CC BY
12DOI 10.24412/CL-37135-2023-1-9-14
COMPLEX APPROACH TO IN VITRO AND IN VIVO MONITORING OF THE DEGRADATION OF IMPLANTS BASED ON ESTER COPOLYMERS USING MR AND
FLUORESCENCE IMAGING
ASTEMIR LIKHOV1, ASIYA SAYDASHEVA1, | NATALIA KAZACHKINÄf, DMITRY DEMIN3, VERONIKA VOLODINA1, VICTORIA DEEVA1, DANIIL ANTONOV1, ILYA SOLOVYEV1, NADEZHDA MARYNICH1, IRINA MEEROVICH1, SOFYA ATZIGEIDA,2 DARIA TUCHINA1,2, , VALERY TUCHIN 1 2, AND VICTORIA ZHERDEVA1*
1RC of Biotechnology of the RAS, Moscow, Russia;
2
Saratov State University, Saratov, Russia; MIREA — Russian Technological University, Moscow, Russia
vjerdeva@inbi .ras.ru
ABSTRACT
Copolymer-based materials are widely used in medicine and biotechnology to produce various medical products, such as implants, prostheses, surgical suture material drug delivery systems, sensors and even triboelectric nanogenerators. The interest in such materials is increasing due to their multifunctional properties, such as the controlled degradation, the ability to stimulate tissue regeneration, the ability to act as carriers for controlled drug release, and triboelectric properties [1-3]. Some examples of copolymer-based materials include polymer nanoparticles [4], implantable triboelectric nanogenerator [5], polymer hydrogel [6], polymer matrix microcontainer [7], stimuli-responsive polymer materials [8].
Bioresorbable copolymers have the ability to degrade and be absorbed by the body over time. It has been noted that such materials might show high variation of degradation behavior [1].
The aim of our work was to monitor the degradation and assess the biocompatibility of ester copolymers in vitro and in vivo, as well as to demonstrate the possibility of visualizing these copolymers in vivo using a bimodal approach involving magnetic resonance imaging (MRI) and fluorescence imaging.
Copolymers were synthetized according to the protocol [2] and with some modification based on the 1,3-propanediol, 1,5-pentadiol, succinic acid and citric acid (Fig.la, 1b). The copolymers were labeled with indocyanine green (ICG) (Fig. 1c) and magnetic resonance agent such as Gd-DOTA (Gadolinium (III) 1,4,7,10-Tetraazacyclododecane-1,4,7,10-tetraacetate) or Gd-citrate complex (Fig.1d, 1e). Fourier-transform infrared spectroscopy (FTIR) analysis showed no structural differences between the labeled polymer samples and the initial polymers.
O O
Figure 1: Compounds used in polymer synthesis: (a) stage 1 - synthesis of monomeric structures from 1,3-propanediol and 1,5-pentadiol - (b) stage 2 - polymerization with the addition of succinic acid or citric acid - (c) ICG - (d) Gd-DOTA - (e) Gd-citrate complex
For all experiments copolymers samples were prepared as 5 mm discs using hole puncher (Fig.2a). The hydrolysis of the copolymers was modeled in vitro using buffered physiological solution (PhS) pH 7.4 and heat
inactivated fetal bovine serum (FBS).. Three discs were used in each vial (Fig.2b) for hydrolysis experiments, and the weight of the discs was measured every 3 days (Fig. 2a). Fluorescence and MRI signal changes (Fig.2c, 2d) have been registered at the same dates every 3 days. Porcine pancreas extract stock solution was prepared, and lipase activity was measured as 17500EU using «Lipase color liquid» (Sentinel Diagnostics, Italy). Enzyme solution was used to imitate lipase-mediated hydrolysis. The incubation conditions with the enzyme were set to maintain the physiological lipase enzymatic activity (0-150 EU) for 3 days at +37Co, after which the incubation solution was replaced with the fresh one.
(b)
500 450 400 5? 350 £ 300 J 250 "S 200
150 100
0 I I I I I I I I I I I I I I I I I I 1 770 785 800 815 830 845 860 875 890 Wavelength, nm
(c)
Figure 2: Polyester copolymers in vitro study: (a) copolymer discs before hydrolysis (up), at 4th day (middle) and 10th day (bottom) of non-enzyme and enzyme-mediated degradation in buffered physiological solution - (b) test tubes for the hydrolysis experiment, containing 3 disk samples with enzyme (upper row) and control (lower row) on the 4th day of experiment - (c) fluorescence spectra of the ICG incorporated into the copolymer structure - (d) saggital MR-imaging for
the Gd release from the sample
The degradation rate of the copolymers differed in buffered PhS and in FBS. Thus, the hydrolysis caused by enzyme-based solution in FBS proceeded approximately 3 times slower than in PhS. According to Gombotz and Pettit [9], polymers may interact with serum proteins, for example by absorbing them, and this can change polymers properties. We hypothesized that the absorption of serum proteins hinders the availability of ester bonds for enzymes. Figure 3 illustrates the changes in signal to noise ratio (SNR) intensity for MRI and fluorescence intensity (FI) measurements.
The MRI signal in PhS increases nonlinearly due to the swelling and subsequent rapid hydrolysis of the copolymers (Fig.3a). When the enzyme is added, the Gd citrate complex releasing becomes more linear, probably, due to the rapid hydrolysis initiation (Fig.3b). The Gd-citrate release kinetic demonstrates the regular day-by-day rate in FBS (Fig.3c), possibly due to the absorption of serum proteins on the polymer. Complete degradation of the polymer in serum occurs on days 24-28, while in the absence of the enzyme, this process may take 50-70 days. ICG release kinetic is nonlinear in PhS and approximate to linear in FBS (Fig.3e, 3f)), but the ICG-release FI in FBS+Lipase (Fig.3 h) is dramatically higher in comparing to FBS only (Fig.3g). We presumed that pH and different conditions in our solutions impact it optical properties as it was mentioned repeatedly [12].
Buffered PhS {
buffered PhS + 1750 EU of lipase I
FBS I
FBS + 1750 EU of lipase [
Buffered PhS {
buffered PhS + 1750 EU of lipase I
FBSI
FBS + 1750 EU of lipase [
Figure 3: Degradation of copolymers in vitro under the different conditions measured by MRI (upper row) and FI (lower row): (a), (e) in buffered PhS - (b), (f) in PhS with addition 1750 EU of lipase - (c), (g) in FBS - (d), (h) in FBS
with addition 1750 EU of lipase
Polymers weight loss was correlated to the MRI and FI change during the whole cycle of enzyme-mediated hydrolysis (Fig.4). In the swelling phase, both the weight of the polymer and the MRI and fluorescence signals increase, while in the degradation phase, they are inversely proportional - as the weight decreases, the signals increase. The tilt angle indicates the correlation coefficient, which was calculated as 1 for MRI, while for changes in fluorescence it is not, as fluorescence strongly depends on the environment [10].
Figure 4: Correlation between copolymers' weight change and MRI & FI signals normalized to their maxima
during all cycle of hydrolisis in FBS.
The MTT test was used to assess the toxicity of the copolymers to cells. The experimental design is illustrated in Figure 5a. Copolymer samples showed no toxicity in relation to the mouse cell lines (A9, C2C12) under physiological pH conditions. However, it was observed that in the absence of buffering agents, such as HEPES, the presence of the polymer in the medium led to a shift towards aciditification (pH 4.0-4.5) (Fig. 5b), negatively impacting cell survival. No significant changes in viability in respective of the polymer concentration in the wells were observed for fibroblast-like cells A9 (Fig. 5c). The co-incubation of the polyester copolymers with myoblast-like C2C12 cells resulted in the intensification of cell proliferation (Fig. 5d). The similar effect was previously observed by Y. Guo et.al. and explained by the citrate enhancement effect for in vivo skeletal muscle regeneration [11].
Seeding cells onto a plate 16-18 hours in advance
transfer of part i
of the nutrient medium 1
and copolymers
M
r conducting an mtt test 1 incubation for 3 days _J
new plate
previous plate
(c)
(a)
O 4,125 8,25 16,5 24,75 33
Ceipolymers, my
■ C2C12 4 days ■ C2C12 7 days ■ C2C12 10 days ■ C2C12 13 days
(d)
Figure 5: (a) Design of the experiment on the toxicity of copolymers to cells: the copolymer samples contained 0.4mkm- pore polyethylene inserts were inserted to the wells with the pre-seeded cells; after 3 days, the inserst with copolymer and part of the nutrient medium were transferred into the new wells with pre-seeded cells, and the cells from the previous plate were sent for the MTT test - (b) diagram for pH medium's changes dependence on the copolymers weght at the 3rd day of incubation with cells- (c, d) viability of C2C12 myoblast-like cells (c) and A9 fibroblast-like cells (d) in the following by MTT-tests from 4th to 13th days of co-incubation with polyester copolymers
in quantity of 0-33 mg/well.
To evaluate the biocompatibility of the copolymers, 5 mm copolymer discs were implanted into BALB/c mice subdorsal. No signs of toxicity or intoxication were observed during all period of observation. Fluorescence images were obtained using planar imaging (Fig. 6a), MRI images were obtained in two modes: T1-weigted gradient echo (T1 3D GRE) (Fig. 6b) and T2-weigted fast spin echo (T2 FSE) (Fig. 6c). ICG fluorescence was detectable up to day 24-28, as well as polymer disks MRI. Similar to the in vitro experiments, an initial increase in signal was observed, followed by a gradual attenuation over time. (Fig. 6d, 6e).
(d) (e)
Figure 6: MRI and fluorescence viualisation of copolymer-based implants in mice Balb/c: (a) pseudo-color staining of ICG-labeled polyester copolymers implanted to mice - (b) MR images, obtained in T1 mode - (c) MR images, obtained in T2 mode - (d) changes in fluorescent intensity in vivo - (e) changes in MRI signal
intensity in vivo
A side effect observed was the increased acidification of the environment surrounding the polymer, accompanied by aseptic inflammation, which was detected as hyperintensive T2 using MRI in BALB/c mice up to 7-8
days post implantation (Fig.6c). Based on our findings, we hypothesized that improving the biocompatibility of the copolymer could be achieved by reducing its citrate content.
Glycerol is widely used as an optical clearing agent (OCA) in tissues [13, 14]. When tissue is exposed to OCA, due to its reversible dehydration and refractive index matching of the tissue scatterers and interstitial fluid, light scattering decreases and optical transmission through the tissue increases. The kinetics of this process is determined by the time of diffusion of tissue water and OCA molecules in the tissue. Thus, from changing the transmission spectrum of a tissue during the action of an agent on it, one can determine the rate of diffusion of molecules in the tissue [13].
Spectral measurements were carried out ex vivo on samples of rat muscle tissue immersed in OCA with the addition of ICG. Figure 7 shows the spectra and time dependences for a number of characteristic wavelengths of collimated transmittance of a tissue sample. In the spectra, absorption bands of ICG near the wavelength of 800 nm and hemoglobin/myoglobin near 550 nm are clearly visible. Tissue transmittance increases with time due to the action of OCA, but to varying degrees at different wavelengths.
55-, 50 £ 45 §40 1 35 i 30 I 25 § 20 1 15 10 5 0
o
90 m 80 m 70 m 60 m 50 m
20 min
10 min 0 min
50 * 40
0
1 30
E
U)
J 20
<D
| 10 ° 0
750 nm 700 nm 800nm 795 nm
600 nm
500 600 700 800 900
Wavelength, nm
20 40 60
Time, min
80
100
(a) (b)
Figure 7: Spectra (a) and kinetic dependences at characteristic wavelengths (b) of collimated transmittance of rat muscle tissue sample during immersion in OCA (70% glycerol, 25% water and 5% DMSO) with the addition of ICG (5*10~3 mg/ml).
The measured spectra of tissue samples were used to calculate the rate of diffusion of agents using the algorithm described in [13-15]. Theoretically, the time dependence of the collimated transmittance of the sample was determined using the Bouguer-Beer-Lambert law:
'(A, t) = exp{-\pa(X) + ps(X, t)] ■ 0,
(1)
Tc(t) x 1 -expI --
where na(A) and (A,t) are the absorption and scattering coefficients of the tissue sample, respectively, cm-1, l is the
thickness of the sample, cm. The effective diffusion coefficient Da of an agent in tissue can be extracted using the following relationships:
t = 4- , (2)
n Da
where t is the characteristic diffusion time of the agent when it is delivered through both surfaces of the sample (two-side diffusion). The efficiency of optical clearing of the samples was also assessed as the ratio of the difference between the initial (ns 0) and minimum (psmin) values of the scattering coefficient to the initial (psj)) value in the sample:
fis_o~fis_min
OCeff = ■
o
(3)
Since the exogenous chromophore ICG with a characteristic spectrum was added to the main endogenous chromophores of muscle tissue (hemoglobin/myoglobin) (Fig. 7a), the OCeff varies in the range of 66% to 86% depending on the selected wavelength. In general, the efficiency of optical clearing of a muscle tissue sample is quite high. At the wavelength of maximum fluorescence intensity of ICG, equal to 820 nm, the intensity of the transmitted light through a layer of muscle tissue, 0.9 mm thick, increases by approximately 25 times, despite some absorption of ICG at this wavelength.
The values of the diffusion coefficients found from the data presented in Fig. 7 for spectral regions characteristic of the strong influence of glycerol, as an OCA, DGly = (5.8±0.6)10-7 - (6.0±0.5)10-7 cm2/s, and the ICG bands, Dicg= (5.8±0.6)10-7 cm2/s, turned out to be close to each other. This may indicate that glycerol and ICG form a molecular complex and move as a single complex in the tissue.
In conclusion, the study demonstrated the potential of fluorescence imaging and MRI to label and visualize ester copolymers in vitro and in vivo. It was found that hydrolysis is the predominant degradation mechanism in buffer solutions and contributes to enzyme-mediated cleavage. In fetal serum, the presence of proteins slows down biodegradation compared to the buffer. Fluorescent and paramagnetic labels were released in two phases, with the weight of the copolymers correlating with the release. MRI and fluorimetry were used to assess polymer biodegradation in vitro
n n
40 min
30 min
0
and in vivo, providing valuable insights into the degradation process, structural changes, and release of labels and degradation products. These methods contribute to a better understanding of the behavior of copolymer materials in biomedical applications, leading to improved predictability and effectiveness of biomedical devices.
In the course of experimental studies, using collimated transmittance spectroscopy, the diffusion rate of one of the degradation products of implants, such as glycerol, as well as the dye-label of the implant, ICG, was measured in ex vivo muscle tissue.
Acknowledgments
We acknowledge Prof. Alexei Bogdanov (University of Massachusetts Medical School, Radiology, Worcester, Massachusetts, USA) for the concept of this study and for the advice. We acknowledge Prof. Savitsky and his lab (Laboratory of physical biochemistry of RC of Biotechnology of the RAS) for providing fluorescence imaging facilities for the in vivo experiments. Fibroblast-like cell line A9 and myoblast-like cell line C2C12 were obtained from the Shared Research Facility "Vertebrate cell culture collection" (Institute of cytology of the RAS, S-Peterburg, Russia).
Funding
This research was funding by Ministry of Science and Higher Education of the Russian Federation grant № 13.2251.21.0009 from 29.09.2021 (Agreement № 075-15-2021-942).
REFERENCES
[1] L. N. Woodard and M. A. Grunlan, "Hydrolytic Degradation and Erosion of Polyester Biomaterials," ACS Macro Letters, vol. 7, no. 8. American Chemical Society, pp. 976-982, 2018.
[2] B. S. Heidari et al., "A novel biocompatible polymeric blend for applications requiring high toughness and tailored degradation rate," J Mater Chem B, vol. 9, no. 10, pp. 2532-2546, 2021.
[3] M. R. A. V. and R. A. V. Zambrano, J. Pawlak, and R. Venditti, "Effects of Chemical and Morphological Structure on Biodegradability of Fibers, Fabrics, and Other Polymeric Materials," Bioresources, vol. 15, no. 4, pp. 97869833, 2020.
[4] L. Palanikumar et al., "Spatiotemporally and sequentially-controlled drug release from polymer gatekeeper-hollow silica nanoparticles," Sci Rep, vol. 7, 2017.
[5] M. Timusk et al., "Study of triboelectric properties and biocompatibility of some selected polyesters: Toward implantable triboelectric nanogenerator devices," Polymer (Guildf), vol. 281, p. 126088, 2023.
[6] C. Kleber, K. Lienkamp, J. Rühe, and M. Asplund, "Electrochemically Controlled Drug Release from a Conducting Polymer Hydrogel (PDMAAp/PEDOT) for Local Therapy and Bioelectronics," Adv Healthc Mater, vol. 8, no. 10,
2019.
[7] R. S. Petersen, L. H. Nielsen, T. Rindzevicius, A. Boisen, and S. S. Keller, "Controlled drug release from biodegradable polymer matrix loaded in microcontainers using hot punching," Pharmaceutics, vol. 12, no. 11, pp. 1-13, 2020.
[8] X. Fu, L. Hosta-Rigau, R. Chandrawati, and J. Cui, "Multi-Stimuli-Responsive Polymer Particles, Films, and Hydrogels for Drug Delivery," Chem, vol. 4, no. 9. Elsevier Inc, pp. 2084-2107, 2018.
[9] W. R. Gombotz and D. K. Pettit, "Biodegradable Polymers for Protein and Peptide Drug Delivery," 1995.
[10] M. Levitus and S. Ranjit, "Cyanine dyes in biophysical research: The photophysics of polymethine fluorescent dyes in biomolecular environments," Q Rev Biophys, vol. 44, no. 1, pp. 123-151, 2011.
[11] Y. Guo et al., "Bioactive biodegradable polycitrate nanoclusters enhances the myoblast differentiation and in vivo skeletal muscle regeneration via p38 MAPK signaling pathway," BioactMater, vol. 5, no. 3, pp. 486-495,
2020.
[12] M. Attia et al. "Optical properties of indocyanine green under ultrasound treatment" J. Photochem. Photobiol., vol. 2, 2020.
[13] V.V. Tuchin, D. Zhu, and E.A. Genina (Eds.), Handbook of Tissue Optical Clearing: New Prospects in Optical Imaging, Taylor & Francis Group LLC, CRC Press, Boca Raton, FL, 2022.
[14] Tuchina D. K., Shi R., Bashkatov A. N., Genina E.A., Zhu D., Luo Q., Tuchin V.V. Ex vivo optical measurements of glucose diffusion kinetics in native and diabetic mouse skin, J. Biophotonics 8 (4), p. 332-346, 2015.
[15]D.K. Tuchina, I.G. Meerovich, O.A. Sindeeva, V. V. Zherdeva, A. P. Savitsky, A. A. Bogdanov Jr, V. V. Tuchin, Magnetic resonance contrast agents in optical clearing: Prospects for multimodal tissue imaging, J. Biophotonics 13 (11), 2020.
