Sialic acid-binding lectin-modified fructose-coated nanoparticles: a promising targeted therapeutic synthetic for breast cancers

Background: Sialic acid-binding lectin (cSBL) specifically kills tumor cells rather than healthy cells. Glycopolymer-coated nanoparticles are selectively ingested by tumor cells because they can interact with the enriched carbohydrate receptors located on the surface of tumor cells. In this context, we synthesized cSBL-modified fructose-coated nanoparticles (LMFN) and cSBL-modified glucose-coated nanoparticles (LMGN) to investigate their anticancer activity in various molecular subtypes of breast cancer cell lines. Methods: The syntheses of fructose-coated nanoparticles and glucose-coated nanoparticles were based on the chemicals of 1,2:4,5-di- O -isopropylidene- β -d-fructopyranose and 1,2:4,5-di- O -isopropylidene- β -d-glucopyranose, respectively. The carbodiimide-based method was employed to synthesize LMFN and LMGN. The antitumor mechanism was explored by cell cycle analysis with flowcytometry and the antitumor activity was assessed by cytotoxicity assay and multiple drug effects analysis. Results: The cytotoxicity assay showed that LMFN had robust antitumor activity against all breast cancer phenotype cell lines whereas LMGN was rarely efficacious to against human epidermal growth factor receptor 2-positive/overexpression (HER2+/overexpression) breast cancer cells. The intrinsic reason for these findings was that the overexpression of fructose transporter, GLUT5, was observed in all breast cancer subtype cell lines but only a paucity of glucose transporter, GLUT1, was expressed in HER2+/overexpression breast cancer cell lines that dampened the uptake of LMGN to these cells. The cell cycle analysis indicated that the anticancer activity of LMFN was achieved by arresting cell cycle in S phase. The multiple drug effects analysis suggested the synergistic effect in the combinations of LMFN and tamoxifen to kill estrogen receptor+


Introduction
Carbohydrate is a crucial source of energy for organisms and an indispensable component which participates in the procedures of cell adhesion, cell-cell recognition and cell proliferation. Cancer cells frequently proliferate actively and grow vigorously through abnormal metabolism. Given the Warburg effect, they are apt to implement active glycolysis and produce a large amount of lactic acid even under the conditions of sufficient oxygen and normal mitochondrial function [1]. Their rapid proliferation and the Warburg effect both consume a myriad of glucose, giving rise to a low glucosemicroenvironment. Therefore, they have to harness other energy surrogates to complete their growth in this landscape; fructose has the same capability as glucose to promote their colony formation and migration [2] and can functionally replace it to maintain cell proliferation when its provision is insufficient. Fructose is the second largest sugar consumed by human body and accounts for more than 40% of sweetener consumption in Western diet [3]. With the highest sweetness among all natural sugars, it is about 1.8 times as sweet as sucrose [4]. Actually, fructose is more prone to be metabolized than glucose as its metabolism bypasses the rate-limiting enzyme of glycolytic pathway and is not controlled by insulin [5].
Recent epidemiological studies indicate that excessive intake of fructose is related to the occurrence and development of some tumors [6][7][8][9]. High intake of fructose increases the incidence rate and accelerates the progression of pancreatic cancer [10,11]. In the study of human breast cancer cell lines, it was uncovered that fructose prompted the proliferation of MCF7 and MDA-MB-231 in a dose-dependent manner beneath the glucose-efficiency condition, but had no effect on normal cells [2]. Glucose transporter 5 (GLUT5) is exclusively responsible for the fructose absorption by cells [12] and is highly expressed in several breast cancer cell lines and tumor tissues while the normal cell lines are devoid of its presentation [2]. Knockdown of it can obviously inhibit the proliferation and growth of MCF-7 cells in fructose-containing medium but not in glucose-containing medium [2].
The selective expression of GLUT5 in cancer tissues implies that GLUT5 can be used as a potential therapeutic target. An in vitro study of Zhao et al [13] demonstrated that fructose-coated nanoparticles (FCN) could complete targeted drugs delivery into MDA-MB-231 cells, thanks to the overexpression of GLUT5 found on their plasma membranes.
Cancer cells recognize fructose molecules on the surface of FCN and induce GLUT5mediated endocytosis. The specific affinity between fructose molecules and GLUT5 spurs these micelles preferentially binding to cancer cells rather than normal cells, thus enabling the drastically higher uptake rate of them by tumor cells than that by normal cells. Increasing the amount of fructose molecules on their surface can increase the uptake rate whereas blocking GLUT5 can decrease it as well as impedes the metastasis, proliferation and apoptosis of cancer cells [14,15].
GLUT1 is a kind of plasma membrane protein that widely exists on the surface of mammalian cell membrane and plays a key role in the process of glucose uptake by tissues and organs [16]. The overexpression of GLUT1 has been identified as an important hypoxia biomarker in malignant tumors and a predictor of tumor angiogenesis. Breast cancer is molecularly classified into five subtypes including Luminal A, Luminal B, Luminal-HER2+, HER2 overexpression and triple-negative breast cancer (TNBC) in terms of the expressed level of estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2(HER2) and Ki-67 [17]. Hitherto, the comparisons of GLUT1 and GLUT5 expressions across these breast cancer subtype cell lines were unclear, and it was 5 an uncertainty whether these cancer cell lines all proliferate in a dose-dependent manner in glucose-containing medium and fructose-containing medium. trastuzumab, the first anti-HER2 monoclonal antibody, kills the HER2+/overexpressed breast cancer cells efficiently because it is against the extracellular domain of HER2 protein that reduces the overexpression of HER2[24]; and tamoxifen competitively inhibits the binding of estrogen and ER, resulting in the blockade of ER-mediated signaling that is the backbone for the growth of ER + breast cancers [25]. However, it was unclear that cSBL acted synergistically or antagonistically with tamoxifen and trastuzumab in repressing the proliferation of ER + breast cancer cells and HER2+/overexpressed breast cancer cells, respectively.
The purpose of this work was to settle those aforementioned questions. To reinforce the 6 interactive activity between cSBL and tumor cells, the glucose-coated nanoparticles (GCN) and FCN were employed as the nanocarrier to synthetize cSBL-modified GCN (LMGN) and cSBL-modified FCN (LMFN). Both synthetics were respectively applied to treat various breast cancer phenotype cell lines to compare their cell killing effects across these cell lines. containing different concentrations of fructose were respectively mixed with an equal volume of 40℃, 1.2% soft agar medium. The mixtures were dripped into 6-well plates to prepare the bottom culture mediums followed by cooling and solidifying at room 7 temperature. The 40℃, 0.7% soft agar medium was added into identical volume of these fructose mediums and followed by being mixed with cell suspension. The mixtures were dropped onto the solidified bottom culture medium and the 6-well plates were placed into 5% CO 2 , 37℃ cell incubator for 10 days. Counting: The cell culture wells were stained by 0.005% crystal violet more than 1 hour. Then, 10 fields in each well were randomly selected to be observed beneath 100-fold microscope and only the cell colonies harboring cell number more than 50 were counted. The soft agar clone formation experiment with reference to glucose was implemented with the same parameters and procedures.

Materials And Methods
Western blot analysis. The harvested cells were washed with PBS followed by fully lysis with adding 4℃ RIPA lysis buffer and 4 times of vibration by an ultrasonic vibrator.
Insoluble substances were removed by a centrifugation at 10,000 rpm for 10 minutes, and protein was quantified by bicinchoninic acid. The obtained protein content was resolved by SDS-PAGE electrophoresis and transferred onto nitrocellulose membrane. The GLUT1 and GLUT5 expressions were respectively detected by correspondent rabbit primary antibody followed by overnight incubation with rabbit secondary antibody and chemiluminescence detection.
Synthesis of FCN. An appropriate proportion of 1,2:4,5-di-O-isopropylidene-β-dfructopyranose, DMAP, pyridine and 0℃ methacrylic anhydride were added into anhydrous dichloromethane. After stirring at room temperature for 48 hours, the organic layer was separated by ice-cold saturated NaHCO 3 solution. The aqueous layer was extracted with dichloromethane and the organic layer was dried with MgSO 4 followed by concentrating under reduced pressure. The product was further purified by flash column chromatography by using ethyl acetate: N-hexane (1:1) as the eluent.
In a Schlenk tube, the purified product was dissolved in 1,4-dioxane and presently AIBN and CPADB were added in and mixed up. Placing the tube in a freeze-pump-thaw for 3 cycles made the mixture be degassed, followed by a polymerization at 70℃ that would be halted with ice water after 15 hours. The polymerized solution was poured into diethyl ether for precipitation and then the harvest viscous polymer was dried under vacuum for 24 hours.
Statistical analysis. Results of measurement data were expressed as means ± SD. All experiments were repeated thrice. Using the GraphPad Prism Version 5.0 software, the statistical comparisons between two groups adopted an unpaired t-test and among multiple groups the One-way ANOVA was used followed by a Bonferroni test to compare differences across the interior-group. A P-value of < 0.05 was considered to be statistically significant. reached 7500 mg/L and 10000 mg/L (Fig. 1B). This inconsistent phenomenon of proliferation implied that the absorption and usage of fructose and glucose by these cell lines might be in diversity.

Results
HER2 + breast cancer cell lines underexpressed GLUT1. GLUT1 and GLUT5 are the glucose transporter and the fructose transporter, respectively, which embed in the cell membrane [12,16]. These 5 breast cancer subtype cell lines all overexpressed GLUT5 with no significance ( Fig. 2A-B)  LMFN synergized with tamoxifen to kill ER + breast cancer cells and with trastuzumab to kill HER2 + breast cancer cells. Herein, LMFN was continuously applied to combine with other agents for analyzing the multiple drug effects. As outlined in Fig. 5A and 5B, the combination of LMFN and tamoxifen synergistically killed MFC7 (CI m =0.54) and MDA-MB-12 361 (CI m =0.37); additionally, LMFN combined with trastuzumab had a synergistic effect on killing BT474 and SKBR3, with CI m value of 0.79 and 0.68, respectively ( Fig. 5C and 5D).

Discussion
Dietary fructose is closely associated with a variety of metabolic diseases [29,30] and is functionally alike to glucose in boosting the proliferation and metastasis of tumor cells [2].
Our work demonstrates that fructose universally prompts the clone formation of these five molecular subtypes of breast cancer cell lines in a dose-dependent manner but the high concentrations of glucose inhibit the proliferation of HER2+/overexpressed breast cancer cell lines, because the overexpression of GLUT5 exists in all breast cancer cell lines whereas GLUT1 is underexpressed in HER2 + breast cancer cell lines.
The uptake of carbohydrates by cells is dependent on the members of GLUT family that are located on their plasma membrane and encoded by SLC2A1 [31,32]. GLUT1 is the first characterized GLUT glucose transporter, particularly focusing on the constant uptake of glucose into cells through facilitative diffusion [33,34]. Research findings have shown that the elevated level of GLUT1 can be observed in myriad types of tumorigenesis procedure and has become an important hypoxia biomarker for malignant tumors [35]. With an exception, our work found that GLUT1 was lowly expressed in HER2+/overexpressed breast cancer cell lines, indicating that GLUT1 cannot be employed as an all-around predictive biomarker for these tumors. By now, it needs to further investigate whether it is a coincidence or there has some detailed mechanisms with pertinent to the HER2 + status restricting the expression of GLUT1.
GLUT5 is the only glucose transporter specifically binding to fructose [36]. Knockdown of it markedly inhibits the growth of several types of breast cancer cells in fructose-containing medium but not in glucose-containing medium [2]. In normal tissues, GLUT5 is mainly 13 presented in small intestine but is underexpressed in the brain, adipose tissue, kidney, testis and skeletal muscle. Previous study had demonstrated that a crowd of breast cancer cells, cervical cancer cells and liver cancer cells overexpress GLUT5 [2], and our work further confirms the elevated level of it in five molecular phenotypes of breast cancer cell lines, suggesting that GLUT5 overexpression can be used as a potential biomarker for predicting tumorigenesis. The high level of GLUT5 protein increases the fructose intake by tumor cells and accelerates the cancer development, thereby the dietary fructose is supposed to be limited and even abstained in cancer patients.
The antitumor activity of the lectin is principally attained by the following paths. (1) Lectin directly bind to various integrins and epidermal growth factor receptor (EGFR) on cancer cell surface and promote the internalization and autophagic degradation of these molecules, inducing caspase-8 dependent cell apoptosis [37]. (2) Lectin downregulates vascular endothelium integrins to inhibit the neovascularization in a dose-dependent manner [37]. (3) Tumor immunogenicity can be improved by lectin due to the decreased expression of B7-H4 that is a negative regulator of T cell mediated immunity [37]. (4) Lectin incurs the downregulation of STMN1 and MCM4 as well as the upregulation of WEE1, RAD1 and ATR, giving rise to the cell cycle arrest in S phase [38]. Recently, glycopolymer-coated nanoparticles have attracted great interest due to the specific binding between them and carbohydrate receptors located on the cancer cell surface, suggesting that these synthetics are potential to deliver agents into tumor cells in a targeted way [13]. Several studies have addressed that the intake of FCN can be substantially observed in tumor cells but rarely in healthy cells [13,41], which is correlated to the overexpressed GLUT5 found on the surface of tumor cells [42,43]. Of note, nanoparticles that are coated with hydrophilic shell manufactured by long polymer chains with high fructose density have great affinity to tumor cells [14]. Since biocompatibility, relative non-toxic property and the capability to cross the blood-brain barrier, GCNs are widely used as the radiosensitizer in radiotherapy and can carry therapeutic biomolecules into brain [44,45].

Ethics approval and consent to participate
This article does not contain any studies with human participants or animals performed by any of the authors.

Consent for publication
Not applicable

Availability of data and material
Not applicable

Competing interests
The authors declare that they have no competing interests