具体实施方式：
[0027]The bioprinting platform employs 3D printing to fabricate optically-tunable photosynthetic matter that mimics coral tissue and skeleton morphology with micron-scale precision (FIGS. 1A-1D). The inventive approach allows replication of any coral architecture, examples of which are shown in FIGS. 7A-7H, providing a variety of design solutions for augmenting light propagation. Fast-growing corals of the family Pocilloporidae are particularly relevant for studying light management. Despite high algal cell densities in their tissues (1×106 cells per cm2 surface area), the internal fluence rate distribution is homogenous, avoiding self-shading of the symbiotic microalgae. The photon distribution is mainly managed by the aragonite skeleton, where light leaks out of the skeleton and into coral tissue, supplying photons deep within the corallite. Additionally, light can enter the coral tissue more easily than it can escape, as low angle upwelling light is trapped by internal reflection due to refractive index mismatches between the coral tissue and the surrounding seawater. The inventive approach mimics these light management strategies to produce a bionic coral formed from sustainable polymers that are highly effective in enhancing microalgal light absorption and growth.
[0028]To precisely control the scattering properties of the bio-inspired artificial tissue and skeleton, a 2-step continuous light projection-based approach, described below, is employed for multilayer 3D bioprinting. Optimization of the printing approach involved balancing between several parameters including printability (resolution and mechanical support), cell survival, and optical performance. The artificial coral tissue constructs were fabricated with a novel bio-ink solution, in which the symbiotic microalgae (Symbiodinium sp.) were mixed with a photopolymerizable gelatin-methacrylate (GelMA) hydrogel and cellulose-derived nanocrystals (CNC), the latter providing mechanical stability and allowed tuning of the tissue's scattering properties. Similarly, the artificial skeleton was 3D printed with a polyethylene glycol diacrylate-based polymer (PEGDA) doped with CNC.
[0029]Key goals to be achieved in material design were: 1) high microalgal cell viability and growth; 2) microscale printing resolution; and 3) optimization of light scattering and biomechanical parameters including material stiffness, porosity and molecular diffusion. The photo-induced, free radical polymerization mechanism underlying the 3D printing technique allowed precise control the mechanical properties via modulating the crosslinking density of the polymerized parts. Any material and fabrication parameters (e.g., light intensity, exposure time, photoinitiator concentration, material composition) that affect the crosslinking density can be employed to tune the mechanical properties of the printed parts. Initially, different concentrations of prepolymer and photoinitiator combinations were evaluated, including glycidal methacrylate-hyaluronic acid (GM-HA), gelatin methacrylate (GelMA), polyethylene glycol diacrylate (PEGDA), and poly(lactic acid), together with the photoinitiators Irgacure 651 and lithium phenyl-2,4,6 trimethylbenzoylphosphinate (LAP). Cell viability and growth were higher in GelMA compared to PEGDA (data not shown), possibly due to favorable diffusion characteristics of GelMA due to its highly porous microstructure, while PEGDA has stronger mechanical stiffness. To take advantage of the properties of both materials, we combined PEGDA with GelMA to make a mechanically robust and tunable hydrogel. GelMA was initially doped with graphene oxide, which enhanced mechanical stability but limited light penetration and cell growth. To avoid UV damage to the algae, photopolymerization is preferably induced using light within the visible spectra. While the 3D printing steps described herein used 405 nm light, the use of visible light for initiating photopolymerization is widely reported in the literature and selection of alternative wavelengths would be within the level of skill in the art.
[0030]To optimize light scattering, we first mixed the hydrogel with different concentrations of SiO2 particles (Sigma-Aldrich, USA) that were in a size range (about 10 μm) to induce broadband white light scattering with high scattering efficiency. However, when mixed into the hydrogels, the SiO2 particle showed a vertical concentration gradient related to the particle sinking speed in the gel. Instead, we used cellulose nanocrystals (CNCs), which exhibit suitable light scattering, mechanical properties and low mass density. CNCs can be considered as rod-shaped colloidal particles (typical length of 150 nm and a width of a few nm in diameter), which have high refractive index (about 1.55 in the visible range). CNCs have been the subject of increasing interest in photonics due to their colloidal behavior and ability to self-assemble into cholesteric optical films. In the 3D bioprinted coral skeleton samples that contain 7% CNCs (w/v), we found that CNCs aggregated to form microparticles with a size range of 1-10 μm. These aggregated microparticles are highly efficient white light scatterers (FIG. 8A). In contrast, the 3D bioprinted bionic coral tissue constructs contained only 0.1% CNCs (w/v), and we did not observe any aggregated CNC microparticles.
[0031]The printing polymer (bio-ink) for the bionic coral tissue constructs was made up of final concentrations of: Marinichlorella kaistiae KAS603 (1×106 cells mL−1), GelMA (5% w/v), LAP (0.2% w/v), food dye (1% v/v), PEGDA (6000 Da; 0.5% w/v), CNC (0.1% w/v), and artificial seawater (ASW; 93.2%). The food dye (yellow, from Wilton® Candy Colors) was added to limit the penetration of polymerization-inducing light into the bio-ink. This leads to higher light absorption relative to scattering and enhanced the spatial resolution of the printing. The food dye is non-toxic and diffuses out after 24 hr.
[0032]To create a digital mask of natural coral surfaces, a spectral-domain (SD) optical coherence tomography (OCT) system (Ganymede II, Thorlabs GmbH, Dachau, Germany) was used to image living corals. FIGS. 7A-7H are USB camera images and respective optical coherence tomography scans of natural corals (FIGS. 7A, 7C, 7E and 7G) and 3D printed replica (FIGS. 7B, 7D, 7F and 5H). Skeletal 3D printed constructs were imaged with an environmental SEM, while 3D printed tissue constructs were photographed with a microscope camera. Scale bar=1 mm (7A, 7B, 7D, 7E, 7G) and 500 μm (7C, 7F, 7H).
[0033]The OCT system was equipped with a superluminescent diode (centered at 930 nm) and an objective lens (effective focal length=36 mm) (LSM03; Thorlabs GmbH, Dachau, Germany) yielding a z-resolution of 4.5 μm and a x-y resolution of 8 μm in water. The imaged coral species (Pavona cactus, Stylophora pistillata, Pocillopora damicornis, Favites flexuosa) were maintained at the Centre Scientifique de Monaco, and corals were imaged under controlled flow and irradiance conditions. For OCT imaging of bare coral skeletons, the living tissue was air brushed off the skeleton. The skeleton was carefully cleaned before imaging the bare skeleton in water. OCT scanning was performed as described previously.
[0034]OCT data was extracted as multiple 16-bit TIFF image stacks and imported into MATLAB® (Matlab 2018a). Image acquisition noise was removed via 3D median filtering. Segmentation of the outer tissue or skeletal surface was done via multilevel image thresholding using Otsu's method on each image of every TIFF stack. The binary files were exported as x,y,z point clouds and converted to a st/file format, which could be sliced into 2D image sequences for bioprinting. If the generated st/files showed holes in the surface mesh, these holes were manually filled using Meshlab (Meshlab 2016).
[0035]The bionic coral design was developed as an optimization between algal growth rates, optical performance and the outcome of optical models (FIGS. 2A-2D, 3A-C; 4A-4C; FIGS. 8A-8F, FIG. 9). The final bionic coral was designed in CAD software (Autodesk 3ds Max, Autodesk, Inc, USA) and was then sliced into hundreds of digital patterns with a custom-written MATLAB program. The digital patterns were uploaded to a digital micromirror device (DMD) in sequential order and used to selectively expose the prepolymer solution for continuous printing. A 405-nm visible LED light panel was used for photopolymerization. A digital micromirror device (DMD) consisting of 4 million micromirrors modulated the digital mask projected onto the prepolymer solution for microscale photopolymerization.
[0036]The basic components of a 3D printing platform 100 for use in an exemplary embodiment of the invention are illustrated in FIG. 11: a light source 10, a computer controller/processor 12, which performs sliced image-flow generation, i.e., “virtual masks” 11, and system synchronization, a digital micromirror device (DMD) chip 13 for optical pattern generation, a projection optics assembly 14, and an multi-axis stage 15 for sample position control. The DMD chip 13 modulates the light and projects the digital pattern generated via computer 12 based on a custom-designed computer-aided design (CAD) model onto the photopolymer solution. The optical pattern is projected through optical lenses 14 and onto the photosensitive biomaterial 16 to fabricate a 3D scaffold. Complex 3D structures are fabricated through a continuous, layer-by-layer polymerization process that is synchronically controlled using a motorized multi-axis stage 15.
[0037]An appropriate light source 10 for use in the 3D printing system can be selected from different sources including a laser (CW or pulsed), arc lamp, and an LED source, which may include an array of LEDs emitting at a single wavelength or across a range of wavelengths. The light source 10 may include controllable parameters, responsive to the computer controller/processor 12, including intensity, iris, aperture, exposure time, shutter, and wavelength. Selection of appropriate operating parameters will depend on the materials used and the desired characteristics of the scaffold and will be within the level of skill in the art.
[0038]The continuous movement of the DMD was synchronized with the projected digital mask to create smooth 3D constructs that are rapidly fabricated without interfacial artifacts. To print the bionic coral, a 2-step printing approach was developed. In the first step, the PEGDA bio-ink was used to print the coral inspired skeleton. The resulting hydrogel was attached to a glass slide surface, washed with DI water and then dried with an air gun. In the second step, the algal cell-containing bio-ink for tissue printing was then injected with a pipette into the skeletal cavities in order to fill the air gaps. The gap-filled skeletal print was repositioned at the identical spot on the bioprinter, and the bionic coral tissue mask was loaded. The z-stage was moved such that the surface of the skeletal print touched the glass surface of the bioprinter.
[0039]Based on optimization via experiments and optical simulations (FIGS. 2A-2D, 3A-3C and 4A-4C), the functional unit of the artificial skeleton was an abiotic cup structure, shaped like the inorganic calcium carbonate corallite (1 mm in diameter and depth) and tuned to redistribute photons via broadband diffuse light scattering (scattering mean free path=3 mm between 450-650 nm) and a near isotropic angular distribution of scattered light (FIG. 4A, FIGS. 8A-8F), similar to the optical properties of the skeleton of fast-growing intact corals. The coral-inspired tissue had cylinder-like projections (200 μm wide and 1 mm long) radially arranged along the periphery of the corallites mimicking coral tentacles, which serve to enhance surface area exposed to light (FIG. 2A). We designed the bionic coral tissue to have a forward scattering cone (FIG. 4A), which enabled light to reach the diffusely backscattering skeleton (FIG. 2A).
[0040]The bionic coral disclosed herein increased the photon residence time as light travelled through the algal culture (FIG. 3A), consequently enhancing the chance of light absorption for photosynthesis by algae located deeper within the structure. As photons travelled through the bionic skeleton, they were redirected into the photosynthetic tissue. The contribution of the scattered light increased with depth, effectively delivering light to the deepest part of the bionic coral (FIG. 4B, FIG. 9). This photon augmentation strategy led to a steady increase of irradiance with tissue depth, which counter balanced the exponential light attenuation by photopigment absorption (FIG. 3C). Compared to a flat slab of biopolymer (GelMA) with the same microalgal density (5.0×106 cells/mL), the fluence rate (for 600 nm light) measured in the photosynthetic layer of the bionic coral was more than 1.5-fold enhanced at 750 μm depth due to the optimised scattering properties of the bionic coral tissue and skeleton, as shown by FIGS. 4B and 4C.
[0041]Three microalgal species were chosen for inclusion in 3D bioprinted polymers: dinoflagellates belonging to the genus Symbiodinium, the green alga Marinichlorella kaistiae, and the diatom Thalassiosira pseudonana. Stock cultures of Symbiodinium strains A08 and A01 (obtained from Mary Coffroth, University of Buffalo) were cultured in F/2 medium in a 12h/12h light:dark cycle under an irradiance (400-700 nm) of 200 μmol photons m−2 s−1. Wild type M kaistiae strain KAS60319 were obtained from Kuehnle AgroSystems, Inc. (Hawaii) and were cultivated at 25° C. in artificial seawater (ASW) medium30 under continuous light from cool white fluorescent lamps (80 μmol photons m−2 s−1). Stock cultures were harvested during exponential growth phase for use in bioprinting.
[0042]In order to evaluate the growth of a commercially-relevant microalgal species in the inventive bionic coral, we cultured the green alga Marinichlorella kaistiae KAS603, the results of which are plotted in FIGS. 5A-5D. Although the main focus of the inventive bionic coral design was to improve light management, it also allows for growing microalgae without the need for energy-intensive turbulent flow and mixing, which is otherwise required for optimal nutrient and light delivery in photobioreactors. This is achieved by the combination of the bionic tissue and skeleton replica morphology, the tissue mechanical properties (average Young's modulus, E=4.3 kPa, FIG. 10) and its porosity (pore size diameter=5-40 μm). We grew M kaistiae KAS603 under no-flow conditions and low incident irradiance (Ed=80 μmol photons m−2 s−1) in our bionic coral, where it reached algal cell densities of >8×108 cells mL“′ by day 12. In the FIG. 5A plot of M kaistiae KAS603 growth in bionic coral, data are means (±SEM, n=3-6 bionic coral prints), and darker and lighter shaded areas indicate 95% confidence and prediction intervals, respectively. This is about one order of magnitude higher than the maximal cell densities reported for this algal species when grown in flasks under continuous stirring. Despite such high algal cell densities, irradiance did not limit growth at depth, and about 80% of the incident irradiance remained at 1 mm depth within the bionic coral tissue construct. FIG. 5B plots vertical attenuation of fluence rate (E0 at 675 nm) at the beginning (day 1) and end of the performance test (day 12). Measurements were performed with O2 microsensors at the center of the corallite cup surface (closed symbols/solid lines) and at a vertical depth of 300 μm (open symbols/dashed lines). Symbols are means (±SEM, n=3-6 bionic coral prints), lines are curve fits.). In comparison, biofilm-based photobioreactors show exponential light attenuation characterized by a virtual depletion of irradiance within 200-300 μm of the biofilm thickness. We observed that M kaistiae KAS603 grew in the bionic tissue as dense aggregates (sphericity 0.75±0.09 SD, diameter=30-50 μm; FIGS. 6A-6C). Algal photosynthesis within the tissue construct yielded a net photosynthetic O2 production of 0.25 nmol O2 cm−2 s−1 at the polyp tissue surface. FIG. 5C plots net photosynthetic rates at day 5, 8 and 11, where lines represent curve fits. Gross photosynthesis within 8-day old bionic coral polyps was enhanced at a depth of 300 μm compared to gross photosynthesis rates measured at the surface of the bionic coral tissue. FIG. 5D plots gross photosynthetic rates at day 5 (dot) and day 8 (rectangle) for the polyp surface (solid) and the polyp depth (open).
[0043]Bionic corals harboring Symbiodinium sp. or M kaistiae KAS603 were cultured under similar conditions as the respective algal stock cultures. Prior to bioprinting, the bio-ink for printing bionic coral tissue constructs was inoculated with cell densities of 1×106 cells mL−1 from exponentially growing cultures. We performed growth experiments with 35 bionic corals harboring M. kaistiae KAS603. The bionic corals were transferred to 6-well plates filled with 3 mL of ASW medium containing broadband antibiotics (penicillin/streptomycin, Gibco) at a concentration of 1:1000. All prints were illuminated with an incident downwelling irradiance (400-700 nm) of 80 μmol photons m−2 s−1 provided by LED light panels (AL-H36DS, Ray2, Finnex) emitting white light. The prints were grown without mixing at 25° C. The ambient growth medium was replenished at day 5 and day 10. Degradation of GelMA-based tissue occurred after about 10-14 days when bacterial abundance was kept low via antibiotic treatment. Such degradation kinetics can be advantageous for more easy harvesting of the highly concentrated microalgae that are contained within the hard PEGDA-based skeleton. In order to obtain values of algal productivity comparable with previous studies we produced an additional subset of bionic corals in slab geometry (FIG. 14, discussed below). For these experiments, bionic corals were illuminated with an incident irradiance of 150 μmol photons m−2 s−1 and supplied with filtered air supplied by an air pump.
[0044]The angular distribution of transmitted light was measured using an optical goniometer. The samples were illuminated using a Xenon lamp (Ocean Optics, HPX-2000) coupled into an optical fiber (Thorlabs FC-UV100-2-SR). The illumination angle was fixed at normal incidence and the angular distribution of intensity was acquired by rotating the detector arm with an angular resolution of 1°. To detect the signal, a 600 μm core optical glass fiber (Thorlabs FC-UV600-2-SR) connected to a spectrometer (Avantes HS2048) was used. To characterize the optical properties, the total transmitted light was measured for different sample thicknesses using an integrating sphere. The samples were illuminated by a Xenon lamp (Ocean Optics, HPX-2000) coupled into an optical fiber (Thorlabs FC-UV100-2-SR), and the transmitted light was collected with an integrating sphere (Labsphere Inc.) connected to a spectrometer (Avantes HS2048). In the case of the skeleton-inspired samples, where the light is scattered multiple times before being transmitted, the light transport can be described by the so-called diffusion approximation. In this regime, the analytical expression, which describes how the total transmission (T) scales with the thickness (L) for a slab geometry, is given as:
T=1lasinh(ze×ltla)sinh(ze×ltla)sinh(L+ze×ltla)(1)
where Ia, It and ze are the absorption length, the transport mean free path and the extrapolation length, respectively. Here, ze quantifies the effect of internal reflections at the interfaces of the sample in the estimation of Ia and It. We quantified ze by measuring the angular distribution of transmitted light, P(μ), which is related to ze by the following equation:
P(μ)=μze+μ12ze+13(2)
where μ is the cosine of the transmission angle with respect to the incident ballistic beam. The theoretical fit is shown in Figure S2C and led to a value of ze=(1.32±0.12). Once the extrapolation length was estimated, the values of Ia and It could be calculated with Eq. (1) (FIGS. 8D, 8E). This was done with an iteration procedure to check the stability of the fit, as described previously. In the bionic coral tissue, the scattering strength of the material is too low and the diffusion approximation cannot be applied. In this regime, the extinction coefficient can be estimated using the Beer-Lambert law (FIG. 8F).
[0045]The refractive index (n) of the bioprinted bionic coral tissue was determined with the optical goniometer to characterize the Brewster angle (θB). A half circle of the material was printed with a diameter of 2 cm and a thickness of z=5 mm. The Brewster angle was calculated according to Snell's law:
n=sin(θi)sin(θr)=sin(θi)sin(θ90-i)=tan(θi)(3)
and Brewster's law:
[0046]θB=arctann2n1(4)
where θi is the angle of incidence, and θr is the angle of refraction. n1 and n2 are the refractive indices of the medium and the surrounding medium, respectively. For the coral-inspired tissue θB ranged between 54.0° and 55.0° yielding a refractive index of n=1.37-1.40.
[0047]Tetrahedral meshes were generated via Delaunay triangulation using the MATLAB based program Iso2mesh that calls cgalmesh. Meshing was performed with different mesh properties varying maximal tetrahedral element volume and Delaunay sphere size in order to optimize simulation efficiency. Settings were optimized for a Delaunay sphere of 1 (10 μm) and a tetrahedral element volume of 5 (50 μm). Generated tetrahedral meshes were used as source architecture for a mesh-based 3D Monte-Carlo light transport simulation (mmclab). The model uses the generated tetrahedral mesh and calculates photon propagation based on the inherent optical parameters, the absorption coefficient μa [mm−1], the scattering coefficient μs [mm−1], the anisotropy of scattering g [dimensionless] and the refractive index n [dimensionless]. The optical parameters were extracted via integrating sphere measurements (see above) and were used to calculate time-of-flight photon propagation in the bionic coral. The probe illumination was a collimated point source with varying source positions.
[0048]To evaluate the mechanical properties of bionic tissue, the Young's modulus of the bionic coral tissue was evaluated with a microscale mechanical strength tester (Microsquisher, CellScale). Each sample was preconditioned by compressing at 4 μm s−1to remove hysteresis caused by internal friction. The compression test was conducted at 10% strain with a 2 μm s−1 strain rate. Cylindrical constructs were 3D printed using the same bio-ink as used to print bionic coral tissue. The Young's modulus was calculated from the linear region of the stress—strain curve. Three samples were tested, and each sample was compressed three times.
[0049]Cell density was determined at the beginning of the experiment (day 0) and then at day 3, day 6, day 10 and day 12 of the growth experiments. To determine cell density, the construct was removed from the growth medium, and any remaining solution attached to the construct was removed with a Kimwipe. Each construct was transferred to a 1.5 mL microfuge tube and the hydrogel was dissolved via adding 600 μL trypsin solution (0.25% Trypsin/EDTA) under incubation at 37° C. for 40 min. This procedure removed the microalgal cells from the matrix allowing for cell counting via a haemocytometer. The accuracy of this approach was verified by printing known cell densities (from liquid culture) and comparing it to the trypsin-based estimates yielding a deviation of <3%. However, the matrix itself is biocompatible and non-toxic and does not need to be removed to harvest algal biomass. Additionally, as the harvesting of lipids and bioproducts (e.g., pigments) relies on solvents that can diffuse into the scaffold, matrix degradation is not required for extraction.
[0050]To compare our cell density estimates with ash free dry weight (AFDW) of algal cell biomass [g], which is a commonly used metric in biofuels research, we determined AFDW using methods described previously. AFDW was on average 3.47×10−11 g cell−1 (±4.6×10−13 SE). The maximal growth rate was obtained from readings of Day 10 and Day 12, yielding 1.47×1011 cells L−1 day−1 or 5.1 g L−1 day−1. The aerial productivity was extrapolated to g m−2 day−1 by accounting for the area occupied by one bionic coral (6 mm in length and width) and the measured productivity per bionic coral.
[0051]Photosynthetic performance of the bionic corals was characterized using Clark-type O2 microsensors (tip size=25 μm, response time <0.2 s; OX-25 FAST, Unisense, Aarhus, Denmark). Net photosynthesis was measured via linear O2 profiles measured with O2 microsensors from the surface into the overlying diffusive boundary layer. The sensors were operated via a motorized micromanipulator (Pyroscience, Germany). The diffusive O2 flux was calculated via Fick's first law of diffusion for a water temperature =25° C. and salinity=30 using a molecular diffusion coefficient for 02=2.255×10−5 cm2 s−1. Gross photosynthesis was estimated via the light-dark shift method. A flow chamber set-up provided slow laminar flow (flow rate=0.5 cm s−1) and a fiber-optic halogen lamp (Schott KL2500, Schott, Germany) provided white light at defined levels of incident irradiance (400-700 nm) (0, 110, 220, and 1200 μmol photons m−2 s−1). Photosynthesis-irradiance curves were fitted to an exponential function42.
[0052]The fluence rate (=scalar irradiance), E0, within the bionic coral was measured using fiber-optic scalar irradiance microsensors with a tip size of 60-80 μm and an isotropic angular response to incident light of ±5% (Zenzor, Denmark). Fluence rate measurements were performed through the tissue at a vertical step size of 100 μm using an automated microsensor profiler set-up as described previously. Depth profiles were measured from the planar tissue surface (i.e. areas distant from the tentacles) into the center of the bionic corallite. Fluence rate was normalized to the incident downwelling irradiance, Ed, measured with the scalar irradiance sensor placed over a black light well at identical distance and placement in the light field as the surface of bioprinted constructs.
[0053]SEM images were taken with a Zeiss Sigma 500 scanning electron microscope. Samples were prepared in two different ways. To image the bionic coral skeleton made of PEGDA, samples were dried at room temperature and sputter coated with iridium (Emitech K575X Sputter Coater). To image the bionic coral tissue made of GelMA, samples were snap frozen with liquid nitrogen, and were then lyophilized in a freeze dryer (Freezone, Labonco) for 3 days. The overall shape could not be maintained, but microscale structures (such as micropores of GelMA) were well preserved. The samples were sputter coated with iridium (Emitech K575X Sputter Coater) prior to imaging on the SEM.
[0054]To characterize microalgal aggregate size and distribution in 3D, a confocal laser scan microscope was used (Nikon Eclipse TE-2000U). Bionic corals were placed on a cover glass and imaged from below with a 641 nm laser. Confocal stacks of chlorophyll a fluorescence were acquired using a pinhole size of 1.2 μm, a vertical step size for z-stacking=1 μm, and a x,y resolution of 0.6 μm. Particle segmentation and visualization of the data was performed in ImageJ and the NIS confocal elements software (Nikon). Particle segmentation was performed via manual thresholding of 229-4095 gray scale values, with a cleaning factor of 6× (this eliminates smaller particles that are not aggregates), hole filling and a smoothing factor of 2×. The segmented particles were analyzed for surface area, volume and particle density per volume.
[0055]Bionic corals enable microalgal cultivation with strongly reduced energy maintenance requirements, as they do not require water mixing and the fluence rate within the biomaterial is up to two-fold enhanced relative to the incident light source (FIGS. 4B, 4C). By comparing our developed bionic corals to various microalgal cultivation platforms, we conclude that no other cultivation system combines excellent light use efficiency with high productivity, low maintenance requirements, excellent water usage, and the ability for spatially controlled cell growth (FIGS. 12A-12C). While traditional systems use glass, excess water, and costly aeration, bionic corals consist of a hydrogel system with minimal water usage, optimized light propagation, and gas flux.
[0056]The table provided in FIG. 13 summarizes performance results comparing the inventive bionic coral against a variety of cultivation platforms as reported in the literature, including open pond, stirred flask, flat panel, vertical column, horizontal tubular, biofilm twin layer, polystyrene foam, rotating algal disk and algal turf scrubber. We compared liquid cultivation platforms and state of the art biofilm-based systems. Light use efficiency was calculated as gram algal biomass (ash free dry weight from the maximum specific growth rate or the maximum reported harvested biomass) per incident irradiance in mol photons of photosynthetically active radiation (400-700 nm)m−2 d−1 (see10). Areal density describes the highest reported densities of algal biomass (AFDW) per m−2 of exposed planar surface area. Water usage describes the amount of water needed accounting for both cultivation medium and dewatering processes. Maintenance accounts for the combined labor needed to operate the system. The bionic coral data stems from two different experimental set-ups. Enhanced productivity and light use efficiency were achieved using additional aeration and an incident light intensity of 150 μmol photons m−2 s−1.
[0057]The high spatial efficiency of the bionic coral system is thus particularly suitable for the design of compact photobioreactors for algal growth in dense urban areas, or as life support systems for space travel. Moreover, bionic corals allow investigation of the cellular activity of specific Symbiodinium strains, while mimicking the optical and mechanical microenvironment of different coral species, thus providing an important tool for advancing animal-algal symbiosis and coral bleaching research. Bionic corals have applications from biological studies to commercial technologies for efficient photon augmentation for sustainable bioenergy and bioproduct generation.
REFERENCES
[0058]1. Hatcher, B. G. Coral reef primary productivity: a beggar's banquet. Trends. Ecol. Evol. 3, 106-111 (1988).[0059]2. Brodersen, K. E., Lichtenberg, M., Ralph, P. J., Kuhl, M., Wangpraseurt, D. Radiative energy budget reveals high photosynthetic efficiency in symbiont-bearing corals. J. R. Soc. Interface 11, 20130997 (2014).[0060]3. Roth, M. S. The engine of the reef: photobiology of the coral—algal symbiosis. Front. Microb. 5, 422 (2014).[0061]4. Wangpraseurt, D., Jacques, S. L., Petrie, T., Kuhl, M. Monte Carlo modeling of photon propagation reveals highly scattering coral tissue. Front. Plant Sci. 7, 1404 (2016).[0062]5. Enriquez, S., Mendez, E. R., Hoegh-Guldberg, 0. Iglesias-Prieto, R. Key functional role of the optical properties of coral skeletons in coral ecology and evolution. Proc. R. Soc. B. 284, 20161667 (2017).[0063]6. Wijffels, R. H., Barbosa, M. J. An outlook on microalgal biofuels. Science 329, 796-799 (2010).[0064]7. Ooms, M. D., Dinh, C. T., Sargent, E. H., Sinton, D. Photon management for augmented photosynthesis. Nat. Commun. 7, 12699 (2016).[0065]8. Sanchez, C., Arribart, H., Guille, M. M. G. Biomimetism and bioinspiration as tools for the design of innovative materials and systems. Nat. Mater. 4, 277 (2005).[0066]9. Lode, A., et al., Green bioprinting: Fabrication of photosynthetic algae-laden hydrogel scaffolds for biotechnological and medical applications. Eng. Life Sci. 15, 177-183 (2015).[0067]10. van Oppen, M. J., Oliver, J. K., Putnam, H. M., Gates, R. D., Building coral reef resilience through assisted evolution. Proc. Natl. Acad. Sci. U.S.A. 112, 2307-2313 (2015).[0068]11. Wangpraseurt, D., et al., In vivo microscale measurements of light and photosynthesis during coral bleaching: evidence for the optical feedback loop? Front. Microb. 8, 59 (2017).[0069]12. Marcelino, L. A., et al., Modulation of light-enhancement to symbiotic algae by light scattering in corals and evolutionary trends in bleaching. PLoS ONE 8, e61492 (2013).[0070]13. Kuhl, M., Cohen, Y., Dalsgaard, T., Jorgensen, B. B., Revsbech, N. P. Microenvironment and photosynthe