3.1 Morphology of ECI-PVAHM
Figure 1 presents optical photos of three PVAHMs, the first of which is an ECI-PVAHM made using a custom UND mold (Fig. 1a). During its preparation process, 2 mL of 10 w% PVA aqueous solution containing a certain amount of E. coli was placed in a closed space. Under the action of gravity, the water molecules permeated out from the nanopores of the bottom dialysis membrane. This process typically takes 12 to 24 h to dehydrate and form a film. In this experiment, to accelerate the rapid dehydration of the PVA solution, a micro fan is set to accelerate the airflow and improve the dehydration efficiency. The dry ECI-PVAHM shown in Fig. 1a has a smooth surface, uniform thickness, is hard and tough, and has a pale yellow color due to the many E. coli cells embedded in the film. The PVAHM of the control sample without E. coli was white, transparent, very tough, and hard (Fig. 1b). Another control sample was formed from 10 w% PVA solution spread in polypropylene boxes or on plates and dried by ordinary evaporation. As shown in Fig. 1c, the formed PVA film has an uneven thickness and appears to be of poor quality.
3.2 Swelling behavior
A hydrogel is a network of cross-linked polymer chains surrounded by an aqueous solution. Superabsorbent hydrogels have attracted extensive attention due to special advantages, such as exceptional hydrophilicity, high swelling ratio, biocompatibility, and abundant availability. This experiment simulated the effect of the environment on the swelling of PVAHMs and evaluated its applicability as a biomaterial. The results showed that all samples exhibited high equilibrium swelling ratios, while the swelling ratios of ECI-PVAHMs with different thicknesses increased rapidly in the first half of the testing period (Fig. 2a). At first, the PVAHM swollen in mL of water for 1 h approached the swelling equilibrium very quickly, which because it was the thinnest hydrogel film; the swelling ratio of the PVAHM in 2 mL reached about 450% and continued to swell; the remaining groups showed similar swelling curves, with swelling ratios between 130% and 250%. The swelling equilibrium was reached in 5 h in the 2 mL group, with a swelling rate of 700%, while the 3 mL, 4 mL, and 5 mL groups reached the swelling equilibrium in 6 h, and the final swelling rates were 510, 360 and 300%, respectively. The above results demonstrate that the UND-based bacteria-loaded PVAHM is a highly absorbent hydrogel with a swelling ratio of 300–800% (Fig. 2b). The swelling ratio of the PVAHM is inversely proportional to the thickness of the membrane. Larger amounts of PVA solution and longer periods of oriented nanopore dehydration may increase the degree of crystallization and a decrease in the porosity of the hydrogel film, resulting in decreased volume and, thus, decreased swelling rate.
3.3 Water stability
The stability of ECI-PVAHM in aqueous medium is very important for the proliferation and growth of immobilized E. coli. Figure 3 is a comparison photo of the appearance and light transmittance of bacteria-loaded and non-bacteria-loaded PVAHMs. It can be seen from these images that the ECI-PVAHM is dark and almost opaque (Fig. 3a) due to the many E. coli cells embedded in the hydrogel film, while the ordinary PVAHM is transparent (Fig. 3b). The sizes of the two membranes are almost the same after reaching swelling equilibrium in water. Therefore, the presence of E. coli cells in the PVAHM did not significantly affect the swelling rate of the bacteria-loaded hydrogel film.
This bacterial-loaded PVAHM has excellent swelling properties, but its stability in water is another key factor as a 3D culture scaffold for E. coli. The cast PVAHM prepared by the usual method has poor mechanical properties and will stretch until the film ruptures when placed in water. Therefore, in this experiment, the non-bacteria-loaded PVAHM was placed in 25°C water that was changed every 2 days, and the stability test was conducted for 60 days. The bar graph in Fig. 3c presents the weight loss after 2 months in water for PVAHMs of four thicknesses prepared from different volumes of PVA solution. Since the PVAHM prepared from 1.0 mL of the solution by unidirectional nanopore dehydration is too thin, in this experiment, the PVAHMs resulting from volumes of 2.0, 3.0, 4.0, and 5.0 mL were tested in water before and after 2 months. The residual amount or dissolution rate after swelling equilibrium was compared. In Fig. 3c, after 60 days of swelling in water, the dissolution loss of the first three PVAHMs with different thicknesses is 12–14%, while the thick films prepared with 5 mL of solution appear to have smaller losses (~ 5%). The difference in the dissolution loss rate of these four groups of samples, especially the average of the first three, is about 13%; it seems that the thinner PVA film has slightly higher losses. However, the dissolution loss rates of these four types of membranes are between 5% and 14%, indicating that the hydrogel membrane is mainly composed of a hydrogen-bonded network structure, which is relatively stable and has good stability after 60 days in water.
3.4 Mechanical tensile properties
The mechanical properties of the bacteria-loaded PVAHMs are also an important indicator. Table 1 shows the mechanical properties of the five thicknesses of PVAHMs prepared with different volumes of PVA solutions. In Table 1, the swelling rate of the PVA hydrogel film in the 1.0 mL group is very large, but the mechanical properties of the film are very poor, so it is not suitable for making biofilms. In the four experimental groups of 2.0–5.0 mL, as the volume of PVA solution used during preparation increased, the change in size after swelling did not differ greatly, and the tensile strength increased with increasing volume, except for the final 5.0 mL experimental group, which decreased. However, the elongation at break of these five groups was between 300–390%. The mechanical properties and elongation at break of the PVAHMs prepared with 2.0–4.0 mL of PVA solution were the best. Therefore, in subsequent biofilm preparation, unless otherwise specified, the biofilms were prepared with 2.0 mL of 10 w% PVA aqueous solution.
Table 1
Tensile strength of ECI-PVAHMs prepared from different volumes of solution
PVA
|
After swelling (mm)
|
Max load
|
Tensile strength
|
Elongation at break
|
(mL)
|
width
|
thickness
|
(Fmax, N)
|
(σ, MPa)
|
(%)
|
1.0
|
10.13 ± 0.50
|
0.34 ± 0.03
|
0.34 ± 0.11
|
0.10 ± 0.03
|
305.09 ± 33.76
|
2.0
|
8.38 ± 3.67
|
0.57 ± 0.04
|
2.10 ± 0.94
|
0.67 ± 0.64
|
340.52 ± 26.00
|
3.0
|
9.38 ± 0.21
|
0.80 ± 0.04
|
6.53 ± 1.37
|
0.88 ± 0.21
|
386.91 ± 19.15
|
4.0
|
9.23 ± 0.31
|
1.03 ± 0.08
|
8.36 ± 2.36
|
0.90 ± 0.29
|
359.85 ± 38.67
|
5.0
|
8.83 ± 0.35
|
1.25 ± 0.06
|
7.18 ± 1.42
|
0.66 ± 0.17
|
318.30 ± 62.74
|
3.5 Stress-strain curves
The mechanical properties of hydrogels are one of the key factors for evaluating their suitability for future applications. The stress-strain tensile curves of PVAHMs prepared with different volumes are notably smooth. As shown in Fig. 4a, the maximum tensile strengths of the ECI-PVAHMs of the 2.0, 3.0, 4.0, and 5.0 mL experimental groups were 0.61, 0.95, 1.2, and 0.81 MPa, respectively, while the tensile strength of the ECI-PVAHM of the 1 mL group was only 0.12 MPa. When the tensile modulus of the PVAHM was calculated using the slope of the linear elastic region of the stress-strain curve, as shown in Fig. 4b, the tensile moduli of the 2.0–5.0 mL ECI-PVAHMs were 0.14, 0.19, 0.24, and 0.22 MPa, while the tensile modulus of the 1.0 mL group of PVAHMs is only 36.55 KPa, which is far less than the mechanical properties of the other groups. This indicated that the mechanical properties of the bacteria-carrying PVAHM prepared with 1.0 mL 10% PVA aqueous solution were poor, and it cannot meet the mechanical properties required for the proliferation of E. coli cells. When other conditions remain unchanged, increased PVA thickness will also increase the tensile strength, reaching the maximum at 4.0 mL, while the tensile strength of the 5.0 mL group decreases. Biofilms were generally prepared with 2.0–4.0 mL of 10 w% PVA aqueous solution.
3.6 Efficiency and stability of cyclic proliferation
To further evaluate the viability and proliferation stability of immobilized bacteria on the porous network scaffold structure, a cyclic culture test was performed every 12 h on the bacteria-loaded PVAHM (ECI-PVA) to detect the proliferation and vitality of the bacteria immobilized in the membrane. The culture medium was refreshed two times each day, and the absorbance of the replaced medium was detected at 600 nm for 40 consecutive cycles, for a total of 20 days of continuous culture and proliferation tests, and the absorbance value after the first culture on the first day is referred to as 100%. The relative bacterial proliferation efficiency for each assay was calculated and plotted. As shown in Fig. 5, after 40 cycles of repeated culturing, the relative viability of the E. coli immobilized on the hydrogel membrane remained at about 91%. Therefore, in this experiment based on the use of UND to prepare the ECI-PVAHM, the size of the porous network and pore size can not only firmly fix the rod-shaped cells of E. coli but also fully ensure the original life activities, biological activities, and strong proliferative capacity of E. coli cells.
3.7 Fluorescence observation of E. coli cells
In this experiment, EGFP-labeled E. coli were used. Therefore, after repeated cell culturing and cyclic proliferation, the bacterial films were observed by sunlight and fluorescence after the first 12 h cycle of culturing and the fortieth cycle (20 d). The proliferation efficiency and viability of this EGFP-labeled ECI-PVAHM or cell can be measured. Figure 6a shows the optical photo of the bacteria-loaded PVAHM after the first 12 h culture. Under an inverted microscope with a magnification of 40⋅, sunlight can pass through the PVAHM, and the bacteria appear as aggregates. Under the light transmitted through the green filter, the EGFP-labeled engineering bacteria emitted a strong green fluorescence at a wavelength of 510 nm since the 0.2-mm-thick hydrogel membrane immobilized many bacteria. The green fluorescence filled the entire ECI-PVAHM (Fig. 6b). Figures 6c-d show the E. coli cells that proliferated and were secure in the culture medium after the fortieth cycle (20 days) of culturing of the biofilm. Immobilized E. coli cells in the membrane last cell division, proliferation, and secretion of bacteria into the culture medium through the porous hydrogel network. The engineered bacterial cells and their aggregates were easily distinguished under a magnification of 400⋅ in sunlight (Fig. 6c), while the bright green fluorescence emitted by these living engineered bacteria could be observed under a green filter (Fig. 6d). These results show that in this ECI-PVAHM network structure based on UND technology, after repeated division and proliferation of immobilized E. coli for 20 days, the E. coli remain highly active. The hydrogel network structure had few adverse effects on the life and proliferation of the cells.
3.8 Continuous-flow culture system of immobilized E. coli cells
A continuous-flow culture system was built for the continuous culture and proliferation of E. coli immobilized in PVAHM. It comprises a temperature-controlled shaker, a stoppered culture bottle with upper and lower mouths, a digital constant-flow pump, a sterile culture liquid pool, a silicone tube, and a fixed clip. The entire culture system is controlled by a mobile phone with the APP software installed. As shown in Fig. 7, several pieces of the bacteria-loaded PVAHM were placed in the culture bottle, shaken for constant temperature cultivation, and the fresh culture solution was quantitatively (timed length) pumped through a digital constant-flow pump. The culture bottle was shaken and cultured at a constant temperature, so that the immobilized E. coli cells in the PVAHM continued to proliferate. After a certain culturing period, the culture solution was pumped out through a constant current pump and pumped into the culture recovery bottle, and the cell proliferation was continuously cultured for one cycle. Then, the operation was repeated, and the fresh culture solution was quantitatively (timed length) pumped into the culture flask through the digital constant current pump and entered the next round of cell culturing and proliferation.
3.9 Microstructure observations
In the UND, 2.0 mL 10% PVA aqueous solution (2.0 mL) will self-assemble the PVA molecular segments into an ordered molecular structure, and the inter- and intra-chain hydroxyl groups form more and stronger hydrogen bonds. After the PVA solution is added to the mold, the bottom dialysis membrane is the only connection to the atmosphere to dehydrate the water downward and reduce the water content. The speed of the final film formation mainly depends on the temperature, RH, and air velocity outside the nanofiltration membrane at the bottom of the mold during dehydration. The preparation conditions in this experiment were controlled at 25°C and 60% RH. A micro fan was added under the mold, but it took about 12 h to dry and form a film. In order to observe the formation of the surface and internal microstructures of the PVAHM, as well as the entrapment and immobilization of E. coli in the hydrogel, samples were taken at two time points, 3 and 9 h before the film was completely formed. These samples were fixed in liquid nitrogen, then transferred to a vacuum freeze dryer for direct lyophilization, and the film-forming samples were processed after 12 h, according to Section 2.6. Finally, the surface structures of the hydrogel film and the microstructures of the longitudinal section or the fracture surface were observed.
Figure 8 shows the cross-sectional SEM images of the UND 3 h before the formation of the ECI-PVAHM. Figures 8a and b are the cross-sections of the 3 h samples at 500⋅ and 2,000⋅, respectively. A typical neatly arranged longitudinal porous network is present, as well as a layered porous structure. The network pore size is ~ 2 µm, and the pore size will further reduce as the dehydration continues. Then, in the sample collected 9 h after dehydration, the concentrated liquid surface of the PVA aqueous solution was magnified by 1,000⋅ and observed by SEM. The dots and rod-shaped objects were almost uniform and dense. These rods are 1–2 µm in length and about 0.2–0.5 µm in diameter; these are E. coli bacteria embedded in the PVAHM (Figs. 8c-d). Figures 8e-f are the cross-sectional images of the ECI-PVAHM samples (12 h) after film formation at 2,000⋅ and 5,000⋅. After further magnification of 10,000⋅ and 20,000⋅ (Fig. 8g-h), the E. coli cells appeared embedded in the pores of the porous hydrogel network (the black part indicated by the red arrow). As a result, these ECI-PVAHMs showed a very regular porous network structure on the surface, with pores about 0.2–1 µm in diameter. These E. coli cells are embedded and confined in these networks and cultured from above. According to the results of the proliferation test, the fixation by these networks hardly affects the normal cell life activities of E. coli, and it is a suitable matrix for cell division and proliferation.
3.10 FTIR spectra
In the FTIR spectrum of the ECI-PVAHM, the infrared absorption peak at 3279 cm− 1 is attributed to the stretching vibration of non-bonded O-H, which reflects the degree of hydrogen bonding between the intramolecular and interchain hydroxyl groups in the PVAHM. In Fig. 9a, the PVAHMs with five different bacterial loadings (0.05, 0.1, 0.2, 0.4, and 0.5 mL) do not show obvious shifts in this absorption band, despite 4.75⋅10− 8–10− 9 CFU E. coli cells embedded and immobilized in the film. Therefore, the E. coli did not significantly affect the secondary structure of ECI-PVAHM. The absorption peak at 2942 cm− 1 is assigned to the C-H stretching vibration, the absorption peak at 1143 cm− 1 belongs to the C-H stretching vibration peak, which is related to the crystallinity of PVA, and the absorption peak at 1083 cm− 1 is the C-O stretching vibration associated with the C-C stretching vibration. These vibrational peaks, that is, the antisymmetric stretching vibration peaks of O-C-C, belong to the amorphous phase. The above-mentioned infrared spectral analysis showed that the PVA hydrogel film does not significantly affect the shift of these absorption bands. This also indicates that 4.75⋅10− 8–10− 9 CFU E. coli cells should be immobilized in the porous network structure of PVA hydrogels formed by UND.
3.11 X-ray diffraction
The bacteria-loaded PVAHMs prepared by UND were analyzed by X-ray diffraction to determine whether the immobilized E. coli in the membrane affected the crystal structure of the PVAHM. The PVA hydrogel film comprises three main parts: the crystalline phase, the amorphous phase of the swollen particles, and water. The high crystallinity of the PVA hydrogel film is due to the many hydroxyl groups of the PVA molecule, and hydrogen bonds are easily formed between these hydroxyl groups within and between the chains to form a crystalline region. These results show that oriented nanopore dehydration can induce the formation of PVA hydrogel films with high crystallinity and excellent mechanical properties; more and stronger hydrogen bonds are generated between the intra- and interchain hydroxyl groups of the PVA molecules. Since the FTIR spectral analysis showed no significant differences due to different concentrations of bacteria in the membrane structure, X-ray diffraction analysis was performed only on the sample with the highest concentration of bacteria. Figure 9b shows that the PVAHM exhibited the main crystal peak of PVA at 2θ = 19.37°, corresponding to monoclinic crystal symmetry. Two smaller characteristic peaks were found at 2θ = 11.78° and 40.53°. After many E. coli were embedded and immobilized in the PVA hydrogel, the basic crystal structure of the membrane did not change significantly.
3.12 Thermal performance
PVA hydrogel is a physically cross-linked gel, and the molecular chains form microcrystalline regions through hydrogen bonds; that is, the physical cross-linking points form a three-dimensional network. These cross-linking points vary with external conditions, such as temperature. The TG curves of ECI-PVAHMs with different levels of bacteria-loading showed three similar weight loss regions25 . The first region occurred from 20–180°C due to the removal and evaporation of weakly bound water; the weight loss was about 5.2%. There was no evident difference due to the number of bacteria present in these ECI-PVAHMs (Fig. 10). The second region (180–450°C) contained two stages of weight loss and is divided into two stages. The differences between these samples were not evident in this stage, which contained three peaks on the DTG curve. The peak in the region of 50–190°C was due to the evaporation of physically weakly bound water and chemically strongly bound water (due to relaxation in the PVA crystalline domains); the weight loss of the film was about 4.56 w%. The second transition zone is around 260–410°C and resulted from the degradation of the PVAHM (the melting of the crystalline domain, i.e., the breaking of hydrogen bonds); two degradation peaks appeared at this stage around 290°C and 360°C, and the differences between the samples were not significant. However, hydrogel membranes with a maximum bacterial load of 0.5 mL at 290°C and 360°C were combined into a single membrane. The peaks of the third stage of these four samples did not differ; they all occurred around 435°C and were attributed to the main chain scission (or carbonization) of the PVA film, which is almost completely carbonized in this stage. There was no difference between these PVAHMs with or without bacterial loading in the DSC curves. Therefore, from the above thermal analysis results, the amount and presence or absence of bacteria did not influence the thermal properties of the ECI-PVAHM.