Self-fluorescent drug delivery vector based on genipin-crosslinked polyethylenimine conjugated globin nanoparticle

Materials Science and Engineering C 71 (2017) 17–24

Contents lists available at ScienceDirect

Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec

Self-fluorescent drug delivery vector based on genipin-crosslinked polyethylenimine conjugated globin nanoparticle Yan Zhang, Lina Mao, Juan Liu, Tianjun Liu ⁎ Institute of Biomedical Engineering, Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin Key Laboratory of Biomaterial Research, Tianjin 300192, China

a r t i c l e

i n f o

Article history: Received 1 July 2016 Received in revised form 23 August 2016 Accepted 26 September 2016 Available online 28 September 2016 Keywords: Self-fluorescence Globin Drug delivery, Fluorescence imaging Genipin-crosslink

a b s t r a c t A kind of self-fluorescent, biocompatible, and low-toxic Genipin crosslinked Globin-PEI nanoparticle (Gb-G-PEI NP) with high enzymolysis-stability and photo-stability was synthesized successfully. The properties of the Gb-G-PEI NP were characterized, including its particle size, surface zeta potential, morphology, paclitaxel (PTX) loading capacity and release. The Gb-G-PEI NPs as imaging probe were investigated by Confocal Laser Scanning Microscope (CLSM) in vitro and by fluorescence imaging system in vivo. Cell imaging results showed that the tumor cell line (HepG-2) had the faster cell uptake rate and metabolism rate than the normal cell line (L-O2), this difference showed its tumor selectivity. MTT assay revealed that the PTX-loaded Gb-G-PEI NPs showed almost the equal potence to tumor cell HepG-2 as the free PTX at the same PTX concentration, while a lower cytotoxicity to normal cell L-O2, suggesting its promising utilization as a drug delivery system. The imaging on mice demonstrated the possibility of the self-fluorescent Gb-G-PEI NPs as probe in vivo. So Gb-G-PEI NPs can be potentially utilized as both tracking marker and tumor cell selective drug delivery system in the biomaterial field. © 2016 Published by Elsevier B.V.

1. Introduction Protein-based nanoparticles with characteristics of good absorbability, non-toxicity, and non-antigenicity, have attracted special interest in biomaterial fields [1,2]. Proteins often used were gelatin, collagen, casein, albumin or whey protein, and these protein-based nanoparticles were usually studied for delivering drugs, nutrients, bioactive peptides and probiotic organisms, etc. [1–5] Globin (Gb), the protein component of Hemoglobin (Hb), was studied extensively as a kind of promising oxygen carriers [6,7], and the past chemical modification on globin included cross-linking of two lysine residues by glutaraldehyde [8] or bis(3,5dibromosalicyl) fumarate [9],the PEGylation of the surface amino groups through urethane linkages [10], isopeptide linkage [11] or propyl maleimide linkage [12],etc. Besides its oxygen carrier ability, globin is also a bioactive “peptide pool” [13], its degradation would produce many active peptides, like hemorphin, ACE inhibitor, antibacterial peptides, so as a biomaterial it possesses more advantage compared with other proteins. However, little work on its potency as drug delivery system was studied. For a delivery system, the surface properties, especially the charged groups, play an important role in their cellular internalization [14,15]. Positively charged particles generally had larger uptake amount than negatively charged or neutral ones [15]. Polyethylenimine (PEI), a cationic linear or branched polymer, with molecular weight ranging between 0.6 and 1000 kDa [16], could be modified on different kinds of ⁎ Corresponding author. E-mail address: [email protected] (T. Liu).

http://dx.doi.org/10.1016/j.msec.2016.09.059 0928-4931/© 2016 Published by Elsevier B.V.

proteins [1,3,17]to form a positive charged system, which enhances the material's cellular adhesiveness and uptake ability. Genipin, the aglycon of geniposide found in traditional Chinese medicine, is an effective naturally occurring crosslinking agent that can react with amino acids or proteins [18,19]. Numerous studies have been performed in genipin crosslinking amine-group-containing biomaterials, like chitosan, gelatin, silk protein, collagen, casein, etc. [18,20–23] Here we used genipin as crosslinker to conjugate PEI with globin. Fluorescence imaging with advantages of simplicity, low-cost and noninvasive, could provide information of the bio-distribution, metabolic characteristics of the delivery systems by real-time tracking of these drug delivery vectors in vitro and in vivo [24]. The imaging fluorescent labels usually used are dyes [25] or quantum dots [26] which were conjugated to or encapsulated in the delivery system to produce optical imaging in vitro and in vivo [27,28]. However sometimes the spontaneous release of these often-toxic labels caused imaging ambiguity and gave the inaccurate information. In this paper, the external fluorescent label was omitted, instead the intrinsic pigment formed by genipin reaction with PEI and globin, was used as fluorescent label to provide accuracy information for the globin–PEI system in vitro and in vivo. Genipin is colorless but its reaction with amino acids or proteins could form blue fluorescent pigments, which was stable with regard to pH, temperature, and light conditions [29,30]. Up to now, most research about genipin was focused on its crosslink mechanism and efficiency, or the properties of the genipin-crosslinked product, like cytotoxicity, resistance to enzymatic degradation, mechanical properties, etc. The optical property of the genipin-crosslinked fluorescent product received less attention [31–33], some work used this pigment fluorescence to indicate the

18

Y. Zhang et al. / Materials Science and Engineering C 71 (2017) 17–24

biodegradation or morphological structure of the materials. No work used this pigment as fluorescent probe to track nanocarrier in vitro or in vivo. This paper would explore the safety of genipin function as natural crosslinker and its product as fluorescent marker in vitro and in vivo. In this research paper, a kind of self-fluorescent, biocompatible, and low-toxic Genipin crosslinked Globin-PEI nanoparticle with high enzymolysis stability and photo-stability was synthesized successfully (Scheme 1). The self-fluorescent was used as a bio-imaging probe to track the cellular uptake and metabolism of Gb-G-PEI nanoparticle by Confocal Laser Scanning Microscopy (CLSM). Also, the self-fluorescent nanoparticle was monitored by in vivo imaging system with mice as models. The cytotoxicity of the developed paclitaxel-loaded Gb-G-PEI nanoparticle was assessed in comparison to free PTX at varying drug concentrations. Results were promising and suggest its application in fluorescent imaging and drug delivery. 2. Experimental section 2.1. Materials The branched Polyethylenimine (PEI, Mw = 1.8 or 10 kDa) was purchased from Aladdin Industrial Corporation (Shanghai, China). Hemoglobin from swine (Hb) was obtained from Hewons Biochem Technologies (Tianjin, China). Genipin (98%) was purchased from Xi'an Plant Bio-engineering Corporation (Xi'an, China). Paclitaxel (PTX) was obtained from Huafeng United Technology Company (Beijing, China). All other chemical reagents were purchased from Tianjin Jiangtian Chemical Technology Corporation (Tianjin, China). Acetonitrile (CH3CN) and trifluoroacetic acid (TFA) (HPLC grade) were purchased from Concord Technology Corporation (Tianjin, China). All reagents were used as received without further purification. Hepatoma carcinoma cell line (HepG-2)and hepatic cell line (L-O2) were purchased from American type culture collection (Maryland, USA) and cultured in DMEM medium enriched with 10% fetal bovine serum (FBS) and 100 U/mL penicillin-streptomycin. Cells were incubated at 37 °C with 5% CO2. ICR mice (4-weeks, male, 20–25 g) were used to evaluate the in vivo imaging. All animal procedures are approved and controlled by the local ethics committee and carried out in accordance with the guidelines Principles for the Care and Use of Laboratory Animals, Peking Union Medical College, People's Republic of China. 2.2. Genipin crosslinked Globin-PEI Nanoparticles (Gb-G-PEI NP) Globin was prepared from hemoglobin by acid-acetone extraction method [34] and characterized by UV–Vis spectra and far-UV CD

spectra. The Gb-G-PEI NP was prepared by genipin cross-linked reaction between globin and PEI. The experiment was as follow: 20 mg globin and 120 mg PEI (Mw = 1800 Da) was dissolved in 7 mL 0.1 mol/L HCl solution, the pH value was adjusted to 5.0 with 0.65 mL 1 mol/L NaOH solution. Then 1.91 mL genipin solution (50 mmol/L) was added, and the final genipin concentration in the system was 10 mmol/L. The system was thoroughly stirred and at time point of 5, 10, 24, 36, 48 or 72 h, 0.5 mL solution was taken out to monitor the reaction progress by both the UV–Vis spectrum and Fluorescent spectrum respectively. The color of the system changed from slight yellow to dark blue, and the solution was extensively dialyzed (MWCO 12,000–14,000 Da) against de-ionized water for 24 h, the water was totally changed 5 times. Dialyzed product was lyophilized to give the Gb-G-PEI1800 NP. To investigate the rate of the crosslink reaction dependence on the concentration of genipin, the following experiments was conducted: 0.32 mL genipin solution was added to Gb and PEI1800 system, the procedure was remained the same as above. Gb-G-PEI10,000 NP (PEI Mw = 10,000 Da) was prepared in the same method as Gb-G-PEI1800 nanoparticles except that the final genipin concentration in the reaction system was 10 mmol/L and reaction time was 48 h at 37 °C. 2.3. The stability of Gb-G-PEI NPs' self-fluorescence Both the photo-stability and enzyme-stability were evaluated for nanoparticle's self-fluorescence. The procedure was as following: To monitor the photo stability, Gb-G-PEI NPs sample (1 mg/mL in water) was irradiated with a 630 nm semiconductor laser (7404, Intense, USA) via an optic fiber (5-cm diameter, 250 mW/cm2), its absorbance and fluorescence were recorded at different time points. To measure the enzyme-stability, Gb-G-PEI NPs sample (1 mg/mL in water) was treated with different concentration of trypsin solution, 1000 μg/mL or 40 μg/mL respectively. UV–Vis and fluorescence spectra were recorded at different times to evaluate the stability of self-fluorescence. 2.4. The paclitaxel loaded Gb-G-PEI1800 NPs 100 mg Gb-G-PEI1800 NPs was suspended in 10 mL H2O–DMSO (1:1 v/v) and kept at 37 °C for 1 h under continuous stirring to swell completely, then 1 mL of the PTX solution (10 mg/mL in DMSO) was added and stirred for 4 h for encapsulation. The resulting assembly suspension was dialyzed (MWCO 3500 Da) against distilled water for 36 h, and the water was totally changed 6 times. The obtained solution in dialysis tube was filtered through 0.45 μm filter and lyophilized to give PTX-loaded nanoparticles. To obtain the PTX-loaded nanoparticles with different drug loading capacity, 3 or 5 mL of the PTX solution (10 mg/mL in DMSO) was added instead. The drug-loaded capacity (LC) and encapsulation efficiency (EE) of PTX was analyzed by HPLC (Waters e2695 Separations Module liquid chromatography, column: VP-ODS C18, 5 μm, 4.6 × 150 mm; acetonitrile/water 70/30 (v/v), 1.0 mL/min, UV detector at 227 nm) and calculated as follow: Loading capacityðLC Þ ¼

mass of loaded drug in carrier  100% mass of carriers

Encapsulation efficiencyðEEÞ ¼

mass of loaded drug in carriers mass of initially added drug  100%

2.5. In vitro drug release

Scheme 1. Scheme for the synthesis of Gb-G-PEI nanoparticles and its cellular uptake process.

A solution of PTX-loaded Gb-G-PEI1800 NPs (1 mL, 5 mg/mL, LC was 8.55 ± 0.18%) was dialyzed (MWCO 3500 Da) against 12 mL of phosphate buffer (10 mmol/L phosphate, 150 mmol/L NaCl, 0.5%

Y. Zhang et al. / Materials Science and Engineering C 71 (2017) 17–24

Tween 80, pH 7.4). At each time point, 3 mL of the extra fluid were taken out for paclitaxel analysis by HPLC and 3 mL of fresh phosphate buffer was added instead. The drug release experiment was conducted at time-point 1, 3, 5, 8, 11, 14, 24, 36 and 48 h respectively. In order to study enzyme degradation on the release behavior of PTX-loaded nanoparticles, 0.2 mL trypsin (30 mg/mL) was added to the PTX-loaded Gb-G-PEI1800 NPs at the beginning of the release process, the other procedure was the same as before. 2.6. Cytotoxicity assay HepG-2 cell line and L-O2 cell line were seeded in a 96-well plate at a density of 0.7 × 104 cells/well in medium, and incubated for 24 h. Then the medium was replaced with 100 μL medium containing 500, 250, 125, 62.5, or 31 mg/L of Gb-G-PEI1800 NPs or Gb-G-PEI10,000 NPs respectively to evaluate the cytotoxicity of the blank nanoparticles. To measure the drug delivery efficiency, PTX or PTX-loaded Gb-G-PEI1800 NPs in 100 μL medium (CPTX: 0.1, 0.4, 1, 2.5 or 10 mg/L) were added to the wells respectively. After incubated for 48 h, the cell viability was tested by MTT assay [35]. 2.7. Cellular uptake and metabolism of Gb-G-PEI1800 NPs studied by confocal laser scanning microscope HepG-2 cell line and L-O2 cell line were seeded respectively at a density of 2 × 104 cells per well in confocal image dishes (NEST, 35 mm) and incubated for 24 h. To analyze the cellular uptake, the growth medium was replaced with Gb-G-PEI1800 NPs solution (set as 0 h), and incubated for another 48 h. During this period, the cellular fluorescent imaging was recorded at 0.5, 1, 2, 4, 6, 10, 14, 24, 36, or 48 h respectively. To determine the cellular fluorescence intensity, cells were washed three times with PBS and incubated with Hoechst 33342 for 10min at 37 °C, and the intracellular fluorescence intensity was analyzed by CLSM. To evaluate the cellular metabolism, first the cells were incubated for 36 h in the growth medium containing Gb-G-PEI1800 NPs, then the cells were washed three times with PBS, fresh cultures medium was added (this time point was set as 0 h) and incubated for another 36 h for metabolism analysis, the intracellular fluorescence was measured every 12 h and analyzed by CLSM. The CLSM images were taken on the ZEISS LSM 710 (Germany) and Hoechst 33342 excited at 405 nm and Gb-G-PEI1800 NPs excited at 633 nm. The mean intensity of the intracellular fluorescence was analyzed by “ZEN 2008” software.

19

CD spectrum revealed that in globin some α-helix and β-sheet was decoiled into β-turn and random coil (Fig. S1B) compared with hemoglobin. This could be rationalized by the destruction of the bonds between histidine residues in the globin and heme, leading to the decrease in spatial structure of α-helix and β-sheet [36]. The self-fluorescence of the Gb-G-PEI NPs was from the oxygen radical-induced polymerization of genipin [19,29], so the genipin concentration and the reaction time became the two dominated factors for the nanoparticle's fluorescence property. Fluorescence and UV–Vis spectra revealed that the maximum optical intensity of the products (IAbs(595) orIF(618)) increased with reaction time or genipin concentration (Fig. S2), the higher genipin concentration and longer reaction time were in favor of the oxygen-induced polymerization reaction to generate the fluorescent blue pigments. The synthetic condition was optimized, and the best reaction condition was 10 mmol/L genipin and reacted for 48 h at 37 °C. Both the Gb-G-PEI1800 NP (PEI Mw = 1800 Da) and Gb-G-PEI10000 NP (PEI Mw = 10,000 Da) were prepared following these optimized synthesis procedure. The size and zeta potential of these nanoparticles were analyzed on Malvern Zetasizer Nano ZS90 Instrument, the results was shown in Fig. 1. The hydration diameter of Gb-G-PEI1800 NP was ~ 223.7 nm in pH 7.4 buffer solution, while globin had the size ~100 nm in 0.1 mol/L HCl (insoluble at pH 7.4). The surface zeta potential of Gb-G-PEI1800 NP was ~+34.7 mV. While the hydration diameter and surface zeta potential of Gb-G-PEI10,000 NP were a little higher than Gb-G-PEI1800 NP in pH 7.4 buffer solution, its size was ~283.3 nm and the zeta potential was ~+37.4 mV. These physical characteristics facilitated the nanoparticles adhesion to the negative charged cell membrane and then the cellular internalization [14,15]. AFM image showed that the average diameter of the nanoparticles was about 150–200 nm (Fig. 1B and C), this was in consistent with the hydrodynamic diameter measurement. Interestingly, the shape of the Gb-G-PEI1800 NP was like a sector, while most of the Gb-G-PEI10,000 NP was sphere. This could be rationalized with the different reactivity of lysine (− NH2) residues at the α-chain and β-chain [37–39], as well as the length of PEI chain. Most of the lysine residue (\\NH2) in the β-chain had higher solvent-accessible surface area than that of in α-chain, so they had higher potential reactivity to conjugate with PEI by genipin. The essence of the genipin crosslinked reaction is the amine moiety addition to the double bonds in genipin, the amine moiety could be from globin and PEI. In the initial process of the crosslink reaction, both globin and PEI were involved, and with the progress of the

2.8. In vivo imaging Male ICR mice (20–25 g) as animal model were used for in vivo imaging.0.2 mL Gb-G-PEI1800 NPs in physiological saline (10 mg/mL) was subcutaneous injected to the back of the mice after gaseous isoflurane anesthesia. Images were taken at 10 min, 3 h, 6 h, 18 h, 24 h and 36 h respectively, using the Xenogen IVIS instrument. The imaging data were: excitation at 595 nm, emission at 635 nm, and the exposure time was 4 s. At the end of the imaging, anesthetized mice were sacrificed and organs were harvested to evaluate the distribution of near infrared fluorescence agent. The fluorescent images were analyzed using the Xenogen Analysis Software. 3. Results and discussion 3.1. Synthesis and characterization of Gb-G-PEI nanoparticle (Gb-G-PEI NP) Globin was obtained from hemoglobin by the acid-acetone extraction method, [34] and the reaction progress was monitored by UV–Vis spectrum. The characteristic absorbance peak of heme at 405 nm was used as an indicator for this reaction, as this peak completely disappeared, hemoglobin was completely transferred to globin (Fig. S1A).

Fig. 1. (A) Size and zeta potential of the nanoparticles; the AFM images of (B) Gb-G-PEI1800 NPs and (C) Gb-G-PEI10,000 NPs.

20

Y. Zhang et al. / Materials Science and Engineering C 71 (2017) 17–24

(A)

35 30

40

25

Gb-G-PEI10000 NP

20

IF(618)

IF(618)

Gb-G-PEI1800 NP

(B) 45

Gb-G-PEI1800 NP

35 30

Trypsin Concentration 40 µg/mL 1000 µg/mL

25

Gb-G-PEI10000 NP

20

15

15

10

0

0.167

1.5

3

6.5

Laser Illumination Time (Hour)

0

4

22

50

Enzymolysis Time (Hour)

Fig. 2. The fluorescence stability of Gb-G-PEI nanoparticle. The maximum fluorescent intensity IF(618) of nanoparticles changed upon different (A) laser illumination times, (B) trypsin concentration as well as enzymolysis times. The values are the mean (±S.D.) of three independent experiments.

reaction, the amine at globin was consumed, and the following reaction would be occurred among PEI. For PEI1800, the molecular skeleton is relative rigid compared with PEI10,000, so the cascade reaction would form a sector structure by one by one connection. While for PEI10,000, the molecule is flexible and long enough to rotate and make a turn, so as one amine moiety anchored on the globin nanoparticle, the other amine on the long chain would attend the reaction, this intermolecular reaction would attenuate the further cascade reaction, so the flexibility of the long chain in PEI10,000 made the reaction in a relative smaller extent compared with that of PEI1800. This resulted in the preservation of the initial morphology of the globin. That is why different reaction capacity in PEI induced the different morphology of the Gb-G-PEI nanoparticles. The stability of the nanoparticles' self-fluorescence was studied under light illumination or enzymolysis. After laser light irradiation at 630 nm (0.25 W) for 6 h, the maximum emission intensity (Fig. 2A) of the nanoparticles changed b 3%, and the fluorescent spectra (Fig. S3) remained almost the same as before, with no obvious spectrum shift. This indicated the self-fluorescence of Gb-G-PEI1800 NPs had high photo-stability. Coincubated the Gb-G-PEI1800 NPs with trypsin (in two conditions: 1000 μg/mL or 40 μg/mL) in pH 7.4 PBS buffer for 50 h, the alterations in both the maximum fluorescence intensity (Fig. 2B) and the whole spectra (Fig. S4) were negligible, indicating the high enzymolysis stability of the self-fluorescence of Gb-G-PEI1800 NPs. The similar phenomena also observed for Gb-G-PEI10,000 NPs. So the self-fluorescence of nanoparticle was both high enzymolysis stability and photo-stability, implied their potential application in tracking the biomaterials in vitro and in vivo. 3.2. Paclitaxel loading and release PTX was loaded into Gb-G-PEI1800 nanoparticles by the hydrophobic interaction. The size, zeta potential, PTX-loaded capacities and encapsulation efficiencies were shown in Table 1. The size of the blank nanoparticle was 223.7 ± 15.5 nm, while it increased with the loading capacity of PTX (LC-PTX), 289.6 ± 22.1 nm at LC-PTX 8.55 ± 0.18%, 328.6 ± 35.1 nm at LC-PTX 15.67 ± 0.64% and 356.2 ± 27.1 nm at LC-PTX 21.76 ± 0.47%. All nanoparticles had positive surface due to the PEI polymer, but the positive potential decreased with the LC-PTX.

This could be explained based upon the surface charge density (σo = Q / S), generally lower surface charge density leaded to the lower surface zeta potential. In this system, Q was charge at the surface of nanoparticle and kept no change upon PTX-loading, while S was the surface area of the nanoparticle and increased with PTX-loading, so the total result of PTX-loading lowered the zeta potential. The encapsulation efficiency changed with the loading capacity, higher loading capacity caused the lower encapsulation efficiency. The in vitro release of the loaded paclitaxel from Gb-G-PEI1800 NPs was followed a normal dynamic model, the result was shown in Fig. 3. Interestingly, PTX was released gentle and sustained from PTX-loaded Gb-G-PEI1800 NPs, there showed non burst release during the investigated period. This indicated PTX was encapsulated in the hydrophobic core of nanoparticle, which was stable during the release process, so the release of PTX from the nanoparticles was a dynamic dissociation process with a controllable rate. The release rate and accumulated amount could be accelerated by the addition of trypsin to degrade the nanoparticles (Fig. S5). In the initial 5 h, there was no obvious difference in release for both cases. However, the efficacy of trypsin became clear after 8 h. After 48 h, the accumulated release amount of PTX from Gb-G-PEI1800 NPs could be ~70% in presence of 5 mg/ml trypsin, while only ~50% in absence of trypsin under the same condition. 3.3. Cellular uptake and metabolism of Gb-G-PEI1800 NPs analyzed by CLSM To track the Gb-G-PEI1800 NPs in cellular uptake and metabolism process, cellular fluorescent images were taken by CLSM with the selffluorescence of nanoparticle as the imaging probe. In this case the fluorescent intensity indicated the relative amount of Gb-G-PEI1800 NPs. Fig. 4A and Fig. 4B demonstrated that the cellular fluorescence intensity (cellular uptake of Gb-G-PEI1800 NPs) increased with the incubation time. Fig. 4D was the quantized value of Fig. 4A and B, at initial 14 h, the mean intensity of the intracellular fluorescence was increased with time, and it was obvious in HepG-2 cell line than that in L-O2 cell line. This implied HepG-2 cell had faster uptake rate for the Gb-GPEI1800 NPs. After 24 h or longer time incubation, minor difference in the fluorescent intensity in these two cell lines was shown, this phenomenon could be explained based upon the balance of the cellular

Table 1 The properties of paclitaxel-loaded Gb-G-PEI1800 NPs. Feeded PTX ratioa

Diameter nmb

Zeta potential mVb

Loading capacityc (%)

Encapsulation efficiencyd (%)

0 10% 30% 50%

223.7 289.6 328.6 356.2

34.7 14.8 11.6 10.2

– 8.55 ± 0.18 15.67 ± 0.64 21.76 ± 0.47

– 86.63 ± 1.65 58.70 ± 2.07 53.61 ± 0.95

a b c d

± 15.5 ± 22.1 ± 35.1 ± 27.1

± 3.2 ± 1.4 ± 1.6 ± 2.5

PTX feed ratio: weight of feeded drug / weight of nanoparticle (Wdrug / Wnanoparticle × 100%) in the loading process. Average diameter and zeta potential determined by dynamic light scattering. Paclitaxel loading capacity (Wdrug / Wnanoparticle × 100%) measured from HPLC. Paclitaxel encapsulation efficiency (Wdrug-loaded / Wdrug-added × 100%) measured from HPLC.

Accumulative Release (%)

Y. Zhang et al. / Materials Science and Engineering C 71 (2017) 17–24

70

21

that in L-O2 cell line. Due to the initial uptake amount was different for these two cell lines, the percentage of fluorescent intensity relative to the initial state was used as an index of its mean metabolism rate at each time point. After 12 h metabolism incubation, only ~ 44% of the fluorescence could be detected in HepG-2 cell line, while ~67% in L-O2 cell line, and the value was about 21% in HepG-2 cell line and ~41% in L-O2 cell line after 24 h metabolism. The ensemble effect of cell internalization and metabolism accounted for why the fast cellular uptake didn't lead to a higher nanoparticle internalized fluorescent intensity with time.

60 50 40 30 20 10 0

10

20

30

40

50

3.4. Cytotoxicity of Gb-G-PEINPs

Release Time (Hour) Fig. 3. The release curves of PTX-loaded Gb-G-PEI1800 NPs with 5 mg/ml trypsin (red line) or no trypsin (black line). The values are the mean (±S.D.) of three independent experiments. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

uptake and metabolism for the nanoparticle. Fig. 4C and E illustrated the cellular metabolism of the Gb-G-PEI1800 NPs in these two cells, the intracellular fluorescence decreased faster in HepG-2 cell line than

(A) 2 hr

6hr

14 hr

HepG-2

The cytotoxicity of the blank nanoparticle is crucial to its application as delivery vehicle. HepG-2 cell lines and L-O2 cell lines were incubated with Gb-G-PEI1800 NPs and Gb-G-PEI10,000 NPs respectively at various concentrations at 37 °C for 48 h, the cell viability was determined by the MTT assay, and the results were shown in Fig. 5. The Gb-G-PEI1800 NPs showed low cytotoxicity (N 90% viability remained) at concentration up to 500 mg/L in HepG-2 cell lines and L-O2 cell lines, however, the Gb-G-PEI10,000 NPs showed obvious cytotoxicity to experimental cell lines at higher concentration, the viability was ~ 80% or ~ 60% for

(B)

2 hr

6hr

14 hr

24hr

36hr

48hr

L-O2 24hr

(C)

36hr

48hr

0 hr

12hr

24 hr

36hr

0 hr

12hr

24 hr

36hr

HepG-2

L-O2

(D)

(E) 6 Cellular Uptake of the Gb-G-PEI1800 NPs

120 Cellular Metabolism of the Gb-G-PEI1800 NPs

4

L-O2 Cell Line HepG-2 Cell line

3

L-O2 Cell Line HepG-2 Cell line

90

Intt / Int 0 (%)

Mean Intensity

5

60

2

30

1 0 0 0.5 1

2

4 6 10 14 24 36 48

Incubation Time Hour

0 0

12

24

36

Incubation Time (Hour)

Fig. 4. The cellular uptake and metabolism behavior were observed by CLSM. CLSM images of (A) HepG-2 cell line and (B) L-O2 cell line showed the cellular uptake behavior of Gb-GPEI1800 NPs; (C) CLSM images showed the metabolism process of Gb-G-PEI1800 NPs in HepG-2 cell line and L-O2 cell line. Panels (D) and (E) were the mean intensity of the intracellular fluorescence and the values are the mean (±S.D.) of five independent CLSM images.

22

Y. Zhang et al. / Materials Science and Engineering C 71 (2017) 17–24

(A) HepG-2Cell Line

(B) L-O2 Cell Line

100

Cell Viability %

Cell Viability %

120

80 60 40

Gb-G-PEI1800 NP Gb-G-PEI10000 NP

20 0

100 80 60 40

Gb-G-PEI1800 NP Gb-G-PEI10000 NP

20 0

Control 31.5

62.5

125

250

Control 31.5

500

62.5

125

250

500

Carrier Concentration (mg/L)

Carrier Concentration (mg/L)

Fig. 5. The cell viabilities of (A) HepG-2 cell line and (B) L-O2 cell line upon incubation with Gb-G-PEI10,000 NPs and Gb-G-PEI1800 NPs for 48 h. The values are the mean (±S.D.) of five independent experiments.

HepG-2 cell line or L-O2 cell line respectively at 500 mg/L. So the following experiments were focused on the Gb-G-PEI1800 NPs. Paclitaxel was loaded into Gb-G-PEI1800 NPs and the cytotoxicity of the PTX-loaded nanoparticles was studied by the MTT assay (Fig. 6). Results showed that cell viabilities of the HepG-2 cell line and L-O2 cell line changed with the concentration of PTX as well as PTX-loaded Gb-GPEI1800 NPs. For HepG-2 cell line, free PTX and that of PTX-loaded GbG-PEI1800 NPs showed similar anti-proliferative activity among all experimental doses (Fig. 6A). However, for L-O2 cell line, free PTX had greater anti-proliferative activity than PTX-loaded Gb-G-PEI1800 NPs (Fig. 6B) at the same PTX dose. The data in Fig. 6A and Fig. 6B revealed that PTX loaded Gb-G-PEI1800 NPs had tumor cell selectivity. For example at 1 mg/L paclitaxel, the viability of L-O2 cell was ~ 47% for free PTX,~74% for PTX-loaded Gb-G-PEI1800 NPs, however, the data in HepG-2 cell line was ~ 44% for free PTX, ~ 50% for PTX-loaded Gb-GPEI1800 NPs. These phenomena could be rationalized by the difference in cellular uptake and metabolism behavior of Gb-G-PEI1800 NPs between tumor cell line (HepG-2) and normal cell line (L-O2). At the same condition, higher cellular uptake in HepG-2 leaded to more internalization of PTX-loaded nanoparticles, a faster metabolism behavior in HepG-2 cell line resulted in the higher intracellular PTX concentration and higher cell death ratio. Usually the cell viability of the drug-loaded nanoparticles was higher than that of the free drug, in this paper, the PTX-loaded Gb-G-PEI1800 NPs showed the similar cell viability to that of the free drug at the same PTX dose in tumor HepG-2 cell line (Fig. 6A), but it showed lower cytotoxicity to normal cell LO2 at same PTX dose compared with free PTX, this tumor cell selectivity made it a good candidate as delivery vector. 3.5. Optical imaging in vivo Whole animal imaging was performed in ICR mice to track implanted Gb-G-PEI1800 NPs by subcutaneous injection at the back. To

(A) HepG-2 Cell Line

track the in vivo process of the Gb-G-PEI1800 NPs, fluorescence imaging was taken from 10 min to 36 h using the self-fluorescence. As seen in Fig. 7A, the fluorescence intensity at the injection site was decreased with time. The relative value in Fig. 7B demonstrated that, ~50% of fluorescence was dislodged after 6 h, and after the permeation and distribution for 24 h, only ~ 10% fluorescence in the injection site could be observed, implying the rapid permeation and metabolism of these GbG-PEI1800 NPs. After 36 h, the mice were sacrificed and the organs, including heart, liver, spleen, kidney, and lung were harvested for ex vivo analysis of material biodistribution (Fig. 7C), strong fluorescence could be seen clearly in the kidney, while a little in the lung. The data revealed that the Gb-G-PEI1800 NPs was mainly expelled out via kidney after metabolism.

4. Conclusion A kind of self-fluorescent, water-soluble, biocompatible and lowtoxic genipin crosslinked globin-PEI nanoparticle (Gb-G-PEI NP) was synthesized. Its self-fluorescence was stable upon laser irradiation at 630 nm (0.25 W) for 6 h or the 1000 μg/mL trypsin enzymolysis for 50 h, and this facilitated the track of permeation and bio-distribution of Gb-G-PEI NP in vitro and in vivo. The cellular uptake and metabolism studied by tracking the self-fluorescence of the Gb-G-PEI NPs, revealed that the tumor cell HepG-2 had higher uptake and faster metabolism rate than normal cell L-O2. This difference made the PTX-loaded GbG-PEI1800 NPs have tumor cell selectivity. Cytotoxicity results revealed that the PTX-loaded Gb-G-PEI1800 NPs showed the similar potence as the free PTX to tumor cell HepG-2, lower cytotoxicity to normal cell LO2 at the same PTX concentration. This suggested its potential utilization as a drug delivery system. In a word, the Gb-G-PEI1800 NPs could play double roles as imaging marker and tumor selective drug delivery system.

(B) L-O2 Cell Line 120

Cell Viability %

Cell Viability %

100 80 60 40 20 0

PTX Gb-G-PEI1800 NPs PTX-Loaded Gb-G-PEI1800 NPs Blank

0.1

0.4

1

2.5

PTX Concentration (mg/L)

10

100 80 60 40 20 0

PTX Gb-G-PEI1800 NPs PTX-Loaded Gb-G-PEI1800 NPs Blank

0.1

0.4

1

2.5

10

PTX Concentration (mg/L)

Fig. 6. The cell viabilities of (A) HepG-2 cell line and (B) L-O2 cell line after incubation for 48 h with PTX, PTX-loaded Gb-G-PEI1800 NPs and Gb-G-PEI1800 NPs, respectively. The values are the mean (±S.D.) of five independent experiments.

Y. Zhang et al. / Materials Science and Engineering C 71 (2017) 17–24

23

(B)

Relative Intensity (%)

(A)

100

(C)

80 60 40 20 0 0

8

16

24

32

Time (hour) Fig. 7. (A) Fluorescence imaging of Gb-G-PEI1800 NPs in vivo at different time, with representative one of the three in each group. (B) The relative fluorescence intensity in the injection site to track the permeation rate of the nanoparticles. (C) Ex vivo image with organ distribution of Gb-G-PEI1800 NPs.

Acknowledgment This work was supported by the Key Technologies R & D Program of Tianjin (12ZCDZSY11900) and Peking Union Medical College Postdoctoral Fund. Appendix A. Supplementary data The UV–Vis spectra, fluorescence spectra, and GPC images of the GbG-PEINPs which were not included in the body of this manuscript were showed in the support information. Supplementary data associated with this article can be found in the online version, at http://dx.doi. org/10.1016/j.msec.2016.09.059.

References [1] A.O. Elzoghby, W.M. Samy, N.A. Elgindy, Albumin-based nanoparticles as potential controlled release drug delivery systems, J. Control. Release 157 (2012) 168–182. [2] M. Murata, S. Narahara, T. Kawano, N. Hamano, J.S. Piao, J.-H. Kang, K. Ohuchida, T. Murakami, M. Hashizume, Design and function of engineered protein nanocages as a drug delivery system for targeting pancreatic cancer cells via neuropilin-1, Mol. Pharm. 12 (2015) 1422–1430. [3] K. Numata, D.L. Kaplan, Silk-based delivery systems of bioactive molecules, Adv. Drug Deliv. Rev. 62 (2010) 1497–1508. [4] A.O. Elzoghby, Gelatin-based nanoparticles as drug and gene delivery systems: reviewing three decades of research, J. Control. Release 172 (2013) 1075–1091. [5] Z. Ruszczak, W. Friess, Collagen as a carrier for on-site delivery of antibacterial drugs, Adv. Drug Deliv. Rev. 55 (2003) 1679–1698.

[6] P. Cabrales, M. Intaglietta, Blood substitutes: evolution from non-carrying to oxygen and gas carrying fluids, ASAIO journal (American Society for Artificial Internal Organs: 1992) 59 (2013) 337. [7] C. Natanson, S.J. Kern, P. Lurie, S.M. Banks, S.M. Wolfe, Cell-free hemoglobin-based blood substitutes and risk of myocardial infarction and death: a meta-analysis, JAMA 299 (2008) 2304–2312. [8] A. Alayash, Hemoglobin-based blood substitutes and the hazards of blood radicals, Free Radic. Res. 33 (2000) 341–348. [9] J.A. Walder, R.H. Zaugg, R.Y. Walder, J.M. Steele, I.M. Klotz, Diaspirins that crosslink .beta. chains of hemoglobin: bis(3,5-dibromosalicyl) succinate and bis(3,5dibromosalicyl) fumarate, Biochemistry 18 (1979) 4265–4270. [10] K. Nho, D. Glower, S. Bredehoeft, H. Shankar, R. Shorr, A. Abuchowski, PEG-bovine hemoglobin: safety in a canine dehydrated hypovolemic-hemorrhagic shock model, biomaterials, Artificial Cells and Immobilization Biotechnology 20 (1992) 511–524. [11] Z. Abassi, S. Kotob, F. Pieruzzi, M. Abouassali, H.R. Keiser, J.C. Fratantoni, A.I. Alayash, Effects of polymerization on the hypertensive action of diaspirin cross-linked hemoglobin in rats, J. Lab. Clin. Med. 129 (1997) 603–610. [12] B. Manjula, A. Tsai, R. Upadhya, K. Perumalsamy, P. Smith, A. Malavalli, K. Vandegriff, R. Winslow, M. Intaglietta, M. Prabhakaran, Site-specific PEGylation of hemoglobin at Cys-93 (β): correlation between the colligative properties of the PEGylated protein and the length of the conjugated PEG chain, Bioconjug. Chem. 14 (2003) 464–472. [13] V.T. Ivanov, A.A. Karelin, M.M. Philippova, I.V. Nazimov, V.Z. Pletnev, Hemoglobin as a source of endogenous bioactive peptides: the concept of tissue-specific peptide pool, Pept. Sci. 43 (1997) 171–188. [14] A. Verma, F. Stellacci, Effect of surface properties on nanoparticle–cell interactions, Small 6 (2010) 12–21. [15] J. Cho, F. Caruso, Investigation of the interactions between ligand-stabilized gold nanoparticles and polyelectrolyte multilayer films, Chem. Mater. 17 (2005) 4547–4553. [16] S. Abbasi, A. Paul, S. Prakash, Investigation of siRNA-loaded polyethylenimine-coated human serum albumin nanoparticle complexes for the treatment of breast cancer, Cell Biochem. Biophys. 61 (2011) 277–287. [17] A. Kichler, Gene transfer with modified polyethylenimines, Journal of Gene Medicine 6 (2004) S3–S10.

24

Y. Zhang et al. / Materials Science and Engineering C 71 (2017) 17–24

[18] Y.-H. Lin, S.-C. Tsai, C.-H. Lai, C.-H. Lee, Z.S. He, G.-C. Tseng, Genipin-cross-linked fucose–chitosan/heparin nanoparticles for the eradication of Helicobacter pylori, Biomaterials 34 (2013) 4466–4479. [19] J. Li, N. Ren, J. Qiu, X. Mou, H. Liu, Graphene oxide-reinforced biodegradable genipincross-linked chitosan fluorescent biocomposite film and its cytocompatibility, Int. J. Nanomedicine 8 (2013) 3415–3426. [20] K.-Y. Chen, C.-H. Yao, Repair of bone defects with gelatin-based composites: a review, Biomedicine 1 (2011) 29–32. [21] S.S. Silva, A. Motta, M.T. Rodrigues, A.F. Pinheiro, M.E. Gomes, J.F. Mano, R.L. Reis, C. Migliaresi, Novel genipin-cross-linked chitosan/silk fibroin sponges for cartilage engineering strategies, Biomacromolecules 9 (2008) 2764–2774. [22] H.G. Sundararaghavan, G.A. Monteiro, N.A. Lapin, Y.J. Chabal, J.R. Miksan, D.I. Shreiber, Genipin-induced changes in collagen gels: correlation of mechanical properties to fluorescence, J. Biomed. Mater. Res. A 87A (2008) 308–320. [23] F. Song, L.-M. Zhang, C. Yang, L. Yan, Genipin-crosslinked casein hydrogels for controlled drug delivery, Int. J. Pharm. 373 (2009) 41–47. [24] S. Bhaskar, F. Tian, T. Stoeger, W. Kreyling, J.M. de la Fuente, V. Grazú, P. Borm, G. Estrada, V. Ntziachristos, D. Razansky, Multifunctional nanocarriers for diagnostics, drug delivery and targeted treatment across blood-brain barrier: perspectives on tracking and neuroimaging, Particle and Fibre Toxicology 7 (2010) 1–25. [25] S. Luo, E. Zhang, Y. Su, T. Cheng, C. Shi, A review of NIR dyes in cancer targeting and imaging, Biomaterials 32 (2011) 7127–7138. [26] W.J. Mulder, R. Koole, R.J. Brandwijk, G. Storm, P.T. Chin, G.J. Strijkers, C. de Mello Donegá, K. Nicolay, A.W. Griffioen, Quantum dots with a paramagnetic coating as a bimodal molecular imaging probe, Nano Lett. 6 (2006) 1–6. [27] C. Ego, D. Marsitzky, S. Becker, J. Zhang, A.C. Grimsdale, K. Müllen, J.D. MacKenzie, C. Silva, R.H. Friend, Attaching perylene dyes to polyfluorene: three simple, efficient methods for facile color tuning of light-emitting polymers, J. Am. Chem. Soc. 125 (2003) 437–443. [28] X. Yin, F. Guo, R.A. Lalancette, F. Jäkle, Luminescent main-chain organoborane polymers: highly robust, electron-deficient poly(oligothiophene borane)s via stille coupling polymerization, Macromolecules (2016), http://dx.doi.org/10.1021/acs. macromol.5b02446.

[29] M.F. Butler, Y.F. Ng, P.D. Pudney, Mechanism and kinetics of the crosslinking reaction between biopolymers containing primary amine groups and genipin, J. Polym. Sci. A Polym. Chem. 41 (2003) 3941–3953. [30] Y.-S. Paik, C.-M. Lee, M.-H. Cho, T.-R. Hahn, Physical stability of the blue pigments formed from geniposide of gardenia fruits: effects of pH, temperature, and light, J. Agric. Food Chem. 49 (2001) 430–432. [31] Y.-J. Hwang, J. Larsen, T.B. Krasieva, J.G. Lyubovitsky, Effect of genipin crosslinking on the optical spectral properties and structures of collagen hydrogels, ACS Appl. Mater. Interfaces 3 (2011) 2579–2584. [32] H. Chen, W. Ouyang, B. Lawuyi, S. Prakash, Genipin cross-linked alginate-chitosan microcapsules: membrane characterization and optimization of cross-linking reaction, Biomacromolecules 7 (2006) 2091–2098. [33] J. Qiu, J. Li, G. Wang, L. Zheng, N. Ren, H. Liu, W. Tang, H. Jiang, Y. Wang, In vitro investigation on the biodegradability and biocompatibility of genipin cross-linked porcine acellular dermal matrix with intrinsic fluorescence, ACS Appl. Mater. Interfaces 5 (2013) 344–350. [34] M. Morrison, S. Horie, Determination of heme a concentration in cytochrome preparations by hemochromogen method, Anal. Biochem. 12 (1965) 77–82. [35] L. Zhao, L. Mao, G. Hong, X. Yang, T. Liu, Design, synthesis and anticancer activity of matrine–1H-1, 2, 3-triazole–chalcone conjugates, Bioorg. Med. Chem. Lett. 25 (2015) 2540–2544. [36] V. Kunitsyn, L. Panin, L. Osipova, L. Tabikhanova, T. Churkina, A. Rozumenko, Changes in the structural ordering of hemoglobin under extreme conditions of the Arctic region, Journal of Asian Scientific Research 5 (2015) 92–95. [37] Y.-D. Hu, J.-S. Bai, K.-F. Jiao, F.-R. Bu, S. Li, Study on polyethylene glycol modified porcine hemoglobin using computer modeling, Zhongguo Shi Yan Xue Ye Xue Za Zhi 9 (2001) 263–267. [38] S. Sivan, N. Lotan, Molecular engineering of proteins with predefined function. Part I: design of a hemoglobin-based oxygen carrier, Biomol. Eng. 20 (2003) 83–90. [39] D.A. Estrin, L. Capece, Computer simulation of chemical reactivity of heme proteins, Biophysical Journal 100 (2011) 379a.