Morphological observations
In this research, commercial NaY zeolites were selected as the hemostatic components for the in-situ biosynthesis process. As shown in Fig. 2a, the SEM image of NaY zeolites showed micro-nano size with an average diameter of 784.6 ± 137.8 nm and the right histogram displayed the diameter distribution of NaY zeolites (520–1300 nm). The SEM image in Fig. 2b showed the typical 3D nanoscale network structure of freeze-dried BC/zeolite NFAs, consisting of BC nanofibers with an average diameter of 75.2 ± 20.5 nm. Besides, the irregular pores were observed in Fig. 2b, which were formed during freeze-drying process due to the sublimation of ice encapsulated in the BC nanofibrous network (Wu & Meredith, 2014). These pores with an average dimeter of 230.9 ± 77.5 nm and a distribution ranging from 90 to 430 nm, were smaller than the size of NaY zeolites. In Fig. 2c (left), the cross-section of BC/zeolite NFAs exhibited a sandwich-like structure, in which the zeolite particles were tightly encapsulated in interlayer by BC nanofiber layers on both sides. By using a higher magnification (Fig. 2c right), the structure of zeolite particles entangled with BC nanofibers was observed, which might further enhance the binding force between zeolites and BC nanofibers. These results demonstrated a sandwich-like structure that NaY zeolite particles were firmly embedded into BC nanofibrous networks via in-situ biosynthetic process.
In order to further confirm the inherent sandwich-like structure, EDS analysis was performed on the cross-section of the BC/zeolite-Ca NFAs. As shown in Fig. 3, the SEM image was consistent with Fig. 2c, confirming the good stability of the sandwich-like structure. On the other hand, the spot signals in C element mapping formed two well-defined outer-layers, corresponding to the BC layers in the SEM image. Meanwhile, the spot signals distribution in O element mapping indicated that the O was presented in whole BC/zeolite-Ca NFAs. However, the spot signals of Al, Si, and Ca formed a clearly defined meddle-layer in their element mappings, respectively, corresponding to the zeolite-Ca layer in the SEM image.
Characterization of BC/zeolite NFAs
As seen in Fig. 4a, the FTIR spectrum was used to characterize the functional groups of zeolite, BC, and BC/zeolite composites. The characteristic peaks of BC around 3343 cm− 1, 2895 cm− 1, and 1055 cm− 1 were ascribed to -OH, -CH2, and C-O-C stretching vibrations (Li et al., 2017). Compared with BC spectrum, the characteristic peak at 790 cm− 1 was only found in zeolite and BC/zeolite composites spectra, which was associated with Si-O-Al groups in zeolite particles (Salim & Malek, 2016). The crystallinity of the purchased and biosynthesized materials was investigated by XRD (Fig. 4b). The purchased zeolite particles showed intense peaks at 2θ = 6.25°, 15.75°, 23.75°, 27.15°, and 31.50°, which confirmed the typical faujasite (IZA) structure of NaY zeolite phase (Salim & Malek, 2016). Meanwhile, these characteristic diffraction peaks were found in BC/zeolite and BC/zeolite-Ca spectra, and showed no significant changes, indicating that the in-situ biosynthetic process and calcium ion-exchanged process did not affect the crystallinity of NaY zeolite. The biosynthesized BC showed three characteristic diffraction peaks at 2θ = 14.9°, 17.2°, and 23.1°, whereas these peaks were not found in BC/zeolite and BC/zeolite-Ca spectra, which might be attributed to the weaker intensity of BC characteristic peaks and the less content in BC/zeolite composites (Li et al., 2017). Thus, TGA was performed to investigate the respective proportions of BC and zeolite in the BC/zeolite composite (Fig. 4c). As seen in TGA and DTG curves, the weight loss from 30℃ to 200℃ was mainly caused by water evaporation. The decomposition of BC started at around 295℃ and ended at around 400℃, and the weight loss was about 14.2%. In addition, the TGA curves exhibited a minor weight loss from 400℃ to 800℃, which might be due to the presence of -OH groups on the zeolite (Chaibi, Boucheffa, & Bendjaballah-Lalaoui, 2021). According to the TGA, the zeolite loading in biosynthesized BC/zeolite NFAs was 82.6 ± 5.7wt% (measured from five TGA data), indicating the high loading ability via in-situ biosynthesis.
To investigate the stability of the BC/zeolite NFAs, we tested the residual weight of zeolite after three different treatments, including sonication for 10 min, soaking for 24 h, and stirring for 24 h. As shown in Fig. 4d, after processing in the three ways, the residual weights of zeolite were 95.4 ± 2.7%, 99.9 ± 0.1%, and 98.6 ± 1.2%, respectively, manifesting that the zeolite particles were firmly anchored into the BC nanofibers. The swelling ratios of BC, BC/zeolite, and BC/zeolite-Ca NFAs were measured and exhibited in Fig. 4e. BC NFAs displayed rapid water absorption from 0 to 10 min, and sustained absorption to a maximum value (3152 ± 111%) within two hours. For BC/zeolite and BC/zeolite-Ca NFAs, the swelling ratios exhibited similar curves to that of BC NFAs but showed lower values at the same testing moments. Not surprisingly, the embedded zeolite particles with lower absorption than BC significantly increased the weight of BC/zeolite and BC/zeolite-Ca NFAs. In addition, the swelling ratios of BC/zeolite-Ca were slightly lower than BC/zeolite, which may be ascribed to the porosity reduction of BC NFAs during continuous freeze-drying process. As shown in Fig. 4f, we recorded the images of fresh blood dropping onto BC/zeolite-Ca NFAs within 30 s, to further observe the blood absorption behavior of the aerogel. Upon contact, the blood droplet immediately spread onto the BC/zeolite-Ca NFAs and was rapidly and completely absorbed. Remarkably, the average pores size of BC layer (230.9 ± 77.5 nm) was far smaller than hemocytes including RBCs and platelets (Guo, Dong, Bang, & Li, 2021). Therefore, we speculate that the water in the blood is absorbed, but most of the blood components are concentrated and assembly onto the BC/zeolite-Ca NFAs.
BCI and hemolysis assay, and hemocytes adhesion
The BCI assay was used to assess the procoagulant activity of the hemostats, with lower BCI values generally indicating better procoagulant performances (Yin et al., 2020). The BCI values were measured when the whole blood contacted with hemostats for 5 min, and the results were shown in Fig. 5a. In comparison with the blank control (BCI = 100 ± 1.0), the BCI value of BC NFAs decreased to 82.0 ± 1.8. It demonstrated that BC NFAs could absorb the water from blood and concentrate the blood components to promote blood clotting (Edwards & Prevost, 2011). BC/zeolite NFAs displayed a lower BCI value (73.6 ± 1.2) than BC NFAs, which probably caused by the release of Na+ from NaY zeolites (Yu et al., 2021). After calcium ion-exchange process, the BCI value of BC/zeolite-Ca NFAs dropped significantly to 27.3 ± 1.5, displaying the best procoagulant performance in the assay. As we know, zeolite is a kind of microporous aluminosilicate that can accommodate Ca2+ during calcium ion-exchange process, which play a significant role in blood coagulation cascade (Shang et al., 2021). Based on this, we believe that the strong water absorption and Ca2+ release of BC/zeolite-Ca NFAs are two crucial factors in prompting the procoagulant performance. In addition, the BC NFAs were treated via calcium ion-exchange process, labeled as BC-Ca. The BC-Ca exhibited a similar BCI value to BC, manifesting that BC has no capacity for Ca2+ exchange. Thus, the BC-Ca NFAs will not be further evaluated and discussed later.
The hemolysis activity assay was used to evaluate the hemocompatibility of hemostats. In general, an ideal and safe hemostat should induce less than 5% hemolysis after contact with blood cells (Peng et al., 2021). As shown in the inset in Fig. 5b, all the NFAs displayed transparent supernatants similar to PBS (negative control), while the DI water (positive control) displayed a red supernatant. It demonstrated that the RBCs were ruptured in DI water, whereas the RBCs remained intact after contacted with these NFAs or in PBS. Furthermore, the hemolysis ratios of BC, BC/zeolite, and BC/zeolite-Ca NFAs were all less than 5%, manifesting that these NFAs possessed good hemocompatibility.
To further reveal the hemostatic mechanism, the micro morphology of whole blood clot (Fig. 5c), RBCs adhesion (Fig. 6a), and platelets adhesion (Fig. 6b) were observed by SEM. As shown in Fig. 5c, a large number of hemocytes and plasma components aggregated on the surface of BC/zeolite-Ca NFAs, forming a stable blood clot. Besides, massive RBCs (red arrows) and multi-tentacled platelets (blue arrows) were observed in this magnified SEM image. The SEM images in Fig. 6a showed the adhesion of RBCs to BC NFAs and BC/zeolite-Ca NFAs, and more RBCs were observed on BC/zeolite-Ca NFAs than on BC NFAs. We presume that the positively charged Ca2+ released from BC/zeolite-Ca NFAs could significantly enhance the hemagglutination reaction. Furthermore, comparing the morphologies of platelets on BC NFAs and BC/zeolite-Ca NFAs in Fig. 6b, it was obvious that the platelets on BC/zeolite-Ca NFAs deformed and stretched out a lot of spiny pseudopodia. This phenomenon indicated that the platelets could be activated on BC/zeolite-Ca NFAs but not on BC NFAs. In the platelet adhesion test, PRP was obtained from anticoagulant blood, thus Ca2+ released from BC/zeolite-Ca NFAs could neutralize the anticoagulants in PRP and further participate in the activation process of platelets (Yeung, Yamashita, & Prakriya, 2017).
In vitro clotting time assay
In vitro blood clotting performances of BC, BC/zeolite, and BC/zeolite-Ca NFAs were evaluated by monitoring the clotting time of whole blood and plasma (Fig. 7). As seen in Fig. 7a, the clotting time of whole blood was 7.5 ± 0.3 min in the absence of any hemostatic material. After contact with BC and BC/zeolite NFAs, the clotting time decreased to 6.5 ± 0.3 min and 6.4 ± 0.4 min, respectively, which could be ascribed to the strong absorbability of the BC or BC/zeolite NFAs. After contact with BC/zeolite-Ca NFAs, the clotting time dramatically decreased to 2.2 ± 0.3 min, due to BC/zeolite-Ca NFAs could rapidly absorb the water form blood to concentrate the blood components and release Ca2+ to further accelerate the coagulation cascade. The results of plasma clotting time displayed a similar regularity to that of whole blood. As seen in Fig. 7b, the clotting time of plasma was 10.7 ± 0.2 min in the absence of any hemostatic material, which was longer than that of whole blood. This phenomenon could be attributed to the lack of platelets in plasma to participate in the coagulation cascade (Yu et al., 2021). It was worth noting that the plasma clotting time decreased to 7.2 ± 0.4 min and 7.1 ± 0.4 min after adding BC NFAs and BC/zeolite NFAs into the centrifuge tubes, respectively. After adding BC/zeolite-Ca NFAs into the centrifuge tubes, the plasma clotting time decreased more significantly to 2.4 ± 0.2 min. In addition, as seen in Fig. 7b, the optical images displayed the whole blood clot and plasma clots at the test time of 2.5 min. Due to the rapid hemostatic effect of BC/zeolite-Ca NFAs, the stable whole blood and plasma clot were formed and firmly stuck to the surface of the aerogel and the wall of the centrifuge tube, respectively. These results demonstrated that the BC/zeolite-Ca NFAs possessed excellent hemostatic properties to reduce blood loss in emergency situations.
Biocompatibility evaluation
Rat fibroblast cells (L929) were utilized to evaluate the biocompatibility of these NFAs. As seen in Fig. 8a, the cell viability of BC NFAs, BC/zeolite NFAs, and BC/zeolite-Ca NFAs were 94.9 ± 3.3%, 94.4 ± 1.6%, and 94.7 ± 6.9%, respectively. The results indicated that the extracts of the three kinds of NFAs were not cytotoxic. In addition, the morphologies of L929 cells after co-cultured with the extracts of BC/zeolite-Ca NFAs were observed by fluorescence staining. As shown in Fig. 8b, the live cells were stained green and exhibited oval or spindle like morphologies in the two groups (control and BC/zeolite-Ca). The BC/zeolite-Ca group showed a similar live situation of L929 cells to the control group, manifesting that BC/zeolite-Ca NFAs possessed good cytocompatibility. In conclusion, the results of cell viability and fluorescence staining assay prove that the in-situ biosynthetic and Ca2+ exchanged BC/zeolite-Ca NFAs were biocompatible and can be used as hemostats.
Hemostatic mechanisms of BC/zeolite-Ca NFAs
The hemostatic mechanisms involving concentration of blood components and activation of coagulation cascade on the surface of BC/zeolite-Ca hemostat is presented in Fig. 9. Owing to the high absorbability of BC/zeolite-Ca NFAs, the water in blood can be rapidly absorbed and result in the concentration of blood components when blood contacts with the NFAs. Notably, the blood components, such as platelets and prothrombin can easily contact or close proximity to interlayered zeolite-Ca hemostatic agents due to the 3D porous structure of outer BC nanofibers. The granular zeolite-Ca can effectively initiate and propagate the coagulation cascade via the intrinsic coagulation pathway, leading to the formation of fibrin clot (Shang et al., 2021). In addition, calcium ions released from zeolite-Ca can accelerate platelets activation, and the activated platelets rapidly adhere and aggregate blood cells and other blood components to form a firm thrombus. The thrombus that quickly forms at a bleeding wound can arrest flow of blood components to achieve hemostasis.