nanobody forms two hydrogen bonds with selected sequence of MUC1 VNTR. (D) A schematic representation of anti‐MUC1 nanobody expressing cassette, comprising a stII signal peptide, the anti‐MUC1 nanobody coding sequence and a 6× His‐tag moiety included in the pET‐32 EK/LIC vector. (E) Results of colony PCR confirming correct transformation of bacterial cells; 1: DNA ladder; 2: negative control (water); 3: positive control (pET32/LIC vector containing anti‐MUC1 nanobody coding sequence); 4–6: positive colony. Experiment was performed in triplicate. (F) SDS/PAGE analysis confirmed enhancement of anti‐MUC1 nanobody expression at ~ 18 kDa following induction with IPTG. 1: Protein Marker; 2: IPTG noninduced; and 3: IPTG‐induced expression of nanobody. Experiment was performed in triplicate. (G) SDS/PAGE analysis of purified anti‐MUC1 nanobody demonstrating a single band at ~ 18 kDa for five different purified samples. 1: protein marker; 2–5: purified anti‐MUC1 nanobody. Note that the protein ladder used herein was different from the one in panel F. Experiment was performed in triplicate. (H) A single positive blot at predicted position in western blot confirmed correct identity of nanobody. 1: protein ladder; 2: negative control; 3: purified anti‐MUC1 nanobody. Experiment was performed in triplicate.

Cloning, expression, and purification of recombinant anti‐MUC1 nanobody

Codon optimized sequence was amplified with PCR using a set of primers as depicted in Table 1 and then purified and concentrated with a gel extraction kit (Catalog number: 28506, Qiagen, Hilden, Germany). Achieved amplicons were cloned into the LIC site of the pET‐32 EK/LIC vector according to the manufacturer’s instructions and used to transform NovaBlue GigaSingles™ competent cells. The recombinant clones were then randomly selected and underwent colony PCR and sequencing for confirming successful cloning. Recovered plasmids from positive colonies were used to transform protein expression strain, BL21 (DE3) competent cells. Freshly transformed single colonies were then inoculated in to 5 mL lysogeny broth medium (LB medium, Catalog number: 12795027, Thermo Fisher Scientifics) containing 50 µg·mL−1 ampicillin and incubated in a shaking incubator (~ 180 rpm) at a range of temperatures (27, 30, and 37 °C) to reach an optical density (OD)600 of 0.4–0.6. At this point, bacteria were induced with various concentrations of isopropyl β‐D‐1‐thiogalactopyranoside (IPTG, Catalog Number: I6758, Sigma), different temperatures, and incubation periods to obtain the optimal expression. Afterward, cells were harvested by centrifugation at 9000  for 20 min and subjected to periplasmic extraction using cold osmotic shock method. Pellets were resuspended in ice‐cold 1× Tris EDTA Sucrose buffer (TES Buffer; 0.2 m Tris/HCl pH 8.0, 0.5 mm EDTA, and 0.5 m sucrose; Catalog number: MB‐058, Rockland Immunochemicals, Pottstown, PA, USA) and incubated for 3 h on ice. Subsequently, 2 volumes of 1 : 4 diluted TES in dH2O were added and incubation continued overnight on ice. After centrifugation at 1008  for 30 min at 4 °C, the supernatant comprising His‐tagged nanobody and other soluble periplasmic proteins was subjected to immobilized metal affinity chromatography (IMAC). Based on results, the optimum temperature yielding highest expression was chosen for purification.

Table 1

Sequence of primers used for amplification of genes in RT‐PCR.

Gene name	Primer sets (5′ → 3′)
MUC1	Forward: AGACGTCAGCGTGAGTGATG Reverse: GACAGCCAAGGCAATGAGAT
β‐actin	Forward: CAGCAGATGTGGATCAGCAAG Reverse: GCATTTGCGGTGGACGAT

Subsequent to the equilibration of Ni‐NTA column (5 mL volume) with 10 column volumes of PBS pH 8 (equilibration buffer), equilibrated supernatant with 0.3 m NaCl and 5 mm imidazole was added to the column with a flow rate of 1 mL·min−1 and washed applying 20 column volumes of washing buffer (equilibration buffer together with 30 mm imidazole, pH 8). In the next step, bound anti‐MUC1 nanobody was eluted using 5 column volumes of elution buffer (equilibration buffer with 300 mm imidazole, pH 8). Elution fractions were then subjected to analysis by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS/PAGE) and immunoblotting.

Characterization of anti‐MUC1 nanobody by SDS/PAGE and immunoblot analysis

Fractions of IPTG‐induced, IPTG noninduced, and purified anti‐MUC1 nanobody were electrophoresed on a 12% SDS/PAGE under reducing condition. Gels were then visualized by Coomassie brilliant blue staining. Resolved proteins by SDS/PAGE were further transferred to a polyvinylidene fluoride membrane (PVDF; Catalog number: GE10600023, Amersham®, Merck) by means of a semidry Trans‐blot transfer device (Catalog number: 1703848, Bio‐Rad, Hercules, CA, USA) according to the manufacturer’s protocol. The PVDF membrane was then incubated overnight in blocking buffer (5% w/v skim milk (Merck) in 0.05% v/v PBS‐Tween‐20 solution) and probed with mouse horseradish peroxidase (HRP) conjugated anti‐histidine (anti‐His) tag mAb 1:2000 (Catalog number: ab1187, Abcam, Boston, MA, USA). The signals were developed using 3,3′‐diaminobenzidine tetrahydrochloride (DAB; Catalog number: sc‐216567A, USA) solution according to previously established protocols.

Ethical approval and ethical standards

The project was found to be in accordance with the ethical principles and the national norms and standards for conducting medical research in Iran (Approved ID: IR.ACECR.IBCRC.1397.001) and evaluated by Motamed Cancer Institute‐Academic Centre for Education, Culture and Research. This institution performed its reviews based on United States Public Health Service (USPHS) regulations and applicable federal and local laws.

Cell line authentication

Human breast cancer cells (MCF‐7, T47D, and MDA‐MB231) and PC3 prostatic cancer cells were purchased from Pasteur Institute of Iran (IPI, Iran). Other human cancer cells (SW742, HEPG2, and A549) and normal MCF‐10A cell lines were obtained from Iranian Biological Resource Center (IBRC, Iran). Human skin fibroblast cell line was a kind gift from Dr. Behnam Sadeghi. Mentioned cell lines are routinely applied in research performed in Motamed Breast cancer institute and are subjected to Mycoplasma contamination test every 2 months using Mycoplasma Detection Kit (Mycoalert®, Lonza, Basel, Switzerland). The identity and purity of each cell line were validated prior to the initiation of project and during experiment by frequent performance of short tandem repeats (STR) profiling.

Cell surface antigen‐binding assay

Cell surface antigen‐binding capacity of nanobody was evaluated using a fluorescence‐activated cell sorting (FACS) calibur flow cytometer device (Becton Dickinson, Franklin Lakes, NJ, USA) at FL1 channel. High MUC1‐expressing T47D and A549 cell lines, along with low MUC1‐expressing HepG2 and MUC1‐negative PC3 cell lines, were incubated with 3 nm solution of recombinant anti‐MUC1 nanobody at 4 °C for 1 h. Cells were then thoroughly washed with PBS and Incubated with 10 μg·mL−1 fluorescein isothiocyanate (FITC)‐labeled anti‐His‐tag‐specific secondary antibody (Catalog number: ab1206, Abcam) at 4 °C for another 1 h. Cells were again thoroughly rinsed, resuspended in PBS solution, and subjected to flow cytometry.

Anti‐MUC1 nanobody internalization studies

Confocal microscopy was applied for the determination of anti‐MUC1 nanobody internalization pattern in very highly and highly MUC1‐expressing T47D and MCF‐7 cancer cell lines. Briefly, cells were incubated at 37 °C and 5% CO2 overnight. On the following day, cells were washed with PBS and equilibrated with 30 ng·mL−1 solution of anti‐MUC1 nanobody for 30 min at 4 °C to initiate binding and attachment of nanobody. Excess unbound nanobodies were then washed away by PBS, and for initiation of internalization, cells were re‐incubated at 37 °C for different time periods (15, 45, and 90 min). Both cell lines were then fixed, permeabilized, and stained with FITC‐labeled anti‐6×His‐tag antibody (the source of green fluorescence). Nuclei were then stained with 4′,6‐Diamidino‐2‐phenylindole dihydrochloride (DAPI; Catalog number: D9542, Sigma). An increase in nanobody internalization was followed by monitoring the increase in mean green fluorescent intensity as a function of incubation time until reaching to an equilibrium where no significant increase in fluorescent was observable.

At the end of each time point, cells were fixed with 4% paraformaldehyde and permeabilized using 0.5% Triton X‐100. Internalized antibodies were then visualized following 30‐min incubation with FITC‐labeled anti‐His‐tag mAb. Cells were then washed and imaged on a Nikon A1 confocal laser scanning microscopy (Tokyo, Japan). The time‐course of increase in mean fluorescent intensity of cells was used for evaluation of anti‐MUC1 nanobody’s internalization in cancer cells and equilibrium state was defined as the state where no increase in fluorescent intensity was observed.

Cell viability assay

The in vitro cytotoxicity of anti‐MUC1 nanobodies was assessed on breast, pancreas, prostate, and liver carcinoma cell lines, as well as normal human fibroblast and normal breast cells. Briefly, 5 × 103 cells were seeded in each well of 96‐well micro titer plates, inoculated with 0.2‐mL medium, and incubated at 37 °C and 5% CO2 for 24 h. Cells were then treated with different concentrations of anti‐MUC1 nanobody (200 pm to 10 µm) for 48 h and subjected to (3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5‐diphenyl‐2H‐tetrazolium bromide) (MTT; Catalog number: M2003, Sigma) viability assay [32]. Briefly, cells were incubated with MTT dye for 3 h at 37 °C, rinsed with PBS, and formed formazan crystals that were dissolved in dimethylsulfoxide (DMSO). Absorbance was read at 570 nm using a microplate reader (Bio‐Rad), and IC50 values were defined as the concentration from nanobody that induced a 50% inhibitory effect on cell growth.

Apoptosis assay

Early and late apoptosis or the percentage of apoptotic cells were determined using FITC‐labeled Annexin V/propidium iodide (PI) apoptosis assay kit (Catalog number: 640914, Biolegend, San Diego, CA, USA) according to the manufacturer’s protocol. 4 × 105 cells were seeded into 6‐well plate and divided into two groups: the nontreated control and the anti‐MUC1 nanobody treated group (1, 2, and 3 nm). These concentrations were chosen according to the average of doses resulting in 25%, 50%, and 75% growth inhibitory effects in different studied cell lines. After 48‐h incubation at 37 °C in 5% CO2 humidified atmosphere, cells were detached using 1× trypsin‐EDTA (Catalog number: 25200056, Gibco) and washed by PBS. Cells were then pelleted and resuspended in 100 μL binding buffer of apoptosis assay kit and stained with FITC‐labeled Annexin V and PI. The flow cytometry analysis (BD FACSCalibur®) was carried out on FL1/FL3 channels. Single‐positive FL1 (Annexin V labeled cells) and double‐positive FL1/FL3 cell populations were considered as early and late apoptotic cell population, respectively. Then, total percentage of apoptotic cells was calculated by summing up the percentage of these two subpopulations.

Animal source

A female Balb/c mouse bearing spontaneous mouse mammary tumor (SMMT) and 34 normal Balb/c mice were purchased from Iranian institute of Pasteur (Karaj, Iran).

In vivo Breast cancer tumor model

A female 4‐week‐old Balb/c mouse bearing spontaneous mouse mammary tumor (SMMT) and 34 normal 7‐week‐old Balb/c mice were purchased from Iranian institute of Pasteur (Karaj, Iran). Plastic cages were utilized for housing animals (four mice/cage) in an environment with adjusted temperature of 25–27 °C and a 12 h light/dark cycle. Acclimation of animals was performed for 7 days prior to initiation of the experiment, while food and water were available ad libitum. The project was found to be in accordance with the ethical principles and the national norms and standards for conducting medical research in Iran (Approved ID: IR.ACECR.IBCRC.1397.001) and evaluated by Motamed Cancer Institute‐Academic Centre for Education, Culture and Research. This institution performed its reviews based on United States Public Health Service (USPHS) regulations and applicable federal and local laws.

Spontaneous mouse mammary tumor is an invasive ductal carcinoma, which is instinctively formed in female BALB/c mouse following transplantation of tumor cells [33]. In brief, the primarily formed SMMT tumors were allowed to reach an approximate size of 500 mm3. Then, they were carefully separated, dissected into small pieces with dimensions < 5 mm, and, subsequently, transplanted to 6‐ to 8‐week‐old healthy syngeneic Balb/c mice by surgery. 12 days after transplantation, when tumors reached to the size of 150–200 mm3, a dose of 3 mg·kg−1 from anti‐MUC1 nanobody was injected systemically (i.e., intravenously) to each mouse and repeated every 3 days for a total of seven injections. To evaluate the effect of treatment on tumor growth delay, tumor volume was measured at predetermined intervals according to the following formula: where a, b, and c represent height, width, and length (mm) of the tumor, respectively. The results were then compared to the PBS receiving groups.

Anti‐MUC1 nanobody cross‐reactivity with murine tumor‐associated MUC1

Prior to in vivo studies, cross‐reactivity of the expressed anti‐MUC1 nanobody against mouse tumor‐associated MUC1 was assessed by immunostaining of established tumors in mice. A 500 mm3‐sized SMMT tumor was sectioned and used for assessing immunoreactivity of anti‐MUC1 nanobody with SMMT tumors. In brief, a 500‐mm3 intact primary SMMT was excised from untreated mice, formalin‐fixed, and paraffin‐embedded according to the established guidelines. Tumor specimens were then sliced into 6‐µm‐thick sections and fixed on silane‐coated slides. For staining of prepared slices, each section was first de‐paraffinized in xylene and then rehydrated in a graded serial dilution of ethanol in water. For quenching endogenous peroxidase activity, sections were immersed in 3% hydrogen peroxide in PBS solution for 30 min. Antigen retrieval was carried out in citrate solution (pH 6.0), and blocking was performed with normal goat serum for 30 min. Thereafter, sections were incubated with PBS, recombinant anti‐MUC1 nanobody (0.2 ng·mL−1), or irrelevant recombinant Llama anti‐green fluorescence protein (GFP) nanobody (0.2 ng·mL−1; Catalog number: HPAB‐0342CQ, Creative Biolabs, New York, NY, USA) as negative control for 2 h at room temperature. Sections were then carefully washed with PBS and incubated with horseradish peroxidase (HRP)‐conjugated goat anti‐Llama IgG (Catalog number: A160‐100P, Bethyl Laboratories, Montgomery, UK) for 1 h. Immunoreactivities were illustrated using 3,3ʹ‐diaminobenzidine, and counterstaining was performed with Gill’s hematoxylin (Catalog number: GHS132, Sigma). Imaging from samples was performed using an Eclipse E‐200 Nikon microscope.

In vivo small animal ultrasound imaging

B‐mode ultrasound images were obtained using an ultrasound imaging device (Micromaxx Sonosite, Bothell, WA, USA), equipped with a small animal ultrasound imaging system. The transducer was initially installed on a rail system and then aligned to the tumor’s central plane utilizing a micromanipulator measurement system. Tumor’s height, width, and length were then evaluated using an electronic caliper on them. Ultrasound images were also used for analyzing extend of tumor’s growth, spreading, and invasion.

Cytokine/chemokine profile measurement by multiplex Enzyme‐Linked Immunosorbent Assay (ELISA) array

Alterations in the expression of major pro‐ and anti‐inflammatory cytokines and chemokines in response to treatment with injected anti‐MUC1 nanobody were evaluated by multiplex ELISA kit (Catalog Number: MEM‐004A, Qiagen, Hilden, Germany), and the procedure was performed according to the previously published method [34]. Immediately after completion of treatment period, mice peripheral blood was collected plasma was separated by centrifugation at 2500  for 10 min and kept at −80 °C until further use. For analysis, 50 µL of thawed samples was added to each well of their respective row. The ELISA procedure was continued according to the manufacturer’s instructions, and at the end, absorbance of the 96‐well plate was read at 450 and 570 nm utilizing a plate reader. For correction of autofluorescence, the 570 nm absorbance was subtracted from final absorbance readings. Expression of each cytokine and chemokine was reported as the fold increase/decrease compared to PBS receiving group levels.

Hematoxylin and eosin staining, immunohistochemistry, and quantification of microvessel density

At the end of the 24‐day treatment period, tumors equivalent in size were separated by autopsy and prepared for immunohistochemical staining using anti‐Ki‐67 monoclonal antibody (ab15580, Abcam), a marker for tumor proliferation; anti‐matrix metalloproteinase 9 (MMP9) monoclonal antibody (Catalog number: ab38898, Abcam), an enzyme involved in inducing angiogenesis and metastasis; anti‐VEGF receptor 2 (VEGFR2) monoclonal antibody (Catalog number: ab2349, Abcam), a receptor with pro‐angiogenic activity upon stimulation; anti‐CD31/platelet endothelial cell adhesion molecule‐1 (PECAM‐1) monoclonal antibody (Catalog number: ab24590, Abcam), recognizing PECAM‐1 expressed on surface of endothelial cells; anti‐CD8 monoclonal antibody (Catalog number: ab209775, Abcam), a specific marker for cytotoxic T cells; anti‐CD4 monoclonal antibody (Catalog number: ab221775, Abcam), a specific marker for helper T cells; and their specific horseradish peroxidase‐conjugated secondary antibody. Counterstaining of all sections was performed with Gill’s hematoxylin. For evaluation of microvessel density, sections were scanned under low magnification for the determination of area with highest vessel density following staining with anti‐CD31 antibody. This area is referred as hotspot. Afterward, under the magnification of ×400, vessel counting was performed. MVD scores were expressed as the number of microvessels per mm2. Each stained endothelial cell, either alone or in the form of clusters, was considered as a single microvessel, even if vessel lumen was absent [35]. Hematoxylin and eosin staining was also performed according to the previously established protocols to evaluate the number of tumor infiltrated lymphocytes (TILs) and the extended of response to therapy [34].

Statistical analyses

The effect of anti‐MUC1 nanobody treatment on survival was evaluated using Kaplan–Meier method and compared applying log‐rank test. graphpad prism software (version 8.0, GraphPad Software) was used for data analysis. Data are reported as means ± standard deviation (SD). Student's t‐test was used for comparing serum levels of cytokines, tumor sizes at the end of the treatment period, intensity of DAB stains in IHC stained sections, number of TILs, and MVD between groups and assessing statistical significance. P‐values lower than 0.05 were considered as statistically significant.

Results

Using AbodyBuilder, the 3D structure of recombinant anti‐MUC1 nanobody was drawn and CDR regions were determined according to IgBLAST analysis (Fig. 1A). Targeted molecular docking depicted successful interactions between recombinant anti‐MUC1 nanobody CDR regions and MUC1 tandem repeats (Fig. 1B,C). The Arginine residue at position 54 and two tyrosine residues at positions 57 and 101 of anti‐MUC1 nanobody, all located in CDRs, play the most prominent role in formation of correct interaction between the nanobody and the selected region of MUC1’s tandem repeats. Furthermore, the threonine residue at position 56 of the anti‐MUC1 nanobody could form a hydrogen bond with MUC1 tandem repeats, making mentioned interaction more compatible.

Construction, Expression, and characterization of the anti‐MUC1 nanobody

For anti‐MUC1 nanobody expression, full‐length anti‐MUC1 nanobody coding sequence was cloned in to the pET32/LIC vector (Fig. 1D) and colony PCR analysis was performed for screening positive colonies bearing the desired construct (Fig. 1E). Based on results, constructs were successfully incorporated in to the clones and subsequent sequencing analysis demonstrated no rearrangement or point mutation in isolated plasmids (Data are not presented herein).

The optimum expression condition for recombinant nanobody was determined to be induction with 0.5 mm IPTG and incubation at 27 °C for 4 h. Extracted recombinant nanobodies from periplasmic space of the IPTG‐induced bacteria were then compared with those of none‐induced bacteria applying 12% SDS/PAGE. Figure 1F represents the occurrence of a band with an apparent molecular weight of ~ 18 kDa. This molecular weight is consistent with the anti‐MUC1 nanobody and the six His‐tag moiety in addition to the enterokinase recognition sequence located in the vector. Purified anti‐MUC1 nanobody demonstrated only a single band with similar molecular weight, proposing that the large part of the expressed anti‐MUC1 nanobodies is presented in their monomeric form (Fig. 1G). This band was further validated by western blotting (Fig. 1H), supporting the fact that the induction and translation of the recombinant protein occurred correctly in the prokaryotic expression system.

MUC1 expression in human cancer cell lines

The MUC1 mRNA transcription patterns of the studied primary normal cells, as well as the other cancerous cell lines, are depicted in Fig. 2A. Considering MUC1 transcription in primary dermal fibroblasts as the reference value, T47D, MCF‐7, and MDA‐MB‐231 cell lines demonstrated a 64‐fold, 1.25‐fold, and 1.15‐fold higher MUC1 transcription values compared to the primary fibroblasts, making them to be categorized as MUC1‐overexpressing cell lines. Contrarily, MUC1 transcription in SW742 colon cancer and A549 lung cancer cell lines was approximately 103 and 108 times lower than that of fibroblast primary cells, making them to be classified as MUC1‐expressing cell lines. Finally, considering MUC1 transcription in HEPG2 and PC3 cell lines, a 1012‐ and 1015‐time lower expression values than reference cell was reported, making them to be subgrouped as MUC1‐negative cell lines.

Fig. 2

Anti‐MUC1 nanobody can selectively bind with MUC1‐positive cancer cells and internalize in a MUC1 expression‐dependent manner. (A) MUC1 mRNA expression pattern is different among various cancerous and normal cell lines. In this context, T47D and MCF7 breast cancer cell lines are MUC1‐overexpressing, while HepG2 and PC3 cell lines were MUC1‐negative. Experiments were performed in triplicate (n = 3), and data are represented as mean ± SD. (B1) Flow cytometry analysis revealed an intensive staining for highly MUC1‐expressing T47D cell lines (B2) intermediate staining for A549 cell lines moderately expressing MUC1 (B3) low staining for HepG2 cell line, expressing MUC1 in low amounts and (B4) almost negligible staining for PC3 cell line, negative for MUC1, with anti‐MUC1 nanobody. (C) Anti‐MUC1 nanobody internalization pattern in T47D and MCF7 cell lines. These cell lines were chosen as both could highly, albeit in significantly different amounts, express MUC1 antigen and become stained with anti‐MUC1 nanobody (data not shown), which will guarantee internalization of nanobody in both. Whole experiments in this section were performed in triplicate. Scale bars: 100 μm for T47D and 10 μm for MCF7.

Anti‐MUC1 nanobody binding assay

To check the specific binding of recombinant anti‐MUC1 nanobody to the MUC1, flow cytometry analysis was performed using MUC1‐overexpressing T47D, MUC1‐expressing A549, and MUC1‐negative HepG2 and PC3 cancer cell lines. As shown in Fig. 2B1,B2, MUC1‐overexpressing T47D and MUC1‐expressing A549 cell lines demonstrated high rates of binding and positive staining with recombinant anti‐MUC1 nanobody. Contrarily, very low rates of binding and staining were associated with MUC1‐negative HepG2 and PC3 cancer cells (Fig. 2B3,B4). These data are highly representative of recombinant anti‐MUC1 nanobody specifically interact and bind with MUC1‐positive cancer cells.

Anti‐MUC1 nanobody‐mediated internalization assay

The extend of anti‐MUC1 nanobody internalization in MCF‐7 and T47D cell lines was analyzed by examining enhancement in cancer cells fluorescent intensity at predetermined time points (15, 45, and 90 min). At the end of each time point, fluorochrome‐labeled secondary anti‐His‐tag antibody was applied for visualization of the fraction of nanobodies remaining on the cell surface. As depicted in Fig. 2C, internalization began very soon and was significantly high even within the first 15 min postincubation with anti‐MUC1 nanobody. For T47D cell line, internalization of nanobody increased as the incubation time prolonged and reached to a maximum at 45‐min time point. No significant enhancement in internalization was observed as the incubation time proceeded from 45 to 90 min, proposing that the nanobody internalization reached to a plateau level within the first 45 min of incubation. Contrarily, in MCF7 cell lines, fluorescent intensity continued to increase up to the end of the 90‐min incubation period. This confirms that anti‐MUC1 nanobody’s internalization takes place in a tMUC1 expression‐dependent manner.

Effect of recombinant anti‐MUC1 nanobody on human tumor and normal cell apoptosis

Apoptosis inducing potency of nanobody in different cell lines was evaluated using annexin‐V/FITC binding assay and flow cytometry. As depicted in Fig. 3A–D, similar to the results of cellular viability assay, for a given concentration from nanobody, total percentage of apoptosis in T47D cell line was significantly higher compared to the other studied cell lines. Apoptosis was also considerably high in MCF7 cell lines following treatment with 2 nm solution of anti‐MUC1 nanobody. Contrarily, the apoptotic rate in low MUC1‐expressing HepG2 and MUC1‐negative PC3 cell lines was either low or negligible. Thus, same concentrations from anti‐MUC1 nanobody interacted more strongly with cells expressing higher number of MUC1 and more potently induced apoptosis. Treating MCF‐10A normal epithelial cell line expressing normal MUC1 could not induce apoptosis or inhibit cellular growth, representing selective interaction of nanobody with truncated MUC1 (tMUC1) and toxicity toward cancer cells.

Fig. 3

Anti‐MUC1 nanobody at nanomolar concentrations could effectively inhibit cancer cell’s growth and induce apoptosis in a MUC1 expression dependent manner. (A) Evaluated by counting Annexin V+/PI− and Annexin V+/PI+ populations in flow cytometry, increasing concentration from nanobody significantly enhanced total number of apoptotic cells in studied cancer cell lines. However, increase in apoptosis was significantly lower in cells with lower MUC1 content including HepG2 and PC3 cell lines. Higher tMUC1 content of cancer cells resulted in enhanced activation of other co‐localized transmembrane receptors including ICAM‐1, E‐selectin, EGFR, and erbB which together inhibit induction of apoptosis. (B) Viability of each cell line following treatment with respective concentration from anti‐MUC1 nanobody. Anti‐MUC1 nanobody could effectively induce cell death in MUC1‐overexpressing and moderately expressing cancer cell lines (P < 0.001). Cell death in T47D cell line was significantly higher compared to the other studied cell lines, *P < 0.05 and ***P < 0.005 following treatment with 2 and 5 nm, respectively. Data were analyzed using two‐way ANOVA followed by Bonferroni’s post hoc test. Data are representative of mean ± SD of each group. (C) Concentration–response curves of anti‐MUC1 nanobody for a number of cancerous cell lines. IC50 values also differed in a MUC1 expression‐dependent manner (P < 0.001). T47D was the most sensitive cell line, while PC3 and HepG2 cell lines were slightly or almost sensitive to anti‐MUC1 nanobody. Normal MCF‐10A and human skin fibroblast cells were insensitive to anti‐MUC1 therapy, further confirming selectivity of anti‐MUC1 nanobody against tMUC1. IC50 values for each cell line have been provided adjacent to their corresponding graphs. For resistant cells, determination of IC50 values was not possible at studied range of concentrations. Data are representative of mean ± SD of each group. Two‐way ANOVA followed by Bonferroni’s post hoc test was used for analysis of results. (D) Apoptotic cell death trends were also similar to those observed with MTT assay in studied cell lines. Note that concentrations required for induction of apoptosis were higher compared to the ones demonstrating inhibitory effects. Data are represented as mean ± SD. Two‐way ANOVA followed by Bonferroni’s post hoc test was used for analysis of results. Whole experiments in this section were performed in triplicate.

The cell growth inhibitory effects of anti‐MUC1 nanobody on T47D, MCF7, MDA‐MB‐231, SW742, A549, PC3, HEPG2, and MCF‐10A cell lines and human normal skin fibroblast is shown in Fig. 3C. Cells were treated with recombinant anti‐MUC1 nanobody for 48 h with concentrations ranging from 200 pm to 10 µm, and cell viability was evaluated with MTT reduction assay. The IC50 values of anti‐MUC1 nanobody for each cell line have been depicted in Fig. 3D. The most significant inhibitory effect of nanobody on cellular viability was observed against T47D cell line, followed by MCF7, MDA‐MB‐231, SW742, and A549 cell lines. Contrarily, inhibitory effect of expressed anti‐MUC1 nanobody on HepG2 and PC3 cell lines was either very low or negligible.

Evaluation of anti‐MUC1 nanobody’s selective cytotoxicity against tMUC1‐expressing cells

To further confirm that observed cytotoxic effects were selective for tMUC1‐expressing cancer cells, MCF‐10A cell line and normal human skin fibroblast cells were treated with increasing concentrations of nanobody. As depicted in Fig. 3C, micromolar concentrations from nanobody (about 103 time higher than those inducing growth inhibitory effects in tMUC1‐expressing cells) could only induce 20–30% reduction in cellular growth of MCF‐10A and normal human skin fibroblast cells. Thus, IC50 values of nanobody for these cells are predicted to be much higher than the range of concentrations applied in this study. Similarly, no significant apoptotic responses were observable with MCF‐10A normal cell line following treatment with applied concentrations from nanobody. These results strongly suggest selective cytotoxicity of the anti‐MUC1 nanobody against tMUC1‐expressing cancer cells but not tMUC1‐negative cancer cells or normal MUC1‐expressing cell lines.

Staining mouse tumor sections with anti‐MUC1 nanobody and irrelevant recombinant Llama anti‐GFP nanobody demonstrated a high reactivity for anti‐MUC1 nanobody toward mouse MUC1 but a very low or negligible one for irrelevant recombinant Llama anti‐GFP nanobody (Fig. 4a–d). This confirmed that the anti‐MUC1 nanobody of present study is also capable of binding with mouse MUC1 antigen, and thus, SMMT model can be used for evaluating nanobody’s tumor suppressing effects in vivo. In addition, negligible staining of sections, observed with PBS and recombinant Llama anti‐GFP nanobody, ignored probability of nonspecific binding of secondary antibody or anti‐MUC1 nanobody to mouse MUC1 antigen, respectively.

Fig. 4

Administration of anti‐MUC1 nanobody in tumor‐bearing mice significantly delayed tumor growth and improved survival without inducing significant cytotoxicity. (A) Staining mouse tumor sections with anti‐MUC1 nanobody demonstrated a high cross‐reactivity for anti‐MUC1 nanobody (b) but not recombinant Llama anti‐GFP nanobody (c) with MUC1 of mouse origin. Semiquantified results obtained by imagej software (Bethesda, MD, USA) analysis have been depicted in (d). No significant positive signals were detected by incubating sections with (a). Tumor sections were incubated with same concentrations of two nanobodies (0.2 ng·mL−1) or PBS and then washed and incubated with secondary HRP‐conjugated goat anti‐Llama IgG antibody. Five fields were acquired and intensity of brown stain (correlated with the extent of marker expression) was obtained using imagej software. Data are represented as mean ± SD for each group and compared using two‐tailed unpaired Student's t‐test. ***P < 0.005 compared to recombinant Llama anti‐GFP nanobody or PBS incubated section (n = 5, P < 0.0001 for both). No significant differences were observed between Llama IgG antibody treated and PBS incubated sections (n = 5, P = 0.4520). Scale bars are equal to 200 µm. (B) Administration of 3 mg·kg−1 anti‐MUC1 nanobody to a group of 7 Balb/C mice, during a 21‐day period follow‐up did not significantly affect body weight and induce any signs of gross toxicity reviewed in the literature. Results were analyzed using unpaired Student’s t‐test at each time interval (for the day 21 post‐treatment, P = 0.7084). Data are represented as mean ± SD. (C) Mean size of the tumors at the end of treatment period was compared using independent Student's t‐test and was significantly lower in group of mice receiving anti‐MUC1 nanobody (n = 7, P = 0.0087). Data are represented as ***P < 0.001 compared to PBS group. (D) Box plot graph demonstrating distribution of tumor weights at the end of therapy. Similarly, the mean weight was significantly lower in group of mice receiving anti‐MUC1 nanobody (n = 6, P = 0.043); *P < 0.05 compared to PBS group. The mean weight of the tumors at the end of therapy was compared using independent Student's t‐test. Note that one mouse was died from each group. Therefore, data have been represented for six mice. (E) Survival analysis performed by Kaplan–Meier method and compared with log‐rank analysis demonstrated a significant increase in survival rates in mice receiving anti‐MUC1 therapy (n = 10 in each group, P = 0.032).

In vivo tumor growth suppression in response to anti‐MUC1 nanobody therapy

Delay in tumor growth was evaluated over a 24‐day period following administration of anti‐MUC1 nanobody with concentration equal to 3 µg·g−1 in comparison with PBS receiving group. Depicted in Fig. 4B, no significant decline (more than 20%) in weight of mice occurred throughout the treatment, confirming that the administered concentration from nanobody was safely tolerated. As shown in Fig. 4C, administration of anti‐MUC1 nanobody could significantly enhance suppression of tumor growth and delay logarithmic phase of tumor growth. Also, the mean weight of tumors at the end of the treatment period was significantly lower for anti‐MUC1 nanobody receiving group compared to PBS group (Fig. 4D). In this context, differences in tumor size became significant from the day 15 post‐treatment and progressively increased up to the end of treatment. Figure 4E represents the results of Kaplan–Meier survival analysis between control and nanobody treated groups, which represents a significant increase in overall survival of mice treated with anti‐MUC1 nanobody.

In parallel, Fig. 5A,B schematically represent the angle between the direction of ultrasound waves and position of the mouse for obtaining transverse and longitudinal scans required for measuring dimensions of tumor (x, y, z). The red hemisphere in both panels is representative of tumor’s spatial direction. Obtained ultrasound images from tumors demonstrated significantly higher rate of invasion and risk of metastasis in PBS group compared to anti‐MUC1 receiving group. In brief, inhomogeneity in structure of tumors, as well as development of waved‐shaped boundaries and scars in ultrasound images, is related to the more invasive and metastatic behavior of tumor. Contrarily, observation of an ovule form tumor with smooth boundaries is representative of a less invasive tumor [36]. As depicted in Fig. 5C, tumors in most of the PBS receiving group mice were inhomogeneous and rough, and represent waved‐shaped boundaries. Contrarily in anti‐MUC1 nanobody, tumors were in most cases smooth and in ovulated form. Furthermore, occurrence of invasion in PBS receiving mice has been shown by white arrow in picture on day 15. Such incidence was not observed in anti‐MUC1 receiving group.

Fig. 5

Ultrasound images confirmed delayed tumor growth in mice receiving anti‐MUC1 nanobody. (A) Schematically representation of transverse and (B) longitudinal scans obtained for measuring dimensions of tumor (x, y, z) and formula used for calculation of volume. (C) Ultrasound scans from tumors obtained on days 0, 15, and 21 of treatment demonstrated a significant reduction in dimensions of tumor. Furthermore, invasion occurred on day 15 in mouse receiving vehicle (white arrow). What’s more, inhomogeneous and waved‐shaped (not in ovule form as for nanobody receiving group) tumors of PBS receiving group represents a more invasive and metastatic behavior.

Immunohistochemistry, quantification of microvessel density, and H&E staining results

Performing hematoxylin and eosin staining (H&E) method, significantly higher rate of response to therapy in anti‐MUC1 nanobody receiving group was observed compare to the control group (Fig. 6Aa,Ab). In this context, counting number of lymphocytes based on the morphology of their nuclei by an expert pathologist, blinded from the experimental procedure, a significantly higher number of TILs were detected in anti‐MUC1 nanobody receiving group compared to the nontreated control one (14.2 ± 1.4 and 7.3 ± 0.7, respectively, P = 0.0016; Fig. 6Ac). Analyzing angiogenesis by evaluation of microvessel’s density in control and treated group, a significant declined vascularization was spotted in anti‐MUC1 nanobody receiving group compared to the control group (11.2 ± 1.1 and 34.7 ± 3.4, respectively, P < 0.0001; Fig. 6Ad,Ae,Af). Also, as depicted in Fig. 6Ba‐k, results of IHC studies demonstrated a significant decline in the expression of MMP9 (0.1 ± 0.01 and 0.013 ± 0.001, respectively, P < 0.0001), Ki‐67 (0.073 ± 0.003 and 0.089 ± 0.004, respectively, P = 0.0098), and VEGFR2 (0.01 ± 0.002 and 0.003 ± 0.0005, respectively, P < 0.0001) antigens was observed in anti‐MUC1 nanobody receiving group in comparison with control, at the end of the 24‐day treatment period.

Fig. 6

Administration of anti‐MUC1 nanobody could significantly decrease concentration of pro‐inflammatory cytokine in plasma and downregulate tumor growth, metastasis and angiogenesis markers in vivo. (A) Panel left: (a) and (b) Administration of anti‐MUC1 nanobody could significantly increase number of TILs in tumor site (n = 5, P = 0.039) compared to control. (c) TILs were counted in five fields (×40 magnification) and reported as mean ± SD; *P < 0.05 compared to PBS group. Panel Right: (d) and (e) MVD was significantly reduced in ani‐MUC1 receiving group compared to control (n = 5, P = 0008). (f) Three areas with highest vascular density were determined for each slide and total score for each slide was reported as microvessel numbers per mm2. MVD was reported as mean ± SD of total scores of five sections; ***P < 0.05 compared to PBS group. (B) Administration of anti‐MUC1 nanobody could significantly downregulate expression of Ki‐67 (marker of proliferating cells; n = 5, P < 0.01) (a) and (f); VEGFR2 (a prominent receptor with pro‐angiogenic effect upon activation; n = 5, P < 0.001) (b) and (g); and MMP9 (a key player in angiogenesis, invasion, and metastasis; n = 5, P < 0.001) (c) and (h); while increasing number of CD4+ (marker of T helper cells) (d) and (i); and CD8+ (marker of cytotoxic T cells; n = 5, P < 0.001) (e) and (j) in tumor environment. (k) For each marker, five fields were acquired and intensity of brown stain (correlated with the extent of marker expression) was obtained using imagej software. Data were reported as mean ± SD for each marker; results were analyzed using two‐tailed unpaired student t‐test. *P < 0.05, **P < 0.01, ***P < 0.001 compared to PBS group. Scale bars are equal to 100 µm. (C) Administration of anti‐MUC1 nanobody could significantly downregulate levels of pro‐inflammatory and angiogenic cytokines including IL‐1α, IL‐2, and IL‐17α in plasma while increasing IL‐1β and GM‐CSF in mice plasma at the end of the 21‐day treatment period. Cytokine’s levels were determined for three mice in each group using sandwich ELISA and reported as mean of fold changes ± SD. Results were analyzed using two‐tailed unpaired Student’s t‐test; **P < 0.01, ***P < 0.001 compared to PBS group.

Alteration in cytokine/chemokine profile in response to anti‐MUC1 nanobody

Peripheral blood collected at the end of the 24‐day treatment period was used for the assessment of pro‐/anti‐inflammatory responses induced by anti‐MUC1 nanobody therapy using multiplex ELISA kit. As demonstrated in Fig. 6C, concentration of pro‐inflammatory and pro‐angiogenic cytokines including IL‐1a, IL‐2, IL‐6, IL‐10, IL‐12, and IL‐17A, in peripheral blood of tumor bearing mice receiving anti‐MUC1 nanobody was significantly lower compared to the PBS receiving group. However, concentrations of IL‐1β and granulocyte‐macrophage colony‐stimulating factor (GM‐CSF) were significantly raised in response to treatment. No significant change in concentration of tumor necrosis factor alpha (TNF‐α) was reportable between groups.

Discussion

In present study, we reported a facile synthesis method of a nanobody against tandem repeats of MUC1 and investigated its tumor suppressive, anti‐angiogenic, and anti‐metastatic effects in vivo and in vitro. MUC1 tandem repeats on epithelia of normal tissues are heavily O‐glycosylated and covered by long branched glycans. Nevertheless, during carcinomas, these long‐chain glycans are substituted with simpler and shorter ones and are expressed in hypo‐glycosylated manner. Different studies have reported that elevation in MUC1 expression level is directly associated with higher risk of invasion and poor prognosis in breast, colon, pancreas, and bladder cancers. Furthermore, MUC1 contributes to the angiogenesis, tumor growth, and development of metastasis [8, 21, 24]. Recently, it has been shown that tandem repeats of MUC1 are capable of activating nuclear factor Kappa‐light‐chain‐enhancer of activated B (NF‐κB), a transcription factor involved in pro‐inflammatory responses, induction of resistance to chemotherapy, tumor progression, invasion, and metastasis [37, 38, 39, 40]. This becomes more important considering that tumor‐associated MUC1 is usually expressed in higher levels during carcinomas and is in its hypo‐glycosylated form. Therefore, tandem repeats of MUC1 protein backbone are more easily and abundantly accessible for the development of new protein–protein interactions, which can result in activation of several signaling pathways, further accelerating tumor growth, angiogenesis, invasion, and metastasis. Consequently, blocking tandem repeats of MUC1 is theoretically an effective way for suppressing tumor propagation [24].

Results presented herein confirm successful production of a periplasmic expressed recombinant nanobody with preserved capacity to bind with MUC1 mucin. The differential binding pattern of anti‐MUC1 nanobody upon interaction with T47D and A549 cell lines confirmed specific binding of anti‐MUC1 nanobody with surface associated MUC1. Furthermore, as correct ligation of anti‐MUC1 antibodies to the epitopes of MUC1 induces internalization of the antibody, similar observations with our nanobody during its incubation with T47D and MCF‐7 cell lines further confirm the specific binding of the nanobody to MUC1 tandem repeats [41, 42]. Also, internalization of the anti‐MUC1 nanobody constructed in current study allows further application of it in production of immunotoxins and antibody drug conjugates (ADCs), which can specifically deliver toxic agents inside tumor cells. Anti‐HER2 and anti‐prostate‐specific membrane antigen (PSMA) antibodies have been successfully applied with this purpose in cancer immunotherapy [43].

tMUC1 is also expressed on surface of Cancer stem cells (CSCs), a specific group of cells, located in tumor niche, which is hypothesized to be the main inducer of tumor recurrence and development of metastases [44, 45]. Whereas this may be an extra advantage, fast rate of clearance from bloodstream as a consequence of nanobodies small molecular size (usually less than 60 kDa, the renal filtration sieve threshold) extremely shortens their half‐life and interaction duration with CSC’s surface antigen in vivo [46, 47]. To address this challenging issue, pharmaceutical scientists are now focusing on dimerization or trimerization of nanobodies (e.g., in the case of caplacizumab), fusing them in their monomeric form with an albumin‐targeting nanobody moiety [48, 49, 50], or administering then with tumor penetrating peptides, which can enhance vascular and tissue permeability through augmentation of endocytosis pathways [51]. In the latter case, fusion of anti‐EGFR nanobodies with iRGD tissue penetrating peptide effectively enhanced anti‐tumoral activity of the nanobody up to a therapeutically meaningful level while also demonstrating synergistically enhanced activity upon co‐administration with chemotherapeutic drugs, T cells, and nanoparticles [52, 53, 54]. Moreover, connecting anti‐EGFR to lactoferrin could effectively enhance its inhibitory activity [55]. In the present study however, to overcome drawbacks associated with low half‐life of anti‐MUC1 nanobody used in its monomeric form, administration time intervals were chosen as short as possible and injection concentrations were kept at their maximal safe and nontoxic levels, determined based on our pilot screening study.

Although changes in glycosylation pattern of MUC1 antigen can alter the affinity and reactivity of antibodies to MUC1 antigen, since the recombinantly expressed nanobody of present study was the same as the one developed in Camelus bactrianus, receiving a homogenized blend of pancreatic and breast cancer tissues, as well as ascites of small cell lung cancer tissues of Iranian adults and two synthetic peptides including TSA‐P1‐24 TSAPDTRPAPGSTAPPAHGVTSAPDTR and A‐P1‐15 APDTRPAPGSTAPPAH as boost up doses [25], the nanobody could have already reacted with hypo‐glycosylated form of MUC1 associated with different types of human cancers, as well as their glycan free peptide epitopes. Moreover, smaller skeletal size of the nanobodies and unique structure of their CDR regions endow them the ability of freely accessing the unreachable sites of the MUC1 antigen and detecting small amino acid sequences as epitope, which are not usually recognized by conventional antibodies or engineered Fab and scFv fragments [56]. Hence, even upon alteration of glycosylation pattern, no changes in complexation of antigen: nanobody will occur.

After confirming the binding ability of the constructed recombinant anti‐MUC1 nanobody to MUC1, efficacy and functionality of the nanobody against variety of cancer cell lines was evaluated with MTT and flow cytometry‐based apoptosis assays. RT‐PCR analysis demonstrated that MUC1 is diversely expressed in different cell lines. Therefore, it is not surprising to observe a range of IC50 values for anti‐MUC1 nanobody with different studied cell lines. The nanomolar range of IC50 values (Table 2) for different studied cell lines represents the high affinity of the constructed nanobody toward its target cell. This also confirms correct folding and expression of nanobody, as unfolding or misfolding of the nanobody during expression or purification process reduces potency of the nanobody.

Table 2

IC50 values of anti‐MUC1 tandem repeats nanobody for different cancer cell lines.

Cell line	IC50 (nm)
T47D	1.2
SKBR3	4.16
MCF7	7
SW742	7.2
MDA‐MB‐231	10
A549	14.3

Cytotoxicity and pro‐apoptotic activities of anti‐MUC1 nanobody on MUC1‐overexpressing cancer cell lines were significantly higher compared to the MUC1‐expressing and MUC1‐negative cell lines. Contrarily, despite of the relatively high expression levels of MUC1 in MCF‐10A and fibroblast normal cell lines, a low apoptotic activity was observed in these cells, which may be attributed to the different glycosylation pattern of MUC1 in normal cells compared to the cancerous one. The shorter and lower in density O‐glycan chains of cancer‐associated MUC1 facilitates reaching of even low levels of antibodies to the protein core of the MUC1. Contrarily, hyper‐glycosylated state of MUC1 in normal MCF‐10A and fibroblast cell lines significantly hampers accession of the nanobody to the targeted epitope [57].

Several mechanisms have been proposed for in vitro and in vivo pro‐apoptotic effects associated with targeting MUC1 glycoprotein with anti‐MUC1 antibody. The loss of cell polarity during cancer progression, and consequently, disorganization of components of cellular membrane results in displacement of MUC1 glycoprotein from apical surface in to whole membrane surface, which, in turn, triggers co‐localization of it with other types of transmembrane receptors including ICAM‐1, E‐selectin, EGFR, and erbB, as well as ECM components, which were not accessible before. In addition, cancer associated form of MUC1 is unique in that it is underglycosylated, and therefore, its core peptides have become accessible for interaction with these receptors. The results of such interactions have shown to be associated with induction of invasion, migration, metastasis, angiogenesis, and inhibition of apoptosis, which have been addressed in numerous reports and review papers [8, 21, 58]. Considering this in mind, blocking of MUC1 core peptide interaction with adjacent receptors and suppression of subsequent induction of cellular signaling pathways may be the main mechanism through which the anti‐MUC1 nanobody results in inhibition of proliferation as well as induction of apoptosis in vitro.

Considering in vivo studies, additional mechanisms may also be involved in observed apoptotic effects. For instance, recently, it has been shown that mAb AR20.5 (OncoQuest Inc., Edmonton, AB, Canada) can demonstrate vaccine like effects and initiate specific immune responses in patients. Through administration of low doses of this antibody, some antibody will react with circulating antigens in blood stream and immune complexes will be formed which can be internalized by dendritic cells and will subsequently active T lymphocytes. These activated cells will in next place internalize in to tumor environment and eradicate tumor cells [59]. We hypothesize that following administration of anti‐MUC1 nanobody in to the blood stream of Balb/c mice, these nanobodies will bound with circulating shed MUC1 glycoproteins and form complexes, which can be more effectively become recognized by antigen‐presenting cells (APCs). Following processing of this complex with APCs, cytotoxic T lymphocytes will become activated in response to interconnection with APCs through MHC type I and will transmigrate to tumor site where they can recognize Muc1‐expressing tumor cells and induce apoptosis and induce killing of tumor cells. Alongside, due to their stability and very small sizes, nanobodies can also directly and in great amounts penetrate in to tumor site and directly reacted with MUC1 on cancer tumor cells and inhibit proliferation, migration, invasion, angiogenesis, metastasis of cancer cell, and induce apoptosis through inhibition of signaling cascades.

Following confirmation of anti‐MUC1 nanobody’s cross‐reactivity with mouse MUC1, administration of this nanobody in SMMT demonstrated a significant tumor growth suppressive effect. A number of factors including the affinity of antibody to the expressed MUC1 on different cell lines and the density of antigen on these cells determines the potency of antibody in killing of antigen‐positive targeted cells [25]. Furthermore, it has been hypothesized that binding of mAbs to MUC1 results in either blockade or stimulation of specific types of cell membrane molecules, which, in turn, mediate inhibition of tumor growth. In another proposal, it has been hypothesized that in the presence of MUC1‐specific mAbs, MUC1 and EGFR may be alternatively trafficked and moved in to lysosomal degradation pathway, which subsequently boost their degradation. These may explain in part why targeting MUC1 signaling pathways with anti‐MUC1 nanobody may significantly suppress cancer proliferation, migration, and metastasis [57, 60, 61].

Application of anti‐MUC1 nanobody could effectively decline serum concentrations of pro‐inflammatory and pro‐angiogenic cytokines downstream of the NF‐κB pathway, including IL‐1a, IL‐2, IL‐10, IL‐12, and IL‐17A. Multiple studies have shown that alteration in pattern of glycosylation affects different functions of MUC1. Most importantly, aberrant glycosylation of MUC1 triggers its endocytosis and accumulation in intracellular compartment, which has been proposed to change its function in intracellular signaling pathways in cancer cells [62]. Expression of MUC1 with tandem repeats has shown to be linked with enhanced chemotoxicity of the innate immune system cells. It has been shown that the tandem repeats of MUC1 play pivotal role in expression of pro‐inflammatory cytokines through activation of NF‐κB pathway and thereby inducing inflammation and progression of cancer [37].

Activation of VEGF receptor 2 via interaction with VEGF protein promotes expression of specific group of matrix metalloproteases, including MMP9, which can initiate degradation of matrix, allowing for sprouting of endothelial cells [63]. Consequently, loss of MMPs from endothelium or inflammatory cells can significantly inhibit angiogenesis. A recent study has shown that MUC1 expression is strikingly correlated with specific groups of angiogenic factors including vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), and platelet derived‐endothelial cell factor, as well as group of receptors specific for angiogenic factors including kinase insert Domain Receptor (KDR) and bFGF receptor 2 [64]. In addition, a direct correlation between MUC1 expression and angiogenesis has been reported in prostate cancer. Our IHC results demonstrated that targeting MUC tandem repeats with anti‐MUC1 nanobody results in significant decline in expression of MMP9 and VEGFR2, proposing an anti‐angiogenic and anti‐metastatic effect for nanobody. Ki‐67 expression is a great marker associated with tumor growth and proliferation, and is widely used as a predictive marker for assessment of patient’s biopsies [65]. A significant decline in expression of Ki‐67 marker in treatment group also proposes the tumor suppressive activity of anti‐MUC1 therapy.

Conclusion

Although present study demonstrated that application of nanobodies against tandem repeats of MUC1 could be an effective way to suppress tumor growth, angiogenesis, invasion, and metastasis, a number of limitations exists in present study. First of all, the in vivo study was not performed on transgenic mice that overexpress human MUC1 in the mouse mammary gland (MMTV‐MUC1 transgenic mice). These mice express human MUC1 gene in a similar pattern to those observed in human organs and develop reproducible spontaneous mammary gland tumors, which can undergo metastasis. They also possess an intact immune system and express a targetable and stable tumor antigen MUC1 [66]. Second, blockade of angiogenesis and metastasis was only assessed by evaluating expression of surface markers. Further in vitro studies including blockade of tube formation assay and Boyden chamber invasion/migration assays seems critical. Third, determination of tumor suppressive effects of nanobody was demonstrated for one administrated concentration and on minimum number of animals. Studying efficacy of other concentrations from nanobody on larger population will be highly informative for confirmation of present studies results. Finally, the blockade of cross talks between MUC1 and related molecular pathways of apoptosis, angiogenesis, and metastasis, as another proof for claims of present study, was not examined herein. Regardless of these limitations, the significant inhibitory effect of anti‐MUC1 nanobody on expression of MMP9, VEGFR2, pro‐inflammatory, and angiogenic cytokines may partly explain therapeutic effects observed with anti‐MUC1 nanobody therapy in vivo. In addition, low IC50 values and high tumor‐specific cytotoxicity of this nanobody further guarantee its safety for application in future clinical studies.

Conflict of interest

The authors declare no conflict of interest.

Author contributions

PM wrote manuscript and analyzed data; BD conducted part of the experiments and wrote the manuscript; NJ, MRE, ASK, SMK, MS, MM, and MSB performed experiments and obtained data, KMA conceived and designed research, FY performed bioinformatic section of the study, and FR involved in analysis of the data; LF contributed in design of the study, supervised the whole project, and provided materials and devices required for the study.

Peer review

The peer review history for this article is available at https://publons.com/publon/10.1002/1878‐0261.13123.

Table S1. List of primers used in present study.

Click here for additional data file.