Cellulose-based hydrogels as vaccine adjuvants

Abstract

Adjuvant compositions, and vaccine compositions utilizing the adjuvant compositions, are described. The adjuvant compositions include a salt-crosslinked TEMPO-oxidized cellulose nanofibril (CNF) hydrogel, and the vaccine compositions further include an antigen or immunogen.

Description

TITLE Cellulose-Based Hydrogels as Vaccine Adjuvants RELATED APPLICATIONS

[0001] This application claims priority to United States Provisional Application No.63 / 622,892 filed under 35 U.S.C. § 111(b) on January 19, 2024, the disclosure of which is incorporated herein by reference in its entirety. STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH

[0002] This invention was made with no government support. The government has no rights in this invention. BACKGROUND

[0003] Billions of dollars are spent every year in the global aquaculture market, and finfish are responsible for about two-thirds of that. Antibiotics have commonly been used for disease mitigation in finfish aquaculture, but have multiple drawbacks including resistance and leaching. Aquaculture loss due to disease is substantial. Solutions other than antibiotics have challenges. For example, vaccines using oil- based adjuvants can cause internal adhesions in the salmon peritoneal cavity. Salmon in particular may need to be vaccinated multiple times due to environmental factors and the condition of the fish.

[0004] Disease outbreaks are a major impediment to aquaculture production and are forecast to continue as the industry grows and the climate warms. Vaccines are integral for disease management in aquaculture but they can be expensive, vary in effectiveness, and come with adjuvant-induced adverse effects causing fish welfare issues and negative economic impacts. Thus, there remains a need in the art for compositions and methods useful for protecting fish from disease. SUMMARY

[0005] Provided is an adjuvant composition comprising a cellulose nanofibril (CNF) hydrogel crosslinked with a cation from a salt, wherein the adjuvant is an injectable solution capable of passing through a 26-gauge needle. In certain embodiments, the CNF hydrogel comprises TEMPO-oxidized cellulose nanofibrils. In certain embodiments, the cation comprises Na+or Ca2+. In certain embodiments, the salt is NaCl or CaCl2.

[0006] Further provided is a method of preparing a vaccine composition, the method comprising mixing a CNF hydrogel with an antigen and optionally one or more pharmaceutically acceptable diluents,carriers, or adjuvants, wherein the CNF hydrogel is a salt crosslinked TEMPO-oxidized CNF. In certain embodiments, the salt crosslinked TEMPO-oxidized CNF is crosslinked with Na+ions or Ca2+ions from the salt. In certain embodiments, the salt is NaCl or CaCl2. In particular embodiments, the NaCl is at a concentration of up to about 114 mM. In particular embodiments, the CaCl2is up to a concentration of about 7.14 mM. In certain embodiments, the CNF hydrogel is present at a concentration of up to about 3 wt%.

[0007] Further provided is a method of vaccinating an animal, the method comprising administering to an animal a vaccine composition comprising an antigen and an adjuvant, wherein the adjuvant comprise a salt crosslinked TEMPO-oxidized CNF hydrogel. In certain embodiments, the salt crosslinked TEMPO- oxidized CNF hydrogel is crosslinked with Na+or Ca2+ions. In certain embodiments, the animal is a fish. In certain embodiments, the animal is a salmon. In certain embodiments, the vaccine composition is an injectable solution capable of passing through a 26-gauge needle. In certain embodiments, the vaccine composition further comprises one or more additional adjuvants, stabilizers, preservatives, surfactants, buffering agents, or culturing substances.

[0008] Further provided is a vaccine composition comprising a salt crosslinked cellulose nanofibril (CNF) hydrogel; and an antigen or immunogen. In certain embodiments, the vaccine composition is an injectable solution capable of passing through a 26-gauge needle. In certain embodiments, the CNF hydrogel comprises TEMPO-oxidized CNF crosslinked with the salt. In certain embodiments, the salt is NaCl or CaCl2. In certain embodiments, the vaccine composition further comprises one or more additional adjuvants, stabilizers, preservatives, surfactants, buffering agents, or culturing substances. In certain embodiments, the vaccine composition comprises bacterin.

[0009] Further provided is a method of stimulating a foreign body response (FBR) in an Atlantic salmon, the method comprising administering to an Atlantic salmon a vaccine composition comprising a citric acid TEMPO-oxidized CNF hydrogel and an immunogen.

[0010] Further provided is a kit for making a vaccine, the kit comprising a first container housing a salt-crosslinked TEMPO-oxidized CNF hydrogel; and a second container housing an antigen or immunogen. In certain embodiments, the second container houses bacterin.

[0011] Further provided is the use of a salt-crosslinked TEMPO-oxidized cellulose nanofibril (CNF) hydrogel as a vaccine adjuvant.

[0012] Further provided is a vaccine composition comprising an amidated TEMPO-oxidized CNF hydrogel; and an immunogen or antigen; wherein the vaccine composition is an injectable solution. In certain embodiments, the vaccine composition has shear-thinning behavior.BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

[0014] FIG.1: Illustration of a CNF hydrogel.

[0015] FIGS.2A-2B: Crosslinked hydrogel rheometry of NaCl (FIG.2A) and CaCl2(FIG.2B) hydrogels.

[0016] FIGS.3A-3B: Dye diffusion out of crosslinked NaCl (FIG.2A) and CaCl2 (FIG.2B) hydrogels.

[0017] FIGS.4A-4C: Photographs showing salmon smolt receiving a fluorescent tag (FIG.4A) and a vaccination (FIG.4B), and abdominal adhesions results at 300 degree days (FIG.4C).

[0018] FIG.5: Log-log plot of mean viscosity (^) in relation to mean shear rate (^) for shear thinning TOCNF formulations exhibiting power-law behavior (n = 3). Error bars are standard deviations from the mean.

[0019] FIGS.6A-6H: SEM photomicrographs of sonicated 2x EDC NHS amidated TOCNF at 250x magnification (FIG.6A), sonicated 2xEDC NHS amidated TOCNF at 1500x magnification (FIG.6B), 1x amidated TOCNF at 250x magnification (FIG.6C), 1x amidated TOCNF at 1500x magnification (FIG. 6D), sonicated 1x amidated TOCNF at 250x magnification (FIG.6E), sonicated 1x amidated TOCNF at 1500x magnification (FIG.6F), sonicated 2x ODA DMF amidated TOCNF at 250x magnification (FIG. 6G), and sonicated 2x ODA DMF amidated TOCNF at 1500x magnification (FIG.6H). Scale bars represent 40 µm under 250x magnification and 4 µm under 1500x magnification.

[0020] FIG.8: FT-IR interferogram documenting changes in absorbance across sonicated variations of amidated TOCNF of the entire absorbance spectrum for each sample (left) and a zoomed in area of the absorbance spectrum between wavelengths 3300 to 1200cm-1(right).

[0021] FIG.9: Kaplan-Maier curves displaying cumulative incidence of time to death (%) for Atlantic salmon vaccinated with amidated TOCNF treatments compared to negative sham controls and positive commercial oil-adjuvanted control during in vivo toxicity study. No mortalities occurred in the sentinel fish during the course of the study. Each vertical step in the curve indicates one or more events (deaths) from time of vaccination (study day). The Gehan-Breslow-Wilcoxon test for comparing cumulative incidence of mortalities (%) across all treatment groups during the course of the study indicates significant differences between the survival curves (p < 0.0001) (n = 60).

[0022] FIG.10: Mean Fulton’s condition factor (K) of Atlantic salmon between formulations over the course of the trial. No significance between formulations (p = 0.0.3406) nor time (p = 0.8028). Dotted red line denotes K = 1.0. Error bars represent the standard error of mean (n = 30).

[0023] FIG.11: Mean specific growth rate (%) of replicate tanks of Atlantic salmon vaccinated by treatment group at 300- and 600- degree-days post-injection. Mixed effects model shows significant differences between time point (p = 0.0006) and treatment (p = 0.0012). Error bars represent standard error of the mean (n = 30).

[0024] FIGS.12A-12B: Hemorrhaging of ventral surface surrounding injection site as seen in Atlantic salmon mortalities injected with vaccine formulated with 2x EDC NHS amidated TOCNF (FIG. 12A) and 2x ODA DMF amidated TOCNF (FIG.12B).

[0025] FIG.13: Internal gross pathology observed in mortalities of Atlantic salmon vaccinated with 2x amidated TOCNF formulations including a.) pale liver, b.) external hemorrhaging surrounding the injection site, and c.) severe hemorrhaging near the formulation through the epidermal layer into the viscera with involvement of pyloric caeca.

[0026] FIG.14: Mean Speilberg scoring of abdominal adhesions (upper left), visceral melanization (upper right), peritoneal melanization (*p < 0.05) (lower left), and abdominal residue (*p < 0.05) (lower right) in Atlantic salmon injected with amidated TOCNF formulations compared to positive and negative controls by ordinary two-way ANOVA at 300- and 600- degree days post-implantation. Error bars are standard error of the mean (n = 30).

[0027] FIG.15: Internal gross pathologies observed at the 300-degree day sampling time point in Atlantic salmon vaccinated with 2x amidated TOCNF formulations including a.) pale liver and b.) edema of the pyloric caeca.

[0028] FIG.16: FT-IR interferogram documenting changes in absorbance throughout the wash process during production of 1x amidated TOCNF of the entire absorbance spectrum for each sample (left) and a zoomed-in area of the absorbance spectrum between wavelengths of 3200 - 300 cm-1(right).

[0029] FIGS.17A-17B: Device to intraperitoneally implant TOCNF hydrogels into Atlantic salmon demonstrating PIT tagging implanter connected to 8-gauge stainless steel luer-lok needle with 3D printed plastic blunt stopper (black) attached to the plunger to deliver a TOCNF hydrogel (FIG.17A), and a close- up of the 8-gauge luer-lok needle pre-loaded with TOCNF hydrogel (FIG.17B).

[0030] FIGS.18A-18B: Observed closed visible incision (FIG.18A) and lesion developed anteriorly from the closed incision site 600- degree days after implantation of TOCNF hydrogels with the improved implantation device (FIG.18B).

[0031] FIG.19: Mean Fulton’s condition factor of Atlantic salmon between formulations over the course of the trial described in Example III herein. No significance between formulations (p = 0.7058) nor time (p = 0.8856). Error bars represent the standard error of mean (n = 8).

[0032] FIG.20: Photographic example of external pathologies observed, namely, a.) unhealed incision, b.) visible lesion, and c.) external proliferative mass during gross necropsy in Atlantic salmon at600-degree days post-implantation with TOCNF hydrogel.

[0033] FIG.21: Mean prevalence (%) of external gross pathology for a.) healed incisions (*p < 0.05), b.) external lesions (letters indicate significant differences between treatments, p < 0.05 with ab indicating no significant differences between a nor b, p > 0.05), and c.) external proliferative masses by treatment group at 300- degree days and 600- degree days post-implantation. Error bars are standard errors of the mean (n = 8).

[0034] FIG.22: Mean Speilberg scoring of a.) abdominal adhesions in Atlantic salmon implanted with TOCNF formulations compared to PBS + Bacterin (*p < 0.0001), and b.) visceral melanization in Atlantic salmon implanted with TOCNF formulations compared to PBS + Bacterin (p = 0.1486) by ordinary two-way ANOVA at 300- and 600- degree days post-implantation. Error bars are standard error of the mean (n = 8).

[0035] FIG.23: Distribution of Spielberg scores (%) for internal abdominal adhesions seen in Atlantic salmon peritoneum at a.) 300- degree days (n = 8), and b.) 600- degree days post-implantation with TOCNF hydrogels (n = 8).

[0036] FIGS.24A-24G: Representative photomicrographs from the body wall of Atlantic salmon at 600- degree days in response to intraperitoneal implantation of TOCNF hydrogels compared to PBS + Bacterin demonstrating normal structure (score 0) (FIG.24A), minimal adhesions and fibrosis (score 1) (FIG.24B), mild adhesions and fibrosis (score 2) (FIG.24C), moderate adhesions (score 3) (FIG.24D), moderate focal fibrosis + / - small granulomas (score 4) (FIG.24E), marked, focally extensive fibrosis with granulomas (score 5) (FIG.24F), and marked transmural inflammation (score 6) (FIG.24G). Scale bars represent 500 µm.

[0037] FIGS.25A-25F: Representative photomicrographs from the coelom of Atlantic salmon at 600- degree days demonstrating normal viscera (score 0) (FIG.25A), minimal adhesions (score 1) (FIG. 25B), mild focal or multifocal adhesions between viscera (score 2) (FIG.25C), moderate multifocal adhesions between viscera, nodular fibrosis (score 3) (FIG.25D), moderate to marked, focal inflammation and nodular fibrosis (score 4) (FIG.25E), marked, multifocal to diffuse inflammation and adhesions (score 5) (FIG.25F), and severe inflammation and diffuse adhesions (score 6) (FIG.25G). Scale bars represent 500 µm.

[0038] FIG.26: Mean pathology scores from histology as a.) replicate tanks of the body wall of Atlantic salmon at 600- degree days post-implantation of TOCNF hydrogels. Significant differences in mean pathology scores in the body wall are observed between fish vaccinated with TOCNF hydrogels compared with the PBS + Bacterin control group (p < 0.0001). No significant differences were observed between replicate tanks (p = 0.1235). Error bars represent the standard error of the mean (n = 4). The chart in b.) shows replicate tanks of the coelom of Atlantic salmon at 600- degree days post-implantation ofTOCNF hydrogels. Significant differences in mean pathology scores in the coelom are observed between fish vaccinated with TOCNF hydrogels compared with the PBS + Bacterin control group (p = 0.0030). No significant differences were observed between replicate tanks (p = 0.9695). Error bars represent standard error of the mean (n = 4).

[0039] FIG.27: Histopathology distribution of scores (%) of a.) the body wall by treatment (n = 8) and b.) pathology distribution of scores (%) of the coelom by treatment (n = 8).

[0040] FIG.28: Heat map of acute mean gross pathology observed in Atlantic salmon parr within the first 72 hours post-implantation comparing a TOCNF hydrogel vaccine to a DPBS + Bacterin only group (n = 4).

[0041] FIG.29: Illustration of citric acid crosslinked TEMPO CNF. The intermolecular interactions are displayed in red.

[0042] FIG.30: Rate of change in supernatant pH of 1.1, 1.3, 1.5, and 1.7 wt% hydrogels related to the average pH of the supernatant.

[0043] FIG.31: Washing of 1.7 wt% hydrogel in PBS after initial DI water washing occurred.

[0044] FIG.32: Rate of drying in mg / min of 20, 29.5, 46, and 63 °C dried hydrogels in comparison to the overall moisture content of the drying hydrogels.

[0045] FIG.33: Dehydration percentage of hydrogels dried at 20, 29.5, 46, and 63 °C.

[0046] FIG.34: Compression strength and modulus of different stages of hydration.

[0047] FIG.35: Rehydration of 10, 30, 50, 70, and 90 minute dried hydrogels at 46 °C in relation to the change in % of initial mass over a two hour period.

[0048] FIGS.36A-36B: Breakdown of hydrogel over 8-week time frame. FIG.36A shows a hydrogel imaged the first day after being submerged in PBS. FIG.36B shows the hydrogel imaged after 8 weeks of submersion in PBS.

[0049] FIGS.37A-37C: Effects of vaccine formulation on mean gene expression from fish head kidney at 300-degree days and 600-degree days post-vaccination. The 2-ΔΔCtmethod was used to compare the expression of IgM (* denotes treatments where p < 0.05 compared to DPBS only group) (FIG.37A), IgT (FIG.37B), and IgD relative to that of β-actin (FIG.37C). Sentinel and DPBS + bacterin groups were not included in statistical analysis. Error bars are standard error of the mean (n = 9). DETAILED DESCRIPTION

[0050] Throughout this disclosure, various publications, patents, and published patent specifications are referenced by an identifying citation. The disclosures of these publications, patents, and published patent specifications are hereby incorporated by reference into the present disclosure in their entirety to more fully describe the state of the art to which this invention pertains.

[0051] In accordance with the present disclosure, cellulose-based hydrogels can be used as adjuvants and drug delivery systems in vaccines. This is particularly useful in the context of vaccines for finfish such as salmon. The cellulose-based hydrogels can reduce adverse reactions and the number of vaccinations needed while maintaining an effective immune response until the salmon is harvested. Thus, further provided herein are vaccine compositions that include cellulose-based hydrogel adjuvants.

[0052] Hydrogels are three-dimensional networks of hydrophilic polymers that can absorb and retain a significant amount of water. Cellulose nanofibril (CNF) hydrogels are materials derived from cellulose, which is a natural biopolymer found in the cell walls of plants and the most abundant organic material on Earth. Cellulose is composed of β-1,4-anhydroglucopyranoside repeated units linked together through covalently bonded oxygen between the C1 of the initial glucose ring and the C4 of the adjoining ring. Cellulose nanomaterials such as cellulose nanofibrils can be isolated from several sources such as wood, cotton, hemp, linen, algal cellulose, bacterial cellulose, and so on.

[0053] CNFs are nanoscale fibers that can be extracted from various cellulose-rich sources, such as wood, cotton, or other plant materials. CNFs have a high specific strength and modulus, low density, hydrophilicity, biodegradability, and the ability for surface modification. CNFs can be obtained through mechanical or chemical treatments applied to cellulose-rich sources. CNFs can be extracted from cellulose pulp through mechanical shearing or through a combination of both mechanical and chemical methods such as high-pressure homogenization, grinding, cryocrushing, and high intensity ultrasonic treatment. Mechanical methods may involve grinding or homogenizing the cellulose material to break it down into nanoscale fibrils, while chemical methods use specific treatments to isolate the nanofibrils. The more shearing that occurs, the more cellulose fibers are cleaved transversely to produce finer sized fibers in the nano- to micron scale. Chemical treatments used to obtain CNF may include acid hydrolysis, oxidative treatments, enzymatic hydrolysis, or alkaline treatments. As one non-limiting example, an oxidative treatment may involve the use of oxidizing agents, such as sodium hypochlorite (NaClO) or 2,2,6,6- tetramethylpiperidine-1-oxyl (TEMPO), optionally in combination with a mechanical treatment. In such an oxidative treatment, the oxidizing agents modify the surface of the cellulose fibers, making them more susceptible to mechanical disintegration into nanofibrils. Oxidative treatments can yield CNFs with improved surface functionalities. Additional processing such as a kraft or sulphite process may be used to facilitate further cleavage and remove undesirable matrix material so as to produce pure cellulose. Cellulose nanofibrils can be used to form hydrogels through various methods such as, but not limited to, physical cross-linking, chemical cross-linking, or a combination of both. As depicted in FIG.1, CNF hydrogels have a closer fiber aggregation compared to CNF due to the addition of salts.

[0054] When CNF is obtained by a TEMPO process, the CNF has distinctive characteristics compared to CNFs obtained through other methods. The TEMPO process involves the use of a radicalinitiator (TEMPO) and NaClO to selectively oxidize the primary hydroxyl groups of cellulose. This results in CNF having surface functionalization (due to the introduction of carboxylate groups on the CNF surface, which imparts negative charges to the CNFs, preventing their aggregation), colloidal stability (from the carboxylate groups, which cause electrostatic repulsion), biocompatibility, and improved purity. In accordance with the present disclosure, TEMPO-oxidized CNF (also referred to as TEMPO CNF) is particularly useful for the creation of CNF hydrogels to be used as an adjuvant in a vaccine composition.

[0055] TEMPO-oxidized cellulose nanofibril (CNF) hydrogels have been used in various veterinary and biomedical applications due to their bioinert nature. Salt bridging enhances the mechanical properties of TEMPO including stability in aqueous solutions. Salts shield the negative charge of TEMPO fibers allowing for closer aggregation of fibers. Closer fiber aggregation and salt bridging results in improved gel strength. However, it is understood that CNFs obtained from non-TEMPO process may also, or alternatively, be used to form the salt-crosslinked CNF hydrogels described herein, and used as adjuvants in vaccine compositions.

[0056] CNF hydrogels can absorb and retain large amounts of water. This is advantageous for drug delivery as it can help in solubilizing and delivering hydrophilic drugs effectively. CNF hydrogels may be transparent, making them suitable for certain optical or medical applications. CNF hydrogels are ideal for drug delivery because of their tunability and ability to encapsulate both water soluble and insoluble molecules for the sustained release of the desired drug. With this sustained release, the immune system can be triggered and create a desired immune response for a longer period of time compared to conventional drug delivery systems.

[0057] CNF hydrogels are also useful for drug delivery because of their biocompatibility and ability to mimic the extracellular matrix. CNF hydrogels are biocompatible and can be well-tolerated by the body, which is important for adjuvant applications as it ensures that the adjuvant does not induce adverse reactions or toxicity. Furthermore, CNF hydrogels may be able to alter their physical properties such as phase transition, swelling, or degree of crosslinking depending upon the stimulus. CNF hydrogels can be responsive to both chemical and physical responses.

[0058] CNF hydrogels can provide sustained release of drugs over time. The three-dimensional network structure of the hydrogel (FIG.1), formed by the cellulose nanofibrils, allows for controlled diffusion of drugs from the gel matrix. The examples herein demonstrate the controlled diffusion of a dye from a CNF hydrogel (FIGS.3A-3B). CNF hydrogels can encapsulate various types of drugs, including both hydrophilic and hydrophobic compounds. This versatility makes CNF hydrogels suitable for a wid range of pharmaceutical applications.

[0059] It may be difficult to crosslink pure CNF to form a hydrogel. However, modified CNF, such as TEMPO-oxidized CNF, is easier for forming a hydrogel. Suitable crosslinkers for forming a hydrogelwith TEMPO-oxidized CNF include, but are not limited to, polyethyleneimine, heavy metals (such as Fe3+and Ca2+), citric acid, dialdehydes, acetals, polycarboxylic acids, and epichlorohydrin / polyepichlorohydrin. A salt-crosslinked TEMPO-oxidized CNF hydrogel may be prepared by mixing a salt with TEMPO- oxidized CNF to crosslink the TEMPO-oxidized CNF. The salt may be, for example, NaCl or CaCl2. It is understood that while the term “salt-crosslinked” is used herein, the actual species accomplishing the crosslinking may be a cation from the salt, such as Na+or Ca2+. The terms “salt-crosslinked” and “salt crosslinked” are used herein to encompass cation-crosslinked embodiments where the cation is provided from a salt. In particular, the cation of a salt may crosslink carboxyl groups on different nanofibrils of the CNF, as illustrated in FIG.1 using Ca2+cations as an example. However, other salts or cross linkers are possible and encompassed within the scope of the present disclosure. For instance, in other embodiments, the TEMPO-oxidized CNF is crosslinked with citric acid to form a CNF hydrogel.

[0060] Advantageously, it has been found that in some embodiments, salt-crosslinked TEMPO- oxidized hydrogels exhibit shear-thinning behavior and are injectable solutions capable of passing through a 26-gauge needle. Shear-thinning hydrogels have improved injectability, moldability, and self-healing characteristics. Shear thinning is attributed to a hydrogel’s ability to rearrange and realign its polymer chains and chemical network when under various stimuli. This rearrangement allows for a local and sustained drug delivery hydrogel that can deliver a higher concentration for a longer period of time at a desired site while minimizing side effects commonly associated with other delivery methods. Thus, salt- crosslinked TEMPO-oxidized CNF hydrogels are useful as vaccine adjuvants. The formulation of a CNF- based hydrogel as an injectable solution should avoid eliciting the foreign body response in an animal injected with a vaccine composition that includes the CNF-based adjuvant.

[0061] In order to achieve the desired injectability and shear thinning behavior, the concentrations of the salt used to crosslink the TEMPO-oxidized CNF should be kept within certain limits. In particular, the salt may be used in a concentration of up to about 7.14 mM when the salt is CaCl2, or up to a concentration of about 114 mM when the salt is NaCl. Notably, these concentrations of salt are significantly less than the concentrations of cation crosslinkers previously used to physically crosslink CNF. In some non-limiting examples, CaCl2 may be used in concentrations of about 7.14 mM, 4.66 mM, or 1.8 mM. In some non- limiting examples, NaCl may be used in concentrations of 114 mM, 69 mM, or 23 mM. Furthermore, when the TEMPO-oxidized CNF concentration exceeds 3 wt%, then the resulting hydrogel may be difficult to push through a 26-gauge needle.

[0062] Furthermore, carboxylic acid groups present on TEMPO-oxidized CNF are easily conjugated through an amide bond. This amide bond occurs when an amine group (-NH2) reacts with a carboxylic acid group (-CONH-), causing the amine group to bond to the carboxylic acid. This process of amide bonding onto the carboxylic acid is referred to as amidation. Through this process, the overall characteristics of theTEMPO CNFs are altered without compromising the overall solubility of the polymer. The amidation of TEMPO CNF acts as a reinforcing structure and improves the dispersibility of cellulose. The amidation of TEMPO CNF may create a branched polymer system and an injectable shear thinning gel. Once the gel is injected and shear forces are removed from the system, the aligned amidated side chains may entangle to recreate the matrix in the peritoneal cavity, allowing for long-term diffusion. The process of amidating TEMPO CNF may utilize toxic residual chemicals that may cause adverse effects to aquatic animals, but with sufficient washing of the final product the gel is able to be injected and stay within fish safely with minimal reactions. In some embodiments, the CNF-based hydrogels described herein include amidated TEMPO CNF.

[0063] Rheometry is the study of the flow and deformation of materials, particularly liquids and soft solids, under the influence of applied forces or stresses. Rheometry measures the rheological properties of materials to understand their flow characteristics and mechanical responses. As shown in the examples herein, the salt-crosslinked TEMPO-oxidized CNF hydrogels with the above-described salt concentrations exhibit shear-thinning behavior (FIGS.2A-2B). This means that the viscosity decreases with an increase in shear rate. In other words, as the salt-crosslinked TEMPO-oxidized CNF hydrogels experience higher rates of shear or deformation, they become less resistant to flow. Certain biological fluids, such as synovial fluid in joints, mucus, and blood are also shear-thinning. This property is beneficial in biological systems because it facilitates easier flow and movement. This property is also beneficial in pharmaceutical preparations for ease of administration. Thus, the salt-crosslinked TEMPO-oxidized CNF hydrogels exhibit shear-thinning behavior which is advantageous in a vaccine composition to facilitate ease of administration of the vaccine composition. In some embodiments, the adjuvant compositions and vaccine compositions described herein are injectable solutions that can pass through a 26-gauge needle. This is highly advantageous because most aquaculture vaccination methods utilize a 26-gauge needle and syringe to administer the vaccination into aquatic animals.

[0064] CNF hydrogels can be combined with other materials or polymers to create hybrid systems with enhanced properties, such as improved drug loading capacity or controlled release profiles. For example, a CNF hydrogel may be loaded with bacterin to vaccinate an animal. Vaccine compositions may include a CNF hydrogel, an antigen or immunogen, and optionally one or more additional adjuvants, stabilizers, preservatives, surfactants, buffering agents, or culturing substances. The vaccine compositions may include an antigen or immunogen bound to CNF hydrogel or otherwise incorporated in the CNF hydrogel. The inherent mechanical strength of CNF contributes to the overall stability and integrity of the vaccine compositions during handling and administration. CNF is also particularly suitable for vaccine compositions because of its biocompatibility.

[0065] The adjuvant and vaccine compositions described herein may also be made available via a kitcontaining one or more key components. A non-limiting example of such a kit is a kit for making a vaccine composition which includes a salt-crosslinked TEMPO-oxidized CNF hydrogel and an antigen or immunogen (such as bacterin) in separate containers, where the containers may or may not be present in a combined configuration. Many other kits are possible and encompassed within the scope of the present disclosure, such as kits including a salt and a CNF hydrogel in separate containers, or kits further including syringes or other devices for administering a vaccine composition. The kits may further include instructions for using the components of the kit to prepare wet wipes. The instructions may be recorded on a suitable recording medium. For example, the instructions may be present in the kits as a package insert or in the labeling of the container of the kit or components thereof. In other embodiments, the instructions are present as an electronic storage data file present on a suitable computer readable storage medium, such as a flash drive. In other embodiments, the actual instructions are not present in the kit, but means for obtaining the instructions from a remote source, such as via the internet, are provided. An example of this embodiment is a kit that includes a web address where the instructions can be viewed and / or from which the instructions can be downloaded.

[0066] EXAMPLES

[0067] Example I – Salt crosslinked TEMPO-oxidized CNF hydrogels

[0068] This example demonstrates the efficacy of NaCl and CaCl2 as crosslinkers to form CNF hydrogels, and the usefulness of the resulting CNF hydrogels as vaccine adjuvants which do not cause a foreign body response in Atlantic salmon.

[0069] CNF hydrogels were prepared according to the following procedures.

[0070] Preparation of salt-crosslinked TEMPO CNF

[0071] Solutions of salt were prepared. 0.9 M for CaCl2 was prepared by dissolving 5 g CaCl2 in 50 ml of water. 5.82 M NaCl was prepared by dissolving 17 g of NaCl in 50 ml of water. The dissolution of each salt was confirmed to be complete before proceeding.

[0072] 25 g of 1.1 wt% TEMPO CNF was placed into a 50 ml falcon tube with a conical bottom. The TEMPO and salt solutions were placed in the refrigerator and allowed to thermally equilibrate for one hour. After being removed from the refrigerator, the desired amount of salt solution was quickly added to the solution of TEMPO. Table A below provides the details of the samples created for fish trials.

[0073] Table A – Salt-crosslinked TEMPO CNF hydrogel used for fish trials Formulation Volume of Volume of Mass of TEMPO CNF 0.9M CaCl25.82M NaCl (1.1wt%) [0074vortexing falcon tube mixer. The falcon tube was mixed for ten seconds on the highest setting, ensuring that the gel was observed forming a vortex inside of the tube. After ten seconds on the mixer, the tube was hand mixed by inverting quickly five times. The mixing step was repeated an additional two times, and then the gel was allowed to rest in the refrigerator for 24 hours without agitation before use.

[0075] Rheometry

[0076] The rheometry of the hydrogel adjuvant compositions was analyzed. FIGS.2A-2B show the crosslinked hydrogel rheometry. As seen in FIGS.2A-2B, the NaCl and CaCl2 hydrogels exhibited shear- thinning behavior.

[0077] The hydrogels better maintained their shape at higher salt concentrations. Even at high salt concentrations, all formulated hydrogels could easily pass through a 26-gauge syringe. After reaching a critical point, the hydrogels exhibited water rejection, clumping and fracturing.

[0078] Dye diffusion out of crosslinked hydrogels

[0079] In vitro characterizations of the hydrogels and hydrogel / antigen (vaccine) formulations were conducted by using fluorescent variants and analyzing diffusion of the fluorescent dye with fluorescence microscopy. FIGS.3A-3B show the diffusion of a dye out of the crosslinked hydrogels. As seen from FIGS.3A-3B, NaCl hydrogels had similar diffusion characteristics across all concentrations, and CaCl2 hydrogels exhibited similar diffusion characteristics.

[0080] Salmon safety trial

[0081] A safety trial was conducted using the salt crosslinked TEMPO-oxidized CNF hydrogels in Atlantic salmon. Salmon smolt were administered a fluorescent tag or a vaccination with CNF hydrogel, as shown in the photographs in FIGS.4A-4B. The 300-day results are shown in FIG.4C. The 300-day results showed no mortalities recorded across all formulations. Most of the hydrogels were recovered from the peritoneal cavity of the salmon. Lower average internal and external abnormalities were observed compared to previous gel formulations. Significant adverse adhesions occurred compared to the negativeand positive control group in the lowest calcium chloride group and the highest calcium chloride group. An edematous cyst-like bubble formed between the junction of the pyloric caeca and digestive tract in all of the 7.15mM CalCl2fish. Without wishing to be bound by theory, it is believed this may be due to an ion imbalance drawing water into the digestive tract. The fish did better with the NaCl formulations. The lowests concentration of the NaCl group (0.023 M) also had the lowest hydrogel retrieval rate (70% of the hydrogels were recovered, 30% of the hydrogels were not recovered). No melanization was observed in any of the formulations. This shows that the salt crosslinked TEMPO-oxidized CNF hydrogel vaccine compositions did not cause a foreign body response in the salmon.

[0082] In sum, the salt crosslinked TEMPO CNF hydrogels were more stable in aqueous solutions than other hydrogels, had greater gel stiffness than other hydrogels, and could impact diffusion rates from the hydrogel matrix. Salt crosslinked TEMPO hydrogels were found to be shear-thinning using cone and plate rheometry (40 mm 2° cone), were easily injectable through a 26-gauge needle, and did not cause a foreign body response in Atlantic salmon.

[0083] Example II – Toxicity and immunogenicity of intraperitoneally injected shear-thinning TEMPO-oxidized cellulose nanofiber hydrogels produced and characterized for antigen delivery in an Atlantic salmon (Salmo salar L.) vaccine

[0084] This example describes the preparation of an injectable shear-thinning vaccine using cellulose nanomaterials (CNM) as a crosslinked adjuvant matrix; the in vitro characterization of the mechanical, structural, and chemical properties of the CNM adjuvanted vaccine formulations using rheology, scanning electron microscopy, and Fourier-Transform Infrared spectroscopy; an in vivo examination of toxicity in Atlantic salmon compared to a commercial adjuvanted Montanide control; and the quantification of immunogenicity using gene expression and serological antigen specific antibody response by indirect ELISA assay compared to commercial adjuvanted Montanide control.

[0085] Materials and methods

[0086] Preparation of vaccine using CNM matrices as an adjuvant

[0087] Bacterin production

[0088] Inactivated bacterial whole-cell vaccines are currently the most cost-effective, safe, and commonly used method of vaccination in aquaculture. Thus, Vibrio anguillarum, the causative bacterial agent of vibriosis in Atlantic salmon, was used to produce a bacterin to investigate vaccine immunogenicity formulated with a novel CNM hydrogel depot in Atlantic salmon. V.anguillarum isolate VE-2021-0143 was obtained from University of Maine Aquatic Animal Health Laboratory (AAHL)’s frozen bacterial stock culture collection. The isolate was streaked onto Tryptic Soy Agar with 5% sheep blood and 1.5% NaCl (BA, Northeast Laboratory Services) and allowed to grow at 16 + 1 °C for 48 hours. A single colony forming unit (CFU) from the plate was selected to inoculate one 10 mL tube of Trypticase Soy Broth +1.5% NaCl (TSB, Becton Dickinson and Fisher Scientific) under aerobic conditions for 48 hours at 16 + 1 °C in a 180-rpm shaker. The 10 mL culture was used to inoculate 500 mLs TSB for culture by incubating at 16 °C while stirring on a magnetic stir plate for 36 hours to OD600nm= 0.784 determined using an Ultraspec10 Cell Density Meter (Biochrom). The cultures were gram stained for purity and serial dilutions from 10-1to 10-10were performed in 2.7 mLs DPBS. Next, 100 µL of the 10-4to 10-10dilutions were plated on BA in duplicate and incubated for 48 hours at 16+1 °C for viable plate counts to determine CFU / mL of each culture. The culture mean was determined to be 2.68 x 109CFU / mL. Cultures were inactivated with 0.02% formalin (1 mL 37% Formalin into 500 mLs culture, Fisher Scientific) and stirred for 48 hours at 22 °C. Quality control was performed by adding 1 mL of V. anguillarum bacterin into 10 mLs of sterile TSB to confirm inactivation. The inactivated bacterin was stored at 4 °C until use.

[0089] Amidated TEMPO-oxidized CNF production

[0090] Amidated TEMPO-oxidized CNF formulations (amidated TOCNF) were developed, optimized, and prepared to formulate a vaccine with the CNM functioning as an injectable shear-thinning adjuvant for Atlantic salmon. First, a 1.1% TEMPO CNF slurry with a fiber length of approximately 1 μm and a diameter of 20 nm (Process Development Center University of Maine) was diluted to 0.5% slurry in deionized (DI) water. The 0.5% TEMPO CNF was stirred at 500-500 rpm and equilibrated to 50 ℃ in a water bath. Next, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC, 95% Fisher Scientific), and N-Hydroxysuccinimide (NHS, 98% Fisher Scientific) were dissolved in 3.0 mL of DI water. The dissolved solution of EDC and NHS was added to the 0.5% TEMPO CNF and stirred until uniformly mixed. The pH was adjusted to 5.5 - 6.0 using 1 M sodium hydroxide (Fisher Scientific) and allowed to stir for 30 minutes. The octadecylamine (ODA, 90% Fisher Scientific) was manually ground with a mortar and pestle to a fine powder. In a separate beaker, the ODA and dimethylformamide (DMF, 99.8% Acros Organics) were combined. Additionally, sonication was performed (Fisherbrand™ Model 505 Sonic Dismembrator) with some formulations as noted at an amplitude of 75 for 2:30 minutes. This was used to compare homogeneity of formulations during in vitro characterization. This ODA-DMF solution was added to the TEMPO-oxidized CNF suspension. The pH was adjusted to 7.5 - 8.0 using 1 M HCl (Fisher Scientific) and 1 M NaOH (Fisher Scientific) and stirred continuously for 24 hours. Afterward, the mixture was centrifuged at 7,000 rpm for 5 minutes and supernatant was discarded. The TEMPO-oxidized CNF pellet was reconstituted in DI water. The centrifugation and wash step was performed a total of three times to remove residual chemicals. Finally, the TEMPO-oxidized CNF pellet was washed with ethanol pH adjusted to 3.0 with 0.1 N HCl to remove any residual unconverted carboxyl and traces of unbound ODA. Afterward, two additional centrifuge steps were performed with subsequent DI water washes. Fourier- Transform Infrared spectroscopy was performed on supernatants collected from the wash steps during production. Each sample was then placed into 3,500 Da snakeskin dialysis with a regenerated cellulosemembrane (Fisher Scientific) for 24 hours to remove any additional residual chemicals. After snakeskin dialysis, the samples were mixed at a ratio of 1:10 amidated TOCNF:Vibrio anguillarum antigen and stored at 4 °C until use.

[0091] Methodology for the amidation of TEMPO CNF

[0092] First, a 1.1% TEMPO CNF slurry with a fiber length of approximately 1 μm and a diameter of 20 nm (Process Development Center University of Maine) was diluted to 0.5% slurry in deionized (DI) water according to Table 1.

[0093] Table 1 – Initial dilution of TEMPO CNF for production of vaccine formulations Original 2x EDC NHS 2x ODA DMF TEMPO CNF 59.09 g 59.42 g 59.13 g Water 70.91 g 70.90 g 70.96 g

[0094] After dilution, the 0.5% TEMPO CNF was stirred at 500-500 rpm and equilibrated to 50 ℃ in a water bath. Next, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC, 95% Fisher Scientific) and N-Hydroxysuccinimide (NHS, 98% Fisher Scientific) were dissolved in 3.0 mL of DI water according to Table 2.

[0095] Table 2 - Dissolving EDC and NHS in 3.0 mL of DI water Original 2x EDC NHS 2x ODA DMF EDC 1.0088 g 2.02 g 1.00869 g NHS 0.7267 g 1.45638 g 0.72680 g

[0096] The dissolved solution of EDC and NHS was then added to the 0.5% TEMPO CNF and stirred until uniformly mixed. The pH was adjusted to 5.5-6.0 using 1 M sodium hydroxide (Fisher Scientific) and allowed to stir for 30 minutes. In a separate beaker, the octadecylamine (ODA, 90% Fisher Scientific) and dimethylformamide (DMF, 99.8% Acros Organics) were combined according to Table 3.

[0097] Table 3 Original 2x EDC NHS 2x ODA DMF ODA 1.131 g 1.157 g 2.26383 g DMF 32.5 mL 33 mL 65 mL

[0098] This ODA-DMF solution was added to the TEMPO CNF suspension and the pH was adjusted to 7.5-8.0 using 1 N HCl and 0.5 N NaOH and stirred continuously for 24 hours. Afterward, this mixture was centrifuged at 7,000 rpm for 5 minutes and supernatant was discarded. The TEMPO CNF pellet was reconstituted in DI water. This centrifugation and wash step was performed a total of three times to remove residual chemicals. Finally the TEMPO CNF pellet was washed with ethanol pH adjusted to 3.0 with 0.1 N HCl to remove any residual unconverted carboxyl and traces of unbound ODA according to Table 4.

[0099] Table 4 – HCl and ethanol wash Original 2x EDC NHS 2x ODA DMF HCl 400 uL 400 uL 240 uL Ethanol 30 mL 30 mL 30 mL

[0100] After this, two additional centrifuge steps were performed with subsequent DI water washes. Each sample was then placed into snakeskin dialysis for 24 hours to remove any additional residual chemicals. After snakeskin dialysis, the samples were mixed at a ratio of 1:10 amidated TEMPO CNF:Vibrio anguillarum antigen. FT-IR analysis was performed on wash supernatants during production of the amidated TEMPO-oxidized vaccine formulations. Results are included in FIG.16.

[0101] Examining in vitro characteristics of amidated TOCNF hydrogel formulations

[0102] Mechanical properties: rheology

[0103] Rheological characterization of changes in viscosity of the formulations under force was performed. A plate and plate rheometer (DHR-3, TA Instruments) with a 4 mm radius and 8 mm sample stage was used to model the material as it flowed through a tube similar to that of a syringe by determining the flow and deformation (mechanical) characteristics of the amidated TEMPO CNF formulations at room temperature (25 °C). The formulations examined were unmodified TOCNF, 1x amidated TOCNF with and without sonication, 2x ODA DMF amidated TOCNF with and without sonication, 2x EDC NHS amidatedTOCNF with and without sonication, and 2x amidated TOCNF with and without sonication. Each formulation was tested in triplicate. The lower plate remained fixed. A 0.5 mL aliquot of material was added. The height was lowered to 1050 μm. Excess material was removed to reduce additional drag forces and subsequent error in measurements. The height was then reduced to 1000 μm before rotational torque was applied. The upper plate continuously rotated with an applied force or shear stress. As the rotational speed (shear rate, ^, sec-1) increased in a step manner, the longer that shear rate was applied to equilibrate the material. Log-log plots were produced to describe the flow behaviors of the formulations under shear force similar to that occurring during injection.

[0104] Structural Properties: Scanning Electron Microscopy

[0105] Scanning electron microscopy (SEM) allows for visualization and analysis of micro- and nanoparticle surface and structural properties. SEM (Zeiss NVision 40 FIB / SEM) was performed at the University of Maine Electron Microscopy Laboratory (Orono, Maine) to visualize the amidated TOCNF formulations. Sonication was used during production of the 1x amidated TOCNF hydrogels formulation to examine homogeneity compared to 1x amidated TOCNF without sonication such that four formulations were examined for comparison by SEM: (1) sonicated 2x EDC NHS amidated TOCNF, (2) 1x amidated TOCNF, (3) sonicated 1x amidated TOCNF, and (4) sonicated 2x ODA DMF amidated TOCNF. Samples were flash frozen in liquid nitrogen and then freeze dried for 30 hours to reduce the presence of ice templating within the formulations. The plate temperature was cycled -34.4, -6.7, 4.4, 15.6, and 32.2 °C for 8, 10, 8, 3, and 3 hours, respectively, with a constant vacuum set to 0.1 Torr. After drying, the formulations were resubmerged in liquid nitrogen and fractured to create a clear cross section. The formulations were attached to SEM stubs using conductive epoxy for imaging. The formulations were metal coated by vacuum evaporating 6 mm of gold / palladium using a Denton DV-502 Rotary Evaporator and then sputter coated with an additional 2 mm of gold / palladium to ensure adequate metal coating. Each sample was imaged at 25x, 100x, 250x, and 1500x magnification. In addition to standard SEM, InLens SEM imaging was used to improve the surface contrast and resolution within the imaged hydrogels for material differentiation of the sample.

[0106] Chemical properties: fourier-transform infrared spectroscopy

[0107] Fourier-Transform Infrared spectroscopy (FT-IR) is a form of spectroscopy useful for investigating chemical structures and bond interactions. FT-IR (ATR accessory-equipped Nicolet iS20) was used to analyze the chemical composition of the final various amidated TOCNF formulations compared to unmodified TOCNF. Additionally, samples were collected for FT-IR throughout production directly after the 24 hour amidation, the first wash, and the third wash steps. This was performed to evaluate the ability of the wash steps to remove residual or trace chemicals and to quantify the level of polymer cross-linking in the 1x amidated TOCNF formulation. The spectrums were recorded on a diamond plate with a resolution of4 cm-1with 128 scans. The Attenuated Total Reflection (ATR) auto-correction parameter was enabled for higher precision. The wavelength of the spectrum was set from 4000 to 400 cm-1. Prior to sample acquisition, the background was set to the same parameters as sample acquisition. Liquid and amidated TOCNF samples (200 μL) were loaded onto the diamond plate and covered to limit evaporation during sampling. Solid samples fully covered the diamond plate in powdered form before the ATR compression arm was engaged. The absorbance spectrums were separated for comparison by graphing manually with offset Y-axis values.

[0108] Examining in vivo toxicity in Atlantic salmon

[0109] In vivo study

[0110] To assess toxicity and immunogenicity of the amidated TOCNF hydrogels formulated as a potential injectable shear-thinning vaccine technology, an in vivo study was performed in Atlantic salmon. Four hundred Atlantic salmon parr weighing approximately 50 g were obtained from Cooke Aquaculture’s hatchery (Bingham, Maine) and transferred to the University of Maine Cooperative Extension Diagnostic and Research Laboratory (Orono, Maine). Approximately 115 fish were arbitrarily distributed into three 490-liter tanks for an average stocking density of 12 kg / m3and allowed to acclimate for 14 days. All experimental procedures were approved by the University of Maine’s Institutional Animal Care and Use Committee under Protocol A2023-02-01.

[0111] The study system consisted of a well water partial flow-through and recirculating system at 12 ± 0.5 °C with a mechanical bead filter, bio-filter, UV disinfection, and eight 490-liter tanks. Each tank was supplied with 2L min-1oxygenated water to maintain a dissolved oxygen level of 8.5 ± 2.0 mg / L. Water and fish parameters which included temperature (°C), dissolved oxygen (mg / L), feed observation, fish appearance, mortalities, ammonia (mg / L), and nitrite (mg / L) levels were documented daily. Tanks were also siphoned daily to remove particulates. The salmon were fed a commercial diet (Bio-Oregon, Westbrook, Maine) twice daily at a feed rate of 1.5% body weight per day.

[0112] Twenty Atlantic salmon parr per triplicate tank (60 fish per formulation, 120 fish per tank) were anesthetized in 100 mgL-1MS-222 buffered with sodium bicarbonate and elastomer tagged with an identifying color before being intraperitoneally injected with a 1 ml syringe and 22-gauge needle (Becton- Dickenson). Injection was slightly posterior to the pelvic fin and perpendicular to the ventral surface on the left side with 100 μL of one of six CNM vaccine formulations versus a commercial adjuvant (Montanide™ ISA 763 A VG adjuvant) and a DPBS negative control for a total of seven treatment formulations co- habitated in three replicate tanks. Unvaccinated sentinel fish were held in separate duplicate tanks. Following injection, fish were immediately returned to the respective tanks for recovery. A DPBS + Bacterin group was vaccinated and added to the study ten days later and sampled on the appropriate number of days post-injection to confirm adverse events did not occur as a result of the bacterin. Final treatmentformulations are listed in Table 5.

[0113] Table 5 – Treatment groups used to inject Atlantic salmon for in vivo toxicity and immunogenicity study Treatment Formulation No. fish No. Total per tank replicate No. tanks Fish 1 Sentinel 60 2 120 2 DPBS Only 20 3 60 3 DPBS + Bacterin 20 1 20 4 Unmodified TOCNF + Bacterin 20 3 60 5 Sonicated 1x Amidated TOCNF + Bacterin 20 3 60 6 Sonicated 2x ODA DMF Amidated TOCNF + Bacterin 20 3 60 7 Sonicated 2x EDC NHS Amidated TOCNF + Bacterin 20 3 60 8 Commercial Montanide™ ISA 763A VG + Bacterin 20 3 60

[0114] Sampling was performed on 30 unvaccinated Atlantic salmon for baseline and at 300- and 600-degree days (12 °C) post-injection by euthanizing fish with a lethal dose of MS-222 supplemented with sodium bicarbonate.

[0115] Observed biometrics

[0116] Fish were monitored twice daily over the study for feeding vigor. Biometric data of weights (g) and fork length (mm) were collected for calculating Fulton’s condition factor (K) and mean specific growth rate (SGR or %G) at Baseline, 300-, and 600- degree days post-injection to compare impact on growth.

[0117] External and internal gross examination

[0118] Gross necropsy was performed on all fish to assess adverse reactions to uptake of and dispersion of amidated TOCNF hydrogels. Thus, complete external and internal gross examinations were performed at pre-injection (baseline), 300-, 600-degree days post-injection, and on all mortalities. External examination evaluated injection healing and presence of external lesions including protruding proliferative tissue. Internal visceral and peritoneal adhesions, melanization, fibrosis, granuloma formation, and TOCNFformulation residue were evaluated and scored based on a modified Speilberg scale. Any other observations deviating from normal were described. Histology tissues were collected and analyzed.

[0119] Histopathology

[0120] At the conclusion of the in vivo study (600-degree days post-injection), tissue samples were collected from 3 fish per treatment group per triplicate tank for a total of 64 samples and fixed into 35 mLs of 10% neutral buffered formalin (Fisher Scientific). Tissues were not sampled for histology from the DPBS + Bacterin control group. Tissues included approximately 1 cm2of body wall at the injection site, the pyloric caeca, liver, a digestive tract, and spleen. Tissues were submitted to the New Hampshire Veterinary Diagnostic Laboratory (NHVDL) for histopathologic processing and evaluation. Tissues were routinely processed, embedded in paraffin, sectioned at 5-mm thickness, mounted on charged slides, and stained with hematoxylin and eosin. Tissue samples were blinded to treatment groups and scored on the appearance of inflammation, fibrosis, and granulation tissue according to the rubric in Table 6, and the presence or absence of formulation was noted.

[0121] Table 6 - Scoring rubric for pathology observations of the body wall and coelom in relation to gross Speilberg scores Speilberg Scores Present example Score Visual a earance of Severit of dama e to Bod wall Coelom al s d d

[0122] Examining In vivo Immunogenicity in Atlantic salmon

[0123] Real-time qPCR assay

[0124] RNA isolation, reverse transcription, and real-time qPCR assays for immune genes were carried out. Total RNA was isolated from the head kidney of Atlantic salmon using TRIzol®reagent (Life Technologies, Carlsbad, CA, USA). Briefly, the samples were placed in 2 ml tubes containing 1 ml of TRIzol®and homogenized in the Omni bead ruptor elite (Kennesaw, GA, USA), then 200 µl of chloroform was added with vigorous mixing before being incubated at 4 ℃ for 10 min. The samples were then centrifuged at 12,000 g at 4 ℃ for 15 min. The supernatants containing RNA (400 µl) were transferred to new 1.5 ml tubes, and the RNA was precipitated by adding 400 µl of isopropanol, which were then mixed gently before incubating overnight at -80 ℃. RNA pellets were collected by centrifuging at 12,000 g at 4 ℃ for 10 min, washed with 1 ml of cold 75% ethanol. The supernatants were discarded, and RNA pellets were air-dried for 5 min before they were resuspended in 100 µl nuclease-free water. The purity and quantity of extracted RNA were assessed by NanoDrop™ Spectrophotometer (Thermo Scientific™, USA) (260 / 230 and 260 / 280 ratio ≥ 1.8). The cDNA of samples was synthesized using PrimeScript™ RT reagent kit (Takara Bio, San Jose, CA, USA), following the manufacturer’s instructions. Real-time quantitative PCR was carried out on QuantStudio 3 (Applied Biosystems) in 20 µl total volume reactions and 500 nmol primers according to the protocol provided by the manufacturer. PCR cycling conditions for all genes were initiated with the denaturation step at 95 ℃ for 30s followed by forty cycles at 95 ℃ for 5s, 60 ℃ for 34s, and 95 ℃ for the 30s, 95 ℃ for 3s, 60 ℃ for 30s. Melting curve analysis was performed to verify that a single PCR product was produced. The relative expression of genes involved in immune response (IgM, IgT, and IgD) were determined using primers designed from Atlantic salmon sequences in the NCBI database. The reference gene β-actin was used to normalize the expression levels of the target genes. All primers of the target and reference genes were synthesized by the Integrated DNA Technologies (IDT, Morrisville, NC, USA). The amplification efficiencies of the target and reference genes were quantified according to the specific gene standard curves generated from 10-fold serial dilutions. After verifying that the primers were amplified with 100% efficiency, the relative expression results were analyzed using the 2−ΔΔCtmethod.

[0125] Serological antigen specific antibody response by indirect ELISA

[0126] Tetrameric immunoglobulin M (IgM) is the prominent immunoglobulin found in salmonid serum with a primary function being systemic immunity. Research has demonstrated Atlantic salmon receiving commercial oil-adjuvanted vaccines stimulate antibody production and have IgM concentrations two- to four- times higher compared to un-vaccinated fish. Thus, an indirect enzyme-linked immunosorbent assay (ELISA) was used to determine serum IgM antibody titers specific to Vibrio anguillarum isolates in Atlantic salmon vaccinated with amidated TOCNF formulations compared to an oil-based commercialvaccine control at 600- degree days post vaccination. Whole cell antigen was produced for ELISA use. Vibrio anguillarum isolate VE-2021-0143 were cultured and inactivated as previously described. The bacterin was centrifuged at 3,500 x g for 15 minutes. The cell pellet was resuspended in an equal volume of DPBS and stored at 4 °C until ready for use.

[0127] Optimization of the ELISA was performed to determine optimal antigen concentration of the V.anguillarum isolate (1x106, 1x107, and 1x108CFU), the optimal primary antibody concentration (0, 1 / 15, 1 / 33, 1 / 50, and 1 / 100 dilutions in DPBS), and the optimal secondary antibody concentration (0, 1 / 2000, 1 / 5000, 1 / 10,000, and 1 / 15,000 dilutions in DPBS).

[0128] ELISAs were optimized and then performed from serum of three fish per treatment per tank replicate (9 fish per treatment) according to Aquatic Diagnostics Ltd. manufacturer’s instructions using whole cell antigen (1.0 x 108CFU) with incubations performed for 24 hours at 4 °C between each coating step. Serial dilutions were performed with fish serum to quantify antibody titer. Fish serum was diluted in duplicate wells (1 / 10, 1 / 20, 1 / 40, 1 / 80, 1 / 160, 1 / 320, 1 / 640, 1 / 1280, 1 / 2560, 1 / 5,120, 1 / 10,240, and 1 / 20,480) in DPBS. Primary antibody, mouse anti trout / Atlantic salmon IgM monoclonal (Aquatic Diagnostics), was diluted to 1 / 33 in the conjugate buffer and 100 µL per well was added. Plates were incubated for one hour at 25 °C. The secondary antibody, goat anti-mouse IgG peroxidase (Sigma-Aldrich), was diluted to 1 / 2000 in DPBS and 100 µL per well was added. Plates incubated for one hour at 25 °C. The chromogenic substrate for oxidation via horseradish peroxidase, 1-Step™ Ultra TMB ELISA Substrate Solution (3,3',5,5' tetramethylbenzidine; Fisher Scientific), was added (100 µL per well) and the reaction was stopped after 10 minutes by adding 50 µL per well of 2M sulfuric acid (Fisher Scientific). Plates were mixed and the absorbance was recorded at 450 nm using a spectrophotometer (BioTek Synergy™).

[0129] Calculations and statistical analysis

[0130] In vitro characterization

[0131] Calculations were performed to determine shear rate (γ),̇ shear stress (τ), viscosity (η), power- law slope, and flow behavior index of the formulations to produce log-log plots. A log-log plot was produced from the viscosity (^) and shear rate (^). A negative linear relationship denoted shear thinning, a positive linear relationship indicated shear-thickening, and a horizontal relationship designated a Newtonian fluid. A second log-log plot from the shear stress and shear rate of the material was produced to calculate the flow behavior index (n) of the polymer matrix. The slope of the linear regression indicated the flow behavior with n < 1 as shear thinning, n = 1 as Newtonian, and n > 1 as shear-thickening. Further, the area under the ODA curves generated from FT-IR analysis were calculated for comparison. Calculations were performed and graphs were produced using Origin 2022b.

[0132] Examining in vivo toxicity

[0133] Sample sizes for the in vivo trial was determined by power analysis to detect differences inthe modified Speilberg scale of tissue reaction and antibody response in Atlantic salmon between the different treatment groups using G Power Version 3.1 software. Degree day was calculated by multiplying the mean water temperature during the course of the study by the number of study days (DD = ((T0+ T1+…) / no. of days) x no. of days). Fulton’s condition factor (K) was calculated using 100WL-3where W is body weight (g) and L is fork length (cm). Mean SGR (%G) was calculated for each treatment by replicate tank using G (%) = ((Ln(wf) - Ln(wi)) x 100) / t where wfis the mean weight (g) of the treatment within a replicate tank at 300- or 600- degree days, wiis the mean weight (g) of the fish at the baseline sample, and is the number of true days at the time point. Fulton’s condition factor and SGR was calculated so that differences in condition and growth of the fish vaccinated with the different formulations could be compared. For adverse events, cumulative incidence (%) was calculated by the number of mortalities per treatment group divided by the number of total fish in the treatment group by 600 degree-days. Cumulative censored (%) was calculated by the number of censored fish per treatment group divided by the number of total fish in the treatment group by 600 degree-days. Mantel-Haenszel Hazard Ratios were calculated from comparisons of the mortality curves of each treatment group to the Montanide commercial adjuvanted control vaccine. Prevalence of adverse reactions (%) ranging from mild to severe as observed using the modified Spielberg scoring growth data were analyzed at each time point by treatment.

[0134] Mean differences in condition factor, SGR, Speilberg scoring, and histology scores were first examined for normality using a Shapiro-Wilk test. Afterward, a two-way ANOVA mixed-effects analysis was used to determine statistical differences. This model was chosen because of unequal replicates in the unvaccinated sentinel and DPBS + Bacterin groups. Groups with a p < 0.05 were considered statistically significant. Where p < 0.05, a pairwise comparison using Tukey’s post-hoc analysis was performed. The adverse mortalities in Atlantic salmon vaccinated by treatment groups were analyzed by Kaplan-Meier time-to-death analysis with mortality curves compared using both Log-Rank (Mantel-Cox) and Gehan- Breslow-Wilcoxon Analysis. All statistical analyses were performed using GraphPad Prism 9.5.1. Results were expressed as mean ± standard error of the mean (SEM) unless otherwise noted so groups with a p < 0.05 were considered statistically significant.

[0135] Examining in vivo immunogenicity

[0136] From ELISA data, mean OD450nmof duplicate wells were calculated. Coefficient of variation (%) was determined between duplicate wells (intra-assay %CV) and duplicate plates (inter-assay %CV) by dividing the standard deviation of the replicates by the mean. %CV less than 15% between inter-assay replicates and 10% between intra-assay replicates were accepted as valid. Titers were determined by end- point titer, or the last doubling dilution, to give a positive OD450nm. Mean titers from ELISA data were compared statistically by two-way ANOVA using GraphPad Prism version 9.5.1.

[0137] Results

[0138] In vitro characteristics of amidated TOCNF hydrogel formulations

[0139] Mechanical properties: rheology

[0140] Rheology was used to determine the shear strength of the amidated TOCNF hydrogel formulations with friction maintained at max shear rate tested (^ = 10 s-1). Formulations with increased reagents (2x EDC NHS and 2x ODA DMF) failed on the rheometer at lower shear rates (^ = 3 s-1) by either sample slipping or sample fracture and could not be further analyzed for rheological properties. All sonicated 2x EDC NHS amidated TOCNF formulations displayed sample slipping with shear rate, ^ < 3 s-1.

[0141] Viscosity (^) and shear rates (^) were logarithmically graphed from amidated TEMPO CNF formulations exhibiting no sample failure. As shear rate increased, viscosity decreased such that all tested variations of amidated TOCNF had a negative linear slope indicating a power-law relationship of fluids displaying shear thinning characteristics. No plateau nor positive linear slope regions were observed, indicating the samples did not contain Newtonian nor shear thickening characteristics (FIG.5).

[0142] The calculated flow behavior index corroborated with the power-law slope results (Table 7). All homogeneous samples that were successfully tested demonstrated shear thinning characteristics as shown by the negative power-law slopes and flow behavior index. These results indicate the formulation to exhibit the most shear-thinning behavior under shear stress was the sonicated 1x amidated TOCNF hydrogel as denoted by the negative power-law slope and the flow behavior index closest to zero.

[0143] Table 7 – Mechanical properties of TOCNF formulations examined by rheology to quantify shear-thinning characteristics Sample Outcome Power-law slope Flow Behavior (Pa •S2) Index0.0334 1x Amidated TOCNF Success -0.88028 0.0493 Sonicated 1x Amidated TOCNF Success -1.10152 0.0151 2x ODA DMF Amidated TOCNF Fractured N / A N / A Sonicated 2x ODA DMF Amidated TOCNF Success -1.0584 0.0263 2x EDC NHS Amidated TOCNF Fractured N / A N / A Sonicated 2x EDC NHS Amidated TOCNF Slipped N / A N / A 2x Amidated TOCNF Success -0.83082 0.1776 Sonicated 2x Amidated TOCNF Fractured N / A N / A

[0144] Structural properties: scanning electron microscopy

[0145] Scanning electron microscopy (SEM) was used to visualize cross sections of 4 amidated TOCNF formulations for surface properties and homogeneity. The SEM images for each amidated TOCNF formulation had observed differences. The sonicated 2x EDC NHS amidated TOCNF formulation showed significant branching and unorganized flaking under 50x magnification (FIG.6A). Under 1500x magnification, this branching was identified as webbing (FIG.6B). The 1x amidated TOCNF hydrogel formulation demonstrated porous surface with organized flakes and slight clumping under 50x magnification (FIG.6C) and under 1500x magnification, this was described as flakes with mild webbing present (FIG.6D). There was less clumping observed when sonication was used during production of the 1x amidated TOCNF hydrogel (FIG.6E), and under 1500x magnification, the formulation appears visually more homogenous with less branching in comparison to the 1x amidated TOCNF with no sonication (FIG. 6F). Several spherical orbs were visualized throughout the sonicated 2x ODA DMF amidated TOCNF formulation (FIG.6G). These orbs appear to be attached to the cellulose nanofibers with branching between some of the clusters (FIG.6H). InLens SEM imaging was used to view the sonicated 2x ODA DMF amidated TOCNF formulation to improve surface contrast and enable differentiation of material (FIGS.7A-7B).

[0146] SEM images of the sonicated 2x ODA DMF amidated TEMPO CNF exhibited significant clumping (FIG.7A), thus InLens SEM imaging was performed under 50x magnification to better visualize the formulation. Variations in grayscale with large sections exhibiting a heterogeneous appearance was observed (FIG.7B) compared to standard SEM imaging under 25x magnification.

[0147] Chemical properties: fourier-transform infrared spectroscopy

[0148] Rheology and SEM results indicated lack of homogeneity of octadeclyamine and TEMPO- oxidized CNF, thus FT-IR was used to determine degree of cross-linking within the amidated TOCNF formulations and to examine for any residual or trace chemicals remaining after production. First, sonicated amidated TOCNF formulations were analyzed in comparison to unmodified TOCNF by FT-IR (FIG.8).

[0149] Consistent areas of the absorbance spectrum across all groups were accounted for by the large concentration of water within the formulations. This was seen at the largest absorbance peak between 3600 - 3100 cm-1corresponding to the presence of O-H asymmetric stretching. The peak at 2300-1900 cm-1was attributed to the combination bond present within water. Finally, the absorbance peak within the range of 1750-1600 cm-1was ascribed to the presence of a phenomenon within water molecules called scissor bending.

[0150] Within all sonicated variations of the amidated TEMPO CNF there were twin peaks present at the wavelength of 2924 cm-1and 2857 cm-1. These twin peaks correlated to the presence of -CH2attributed to ODA. Absorbance peaks at these locations increased with concentration of the chemicals suchthat the area under the curve was calculated and compared (Table 8). Also noteworthy, twin absorbance peaks at wavelength 1450 cm-1were visible with increasing ODA / DMF and EDC / NHS.

[0151] Table 8 – Calculated area under the ODA curve for sonicated amidated TOCNF samples Sample Wavenumber 2924 cm-1Wavenumber 2857 cm-12x Amidated TOCNF 0.39578 0.16039 2x EDC NHS Amidated TOCNF 0.26865 0.13201 2x ODA DMF Amidated TOCNF 0.08786 0.06128 1x Amidated TOCNF 0.03157 0.02082 Unmodified TOCNF 0 0

[0152] Examining in vivo toxicity in Atlantic salmon

[0153] Adverse events

[0154] Twenty-six mortalities occurred during the course of the study period with the three replicate tanks experiencing 6.7, 6.7, and 7.5% total mortality, respectively. Kaplan-Meier time to death analysis to compare the cumulative incidence of mortalities (%) across each treatment group during the course of the study with significant differences observed between survival curves compared using both the Log-Rank (Mantel-Cox) and Gehan-Breslow-Wilcoxon analysis (Log-rank; ^2= 56.97, p < 0.0001 and Gehan- Breslow-Wilcoxon; ^2= 56.96, p < 0.0001) (FIG.9).

[0155] Mantel-Haenszel hazard ratios (HR) were calculated with 95% Confidence Intervals (95% CI) revealing unvaccinated sentinel Atlantic salmon had the lowest risk of mortality when compared to vaccinating with the commercial oil-adjuvanted positive control group (HR = 0.0498; 95% CI, 0.000078 - 3.183; p = 0.3173). Vaccinating Atlantic salmon with the sham negative control groups (DPBS Only and DPBS + Bacterin) reduced risk of mortality compared to vaccinating with the commercial oil-adjuvanted positive control group (DPBS, HR = 0.1353; 95% CI, 0.0027-6.821; p = 0.3173 and DPBS + Bacterin, HR = 0.2636; 95% CI, 0.00285-24.36; p = 0.5637). Additionally, vaccinating Atlantic salmon with unmodified TOCNF + Bacterin also reduced risk of mortality equivalently to the sham negative control when compared to vaccinating with the commercial oil-adjuvanted positive control group (HR = 0.1353; 95% CI, 0.0027- 6.821; p = 0.3173). Intraperitoneally vaccinating with amidated TOCNF increased risk of mortality in Atlantic salmon (HR = 2.719; 95% CI, 0.383 - 19.300; p = 0.3235), however survival curves were not statistically significant compared to the commercial oil-adjuvanted positive control by Gehan-Breslow- Wilcoxon analysis. Increasing amidation of TOCNF increased risk of mortality significantly in 2x ODA DMF amidated TOCNF (HR = 5.461; 95% CI, 1.670-17.860; p = 0.0054) and in 2x EDC NHS amidated TOCNF (HR = 5.713; 95% CI, 1.830-17.830; p = 0.0030) compared to the commercial oil-adjuvantedpositive control. These results are summarized in Table 8.

[0156] Table 8 – Results of the Mantel-Haenszel hazard ratios of vaccinating Atlantic salmon with the sham negative control groups and amidated TOCNF groups compared to the commercial oil-adjuvanted positive control group at 600 degree-days post-injection showing Gehan-Breslow-Wilcoxon analysis Formulation % % Cumulative M-H Odds Ratio (95% Cumulative Survival CI) ^2(df), p-value Incidence (No. (No. Morts / Censored / No. No. Total) Total) Sentinel 0 (0 / 120) 100 (120 / 120) 0.0498 (0.00078-3.183) 2.000 (1), p = 0.3173 DPBS Only 0 (0 / 60) 100 (60 / 60) 0.1353 (0.0027-6.821) 1.000 (1), p = 0.3173 DPBS + B. 0 (0 / 20) 100 (20 / 20) 0.2636 (0.0029-24.36) 0.333 (1), p = 0.5637 Unmodified TOCNF + B. 0 (0 / 60) 100 (60 / 60) 0.1353 (0.0027-6.821) 1.000 (1), p = 0.3173 1x Amidated TOCNF + B. 5.0 (3 / 60) 95 (57 / 60) 2.719 (0.383-19.300) 0.9748 (1) p = 0.3235 2x ODA DMF Amidated TOCNF + B. 16.7 (10 / 60) 83.3 (50 / 60)p = 2x EDC NHS Amidated TOCNF + B. 18.3 (11 / 60) 81.7 (49 / 60) 5.713 (1.830-17.830) 8.828 = 0.0030 Commercial Montanide™ ISA 763A VG 1.7 (1+ B. / 60) 98.3 (59 / 60) Note: M-H: Mantel-Haenszel Odds Ratio; CI: 95% Confidence Interval; B: Bacterin. Value in bold denotes statistically significant differences between mean survival curves in formulations that were intraperitoneally injected Atlantic salmon compared to commercial Montanide control (^ = 0.05)

[0157] Observed biometrics

[0158] Fish maintained vigorous feeding and normal behavior throughout the course of the study. A two-way ANOVA mixed-effects analysis was performed to examine the effect of vaccine formulation and sampling time point on Fulton’s condition factor (K) in the Atlantic salmon parr intraperitoneally vaccinated. Analysis revealed there was not a statistically significant interaction between the effects of formulation nor time point (F (7, 347) = 1.039, p = 0.04031) on K. Simple main effects analysis showed the formulations did not have a statistically significant effect on K (F (7, 347) = 1.135, p = 0.3406) and timepoint did not have a statistically significant effect on K (F (1, 347) = 0.06243, p = 0.8028). It is important to note mean K was greater than 1.0 across all formulations at 300- degree days (KM300= 1.073, SD = 0.141) and at 600- degree days (KM600= 1.077, SD = 0.139) including in the sentinel fish (KMS= 1.122, SD = 0.134) indicating fish maintained healthy body condition as it relates to growth throughout the study (FIG. 10).

[0159] To more closely examine differences in growth between treatment groups, an ordinary two- way ANOVA was performed to analyze the effect of vaccine formulation and sampling time point on SGR (%G) in Atlantic salmon parr intraperitoneally vaccinated. Analysis revealed there was not a statistically significant interaction between the effects of formulation and time point (F (6, 13) = 1.528, p = 0.2447) on %G. Simple main effects analysis showed the formulations had a statistically significant effect on %G (F (6, 13) = 3.393, p = 0.037) and time point had a statistically significant effect on %G (F (1, 13) = 18.76, p = 0.0008). A pairwise comparison using Tukey’s post-hoc analysis revealed significant differences observed between the 300 degree day treatment groups such that SGR was significantly lower in unvaccinated sentinel Atlantic salmon compared to negative sham vaccinated controls (DPBS Only, p = 0.0248; Unmodified TOCNF + Bacterin, p = 0.0167). The DPBS + Bacterin group was not included in the analysis, however it should be noted that SGR was equivalent to DPBS Only and Unmodified TOCNF + Bacterin group at both 300 degree days and 600 degree days (%Gmean = xx, SD = xx at 300- degree days and %Gmean = xx, SD= xx at 600- degree days). While significant differences were observed between treatment groups at 300 degree days post-injection, SGR had equilibrated to positive growth with no observed significant differences by 600 degree days post-injection (FIG.11).

[0160] External and internal gross pathology

[0161] External gross pathology

[0162] External gross necropsy of mortalities during the course of the study largely showed presence of Saprolegnia related lesions and / or infections in ten of the twenty-six mortalities (38.5%) in the 2x ODA DMF and 2x EDC NHS amidated TOCNF groups. Hemorrhaging of the ventral surface surrounding the injection site was also observed in nine out of the twenty-six mortalities (34.5%) in the 2x ODA DMF and 2x EDC NHS amidated TOCNF groups (FIGS.12A-12B).

[0163] At 300-degree days post-injection, external proliferative masses surrounding the injection site were noted as presented in Table 9. No external proliferative masses were observed at the 600-degree day sampling time point

[0164] Table 9 – Prevalence of external proliferative masses surrounding injection site of Atlantic salmon at 300-degree days post-injection Treatment Group No. Proliferative Mass / Total No. Fish (%) Sentinel 0 / 30 (0%) DPBS Only 0 / 30 (0%) DPBS + Bacterin 0 / 10 (0%) Unmodified TOCNF + Bacterin 2 / 30 (6.67%) 1x Amidated TOCNF + Bacterin 3 / 30 (10%) 2x ODA DMF Amidated TOCNF + Bacterin 6 / 30 (20%) 2x EDC NHS Amidated TOCNF + Bacterin 5 / 30 (16.67%) Commercial Montanide™ ISA 763A VG + Bacterin 1 / 30 (3.33%)

[0165] Internal gross pathology

[0166] Internal examination of mortalities that occurred during the in vivo study frequently revealed mottled or pale liver (34.5%) and severe hemorrhaging in the vicinity of the formulation through the epidermal layer to the viscera with involvement of the pyloric caeca primarily in the 2x ODA DMF and 2x EDC NHS amidated TOCNF groups (38.5%) (FIG.13).

[0167] Internal gross pathology results at 300- and 600-degree days post-injection using the Speilberg rubric during the in vivo study are presented here. An ordinary two-way ANOVA was performed to analyze the effect of TOCNF hydrogel formulation and sampling time point on the mean Speilberg score of abdominal adhesions (FIG.14, upper left), visceral melanization (FIG.14, upper right), peritoneal melanization (FIG.14, lower left), and abdominal residue (FIG.14, lower right) in the Atlantic salmon parr intraperitoneally injected with amidated TOCNF formulations at 300- degree and 600-degree days post- injection. Where p < 0.05, a pairwise comparison using Tukey’s post-hoc analysis was performed with significant treatments indicated. Ordinary two-way ANOVA analysis results are presented in Table 10.

[0168] Table 10 – Results of ordinary two-way ANOVA (^ = 0.05) of mean Speilberg tissue reactions in Atlantic salmon vaccinated with test formulations Source of Variation 4 0 0 3; q ; g ; q . denotes statistically significant differences between mean Speilberg scores in formulations that were intraperitoneally injected in Atlantic salmon (^ = 0.05).

[0169] Also noteworthy, severe gross pathologies were frequently noted in the 300-degree day post- injection sampling time point associated with amidated TOCNF formulations that were not present in DPBS with and without Bacterin, unmodified TOCNF, and oil-adjuvanted controls. Severe gross pathologies often noted included pale liver, ascites, and edema of the pyloric caeca (FIG.15). These reactions were not observed at the 600-degree day post-injection sampling time point.

[0170] Example III – Biocompatibility of intraperitoneally implanted TEMPO-oxidized cellulose nanofiber hydrogels for antigen delivery in Atlantic salmon (Salmo salar L.) vaccines

[0171] This example describes the intraperitoneal implantation of a crosslinked TEMPO cellulose nanofiber hydrogel formulated with a Vibrio anguillarum bacterin in Atlantic salmon with macroscopic and microscopic monitoring to 600-degree days post-implantation. Results demonstrated a modified passive integrated transponder tagging (PITT) device allowed for implantation of the hydrogel. However, the Atlantic salmon implanted with the hydrogels exhibited a significant foreign body response (FBR) compared to sham-injected negative controls. The FBR was characterized by gross and microscopic external and visceral proliferative lesions, granulomas, adhesions, and fibrosis surrounding the hydrogelusing Speilberg scoring of the peritoneum and histopathology of the body wall and coelom. Acutely, gross monitoring displayed rapid coagulation of blood in response to the implantation wound with development of fibrinous adhesions surrounding the hydrogel by 72 hours post-implantation consistent with early stage FBR. These results inform on the innate immune response to an implanted biopolymer hydrogel in Atlantic salmon.

[0172] Vaccination by injection is a commonly used delivery method to administer small, concentrated known doses of antigens and can be produced in multivalent form to protect against multiple pathogen strains or diseases within one injection. Most injectable fish vaccines use oil-based adjuvants that act on the adaptive immune system, increasing both the humoral and / or cell-mediated immune responses of the fish. Negative consequences of vaccination can result from up-regulation of the innate immune response, causing inflammation, the formation of adhesions in and around the injection site, and strong infiltration of mononuclear cells. This response can further produce a significant growth penalty and go on to cause welfare issues such as adhesions and internal organ damage. These undesired reactions occur in Atlantic salmon (Salmo salar L.), a species that is extensively farmed as a high valued human food protein source rich in omega-3 fatty acids. Atlantic salmon are susceptible to a host of diseases and are routinely vaccinated by injection which has improved disease prevention and increased productivity. However, salmon are also negatively impacted by the adverse effects attributed to vaccination. Thus, the challenge remains to create a cost-effective and adequately robust immune response for long-lasting protection that minimizes adverse effects.

[0173] Materials and methods

[0174] Historically, porous biomaterials have been the target of research efforts for numerous applications including cell and tissue engineering and regenerative medicine. Recent successes in vaccination strategies and immunotherapies demonstrate strong promise to translate across multiple species with the developing area of biopolymer nanotechnology expanding into the field of vaccinology. In general, biomaterials are currently being investigated to improve the biocompatibility and effectiveness of vaccines and other immunotherapies over a long duration. Polysaccharides such as chitosan, Poly (d,l- lactide-co-glycolic acid) (PLGA), alginates, and others at the nanoscale as vaccine adjuvants show promising results.

[0175] It has been hypothesized that in vivo modulation of host immune cells can be achieved long- term through the use of three-dimensional (3D) porous biomaterial scaffolds and hydrogels. Polymeric scaffolds and hydrogels are 3D polymer networks that swell in water or other aqueous media while remaining insoluble in water. These 3D polymer networks can exist at either the macroscopic or nanoscale size for controlled release of therapeutants such as drugs, nucleic acids, proteins, cells, and other antigens. These implantable engineered materials can serve as local depots to recruit immune cells and / or modulatetheir function. Further, it has been demonstrated that hydrogels provide large porosity and surface area enabling tunable release kinetics of antigens and the ability to deliver multiple biomolecules simultaneously.

[0176] Hydrogels have design flexibility, hydrophilic nature, and high swelling ratio, which creates permeability to nutrients, metabolites, oxygen, and waste materials. Various materials and crosslinking techniques have been investigated to tune the pore size, mechanical strength, and degradation rates of the hydrogels. Further, the use of natural polymers in the production of these delivery systems have been shown to offer advantageous cell adhesion, eventual degradability, reduced inflammatory response, targeted specific responses, and biocompatibility.

[0177] Cellulose nanomaterials (CNM), a type of polysaccharide at the nanoscale, are useful for drug delivery. For example, cellulose nanofibers (CNF) modified during manufacturing to form TEMPO- oxidized CNF can further be physically cross-linked into a biopolymer hydrogel system for drug delivery. It has been demonstrated that TEMPO-oxidized CNF is biocompatible both in vitro and in vivo, however conflicting results on the degree of biocompatibility of cellulose nanomaterials can also be found in literature where an inflammatory immune response was triggered by implanting CNM scaffolds formulated with specific chemical and physical properties. Nonetheless, CNM is an attractive biomaterial for use in aquaculture vaccine systems as an adjuvant / depot. In particular, the nanomaterial is a readily available renewable resource that is biodegradable, biocompatible, cost effective, and highly tunable.

[0178] The safety of unmodified TEMPO-oxidized CNF intraperitoneally injected into Atlantic salmon observed described in the example above prompted research into the use of a citric acid crosslinked TEMPO-oxidized cellulose nanofiber (TOCNF) hydrogel as an in vivo hydrogel system for vaccination. The first step was to examine the biocompatibility of the TOCNF hydrogels in vivo. Biocompatibility is defined by the Food and Drug Administration in ISO 10993-1 as "The ability of a device material to perform with an appropriate host response in a specific situation." and must be considered when using implanted biomaterials for in vivo immunostimulants. This research hypothesized that TOCNF hydrogels formulated with an inactivated bacterial whole-cell antigen (Vibrio anguillarum) would be biocompatible when intraperitoneally implanted into Atlantic salmon. Results would support the application of a new generation of vaccines for aquaculture as well as eventually translate to other finfish species and terrestrial mammals. The objectives of this research sought to address this question through 1.) preparing citric acid crosslinked TEMPO-oxidized cellulose nanofiber hydrogels with a fish pathogen bacterin, 2.) developing a method for the intraperitoneal implantation of TOCNF hydrogels into Atlantic salmon, 3.) examining in vivo biocompatibility by gross and histological examination at 600- degree days post-implantation, and 4.) describing the acute response to implantation.

[0179] Materials and methods

[0180] The use of experimental fish was under scientific research protocols of the University of Maine, Institutional Animal Care and Use Committee (IACUC Protocol #: A2021-01-02), and complied with all relevant international animal welfare laws, guidelines, and policies.

[0181] Preparation of Vaccine Using CNM Matrices as an Adjuvant

[0182] Bacterin production

[0183] Vibrio anguillarum, the causative bacterial agent of vibriosis in Atlantic salmon, was chosen for this example as it is a well-studied and characterized global bacterial pathogen that is commonly used as inactivated bacterial whole-cell antigens in commercial vaccine formulations. A bacterin was produced to formulate the vaccine with a novel CNM hydrogel depot for examining biocompatibility in Atlantic salmon. V.anguillarum isolate VE-2021-0143 was obtained from University of Maine Aquatic Animal Health Laboratory (AAHL)’s frozen bacterial stock culture collection. The isolate was streaked onto Tryptic Soy Agar with 5% sheep blood and 1.5% NaCl (BA, Northeast Laboratory Services) and allowed to grow at 16 + 1°C for 48 hours. A single colony forming unit (CFU) was selected from the plate to inoculate one 10 mL tube of Trypticase Soy Broth + 1.5% NaCl (TSB, Becton Dickinson and Fisher Scientific) under aerobic conditions for 48 hours at 16 + 1°C on a 180-rpm shaker. The 10 mL culture was used to inoculate 500 mLs TSB for culture by incubating at 16 °C while stirring on a magnetic stir plate for 36 hours to OD600nm = 0.817 determined using an Ultraspec10 Cell Density Meter (Biochrom). The culture was gram stained for purity and serial dilutions from 10-1to 10-10were performed in 2.7 mLs DPBS. Next, 100 µL of the 10-4to 10-10dilutions were plated on BA in duplicate and incubated for 48 hours at 16 + 1°C for viable plate counts to determine the colony forming units per milliliter (CFU / mL) of the culture. The culture mean was determined to be 1.22 x 109CFU / mL. The culture was inactivated with 0.02% formalin (1 mL 37% Formalin into 500 mLs culture, Fisher Scientific) and stirred for 48 hours at 22 °C. Quality control was performed by adding 1 mL of V. anguillarum bacterin into 10 mLs of sterile TSB to confirm inactivation. The inactivated bacterin was stored at 4°C until use.

[0184] Hydrogel production

[0185] Hydrogels were developed, optimized, and prepared using citric acid to physically crosslink TEMPO-oxidized cellulose nanofibers and combined with the Vibrio anguillarum bacterin as the antigen to formulate a vaccine. In this approach, the CNM adjuvant functioned as a stationary depot in Atlantic salmon. TEMPO CNF hydrogels were prepared by diluting 1.1 wt% TEMPO CNF (Process Development Center University of Maine) to 0.55 wt% in 109CFU / mL V.anguillarum bacterin and then vacuum filtering the mixture to 1.7 wt%. The 1:1 mixture of 1.1 wt% TEMPO CNF and 109CFU / mL V.anguillarum bacterin was stirred together until homogenous and vacuum filtered using a hydrophobic 0.45 µm nylon membrane filter (Millipore Sigma), a vacuum pump, filtering flask, and Buchner funnel to remove waterfrom the hydrophilic TEMPO CNF and V. anguillarum bacterin mixture. After filtration, the hydrogels were extruded through a modified 1 mL syringe with the Luer-Lok connection removed to mold the hydrogels into 0.1 mL cylindrical shapes. Hydrogels were then added into a 10 wt% citric acid solution for a final concentration of 0.5 mol / L (99.5% citric acid diluted in DI water, Sigma-Aldrich) for 24 hours at 22 + 2 °C to physically cross-link the TEMPO CNF fibers. After cross-linking in citric acid, the hydrogels were washed in DI water once every 24 hours over 2, 4, 6, and 8 days to remove residual citric acid thus raising the pH incrementally over time. The last 24-hour wash was performed with Dulbecco’s PBS (DPBS). This final DPBS wash raised the pH to neutral. More specifically, 2-days of DI water washes was the minimum time required to return the pH to 7.0 ± 0.2 after the final DPBS wash. By day 6, the pH plateaued around 7.0. The 4-day time point was selected as the midpoint and eight days of washing post- production was chosen as the end point because the hydrogel pH had stabilized above 7.0 for multiple days. Afterward, the DPBS wash solutions were examined for presence of antigen using BactoviewTMLive Red fluorescent bacterial (1x, Biotin) staining to confirm antigen did not diffuse from hydrogel during washes. TOCNF hydrogels were dehydrated for 30 minutes at 46 °C (Harvest Maid APS FD-300) to remove approximately 50% of excess water and stiffen the hydrogels to aid delivery. Final hydrogels were equivalent to a 100 µL dose which contained approximately 108CFU of V.anguillarum bacterin per hydrogel and measured 2.5 - 3.0 mm in diameter and 5.0 mm in length. Hydrogels were stored at 4 °C until use. All the hydrogels were rehydrated in DPBS for 24 hours before implantation.

[0186] Implantation of TEMPO-Oxidized CNF Hydrogel

[0187] An in vivo method for intraperitoneal implantation of the hydrogel into Atlantic salmon parr was first investigated. Atlantic salmon parr with an average weight of approximately 20-30 g were obtained from Cooke Aquaculture’s hatchery (Bingham, Maine). Fish were randomly distributed into two- 490-liter tanks for an average stocking density of 14.5 kg / m3and allowed to acclimate for 21 days upon arrival at the University of Maine Cooperative Extension Diagnostic and Research Laboratory (Orono, Maine).

[0188] The study system consisted of a well water partial flow-through recirculating system at 14 ± 2 °C with a mechanical bead filter, bio-filter, UV disinfection, and four- 490-liter tanks. Each tank was supplied with 2L min-1oxygenated water to maintain a dissolved oxygen level of 8.5 ± 2.0 mg / L. Water and fish parameters including temperature (°C), dissolved oxygen (mg / L), feed observation, fish appearance, mortalities, ammonia (mg / L), and nitrite (mg / L) levels were documented daily. Tanks were also siphoned daily to remove particulates. The salmon were fed a commercial diet (Bio-Oregon, Westbrook, ME) twice daily at a feed rate of 1.5% body weight per day.

[0189] Five Atlantic salmon were arbitrarily selected for implantation of one of seven treatment formulations (Table 11) for a total of 35 Atlantic salmon. The TOCNF hydrogels were washed for 4-, 6-, and 8-days during production with DI water for all days except the last 24 hours occurred in DPBS. Thevarious wash times were chosen to compare in vivo effects of pH changes from the citric acid crosslinking. Fish were anesthetized in 100 mg / L tricaine methanesulfonate (MS-222, Syndel USA) buffered with 150 mg / L sodium bicarbonate (Proline USA) until reactions to external stimuli ceased. After sedation, an incision approximately 3.0 - 5.0 mm in length was made slightly posterior to the dorsal fin and parallel to the surface on the right side of the fish with a sterile scalpel. Hydrogel formulations were inserted through the incision into the peritoneum cavity using forceps. The DPBS + bacterin control group was injected into the incision using a 1 mL syringe with 28-gauge needle. The fish were then fluorescently elastomer tagged using a 1 mL sterile syringe and 30-gauge needle along the jawline subdermally.

[0190] Table 11 – Treatment formulations and pH for in vivo delivery study Treatment Hydrogel Bacterin Wash pH After DI Final No. of Fish Formulation Time Water Wash pH 1 1.7% TEMPO CNF None 4 days 2.96 6.8 5 2 1.7% TEMPO CNF None 6 days 3.87 7.0 5 3 1.7% TEMPO CNF None 8 days 4.68 7.2 5 4 1.7% TEMPO CNF 100 uL 4 days 2.96 6.8 5 5 1.7% TEMPO CNF 100 uL 6 days 3.87 7.0 5 6 1.7% TEMPO CNF 100 uL 8 days 4.68 7.2 5 7 None 100 uL None - 7.25 5

[0191] After 600- degree days, all fish were euthanized with 250 mgL-1MS-222 buffered with sodium bicarbonate. Biometric data on weight (g) and length (mm) were collected. Gross necropsy was performed on all fish to assess adverse reactions to uptake of and dispersion of TOCNF hydrogels. Thus, complete external and internal gross examinations were performed at 600-degree days post-implantation. External examination evaluated incision healing and presence of external lesions including protruding proliferative tissue. Internal visceral and peritoneal adhesions, melanization, fibrosis, granuloma formation, and hydrogel residue were evaluated based on the modified Speilberg scale (Table 12).

[0192] Table 12 – Modified Speilberg scale of tissue reaction Score Visual appearance of abdominal cavity on to ng he or s. ns

[0193] Examining In vivo Biocompatibility in Atlantic salmon

[0194] In vivo study

[0195] To assess biocompatibility of the TOCNF hydrogel as a stationary depot, a second in vivo study was performed in Atlantic salmon. Ninety-six Atlantic salmon parr weighing approximately 80 g were obtained from Cooke Aquaculture’s hatchery (Bingham, Maine) and transferred to the University of Maine Cooperative Extension Diagnostic and Research Laboratory (Orono, Maine). The study was performed in duplicate random block design systems. In each replicate system, eight fish per treatment were arbitrarily distributed into five randomized 75-liter tanks for an average stocking density of 8.5 kg / m3with an additional sixth tank as unvaccinated sentinel control fish. Thus, the total study design consisted of twelve 75-liter tanks with 16 fish per treatment (Table 13). Fish were and allowed to acclimate for 14 days.

[0196] The study system consisted of a recirculating well water partial flow-through system at 14.5 °C in two rack systems consisting of six 75-liter tanks per rack, each with a bio-filter, UV disinfection sterilization, and located at the University of Maine Cooperative Extension Diagnostic and Research Laboratory (Orono, Maine). Each tank was supplied with 2L min-1oxygenated water to maintain 9.0 ± 2.0 mg / L dissolved oxygen. Tank and system parameters including temperature (°C), dissolved oxygen (mg / L), feed observation, fish appearance, mortalities, ammonia (mg / L), and nitrite (mg / L) levels were documented daily. Tanks were also siphoned daily to remove particulates. The salmon were fed twice daily a commercial diet (Bio-Oregon, Westbrook, ME) at a feed rate of 1.5% body weight per day.

[0197] Eight Atlantic salmon parr per replicate tank (16 fish per formulation) were anesthetized in 100 mgL-1MS-222 buffered with sodium bicarbonate and elastomer tagged with an identifying color. Due to poor retention and incision healing using forceps in the initial piloting of TOCNF delivery, a 3.0 - 5.0 mm incision was made with a scalpel slightly posterior to the pelvic fin and perpendicular to the ventral surface on the left side and then TOCNF hydrogel formulations (Table 13) were intraperitoneally implanted using a device the research team designed and manufactured by modifying a passive integrated transponder tagging device used to implant PIT tags into Atlantic salmon (Biomark, MK10). A 3D printed plastic blunt stopper was added to the end of the plunger and an 8-gauge beveled stainless steel luer-lok needle (W. W. Grainger) was attached for implanting (FIGS.17A-17B). DPBS + bacterin control group was injected into the incision using a 1 mL syringe with 28-gauge needle as a negative control comparison.

[0198] Following implantation, fish were immediately returned to the respective tanks for recovery and monitored for adverse reactions at 300 and 600- degree days (24 and 48 days respectively at 14.5 °C). Final treatment formulations are listed in Table 13. Because results were inconclusive during the in vivo implantation study, the TOCNF hydrogels were washed for 2-, 4-, and 6-days during production with DPBS. The various wash times were chosen to compare in vivo effects of pH changes from the citric acid crosslinking. For this study, an additional formulation consisted of TOCNF hydrogels washed for 4 days in DPBS and then dehydrated to examine in vivo effects of stiffening the TOCNF hydrogel by removing water for ease in delivery (Table 13). Sampling was performed at 300 and 600- degree days (14.5 °C) post- implantation by euthanizing fish with a lethal dose of MS-222 supplemented with sodium bicarbonate.

[0199] Table 13 – Treatment groups used to implant TOCNF hydrogels for biocompatibility study Treatment Formulation Final No. of fish No. of replicate Total No. pH per tank tanks Fish 1 TOCNF Hydrogel 2-day wash 6.8 8 2 16 2 TOCNF Hydrogel 4-day wash 7.0 8 2 16 3 TOCNF Hydrogel 6-day wash 7.2 8 2 16 4 TOCNF Hydrogel 4 day dehydrated 7.0 8 2 16 5 Bacterin + DPBS 7.25 8 2 16 6 Sentinel Unvaccinated - 8 2 16

[0200] Observed biometrics

[0201] Fish were monitored twice daily over the study for feeding vigor. Biometric data of weights (g) and fork length (mm) were collected for calculating Fulton’s condition factor (K) at the 300 and 600- degree days to compare impact on growth.

[0202] External and Internal Gross Examination

[0203] Gross necropsy was performed on all fish to assess adverse reactions to uptake of and dispersion of TOCNF hydrogels. Therefore, complete external and internal gross examinations were performed at 300 and 600-degree days post-implantation. External examination evaluated incision healing and presence of external lesions including protruding proliferative tissue. Internal visceral and peritoneal adhesions, melanization, fibrosis, granuloma formation, and hydrogel residue were evaluated and scored based on a modified Speilberg scale (Table 14). Histology tissues were collected and analyzed.

[0204] Histopathology

[0205] At the conclusion of the in vivo biocompatibility study (600-degree days post-implantation), tissue samples were collected from 4 fish per treatment group per duplicate tank for a total of 40 samples and fixed into 35 mLs of 10% neutral buffered formalin (Fisher Scientific). Tissues included approximately 1 cm2of body wall at the implantation site, the pyloric caeca, liver, a digestive tract, and spleen. Tissues were submitted to the New Hampshire Veterinary Diagnostic Laboratory (NHVDL) for histopathologic processing and evaluation. Tissues were routinely processed, embedded in paraffin, sectioned at 5-mm thickness, mounted on charged slides, and stained with hematoxylin and eosin. Tissue samples were blinded to treatment groups and scored on the appearance of inflammation, fibrosis, and granulation tissue according to the rubric in Table 14, and the presence or absence of hydrogel was noted.

[0206] Table 14 – Scoring rubric for pathology observations of the body wall and coelom in relation to gross Speilberg scores Speilberg Scores Present study Score Visual appearance of Severity of damage to Body wall Coelom l d d

[0207] Examining Acute Reactions to TEMPO-Oxidized CNF Hydrogel in vivo

[0208] Alongside the in vivo biocompatibility study, a total of twenty Atlantic salmon parr were distributed in an additional third identical rack system for examining possible acute reactions to the TEMPO-oxidized CNF Hydrogel. Five fish held in each of 4 tanks were anesthetized in 100 mgL-1MS-222 buffered with sodium bicarbonate and elastomer tagged with an identifying color. A 3.0-5.0 mm incision was made with a scalpel slightly posterior to the pelvic fin and perpendicular to the ventral surface on the left side of five fish per duplicate tank (10 fish total). A TOCNF hydrogel vaccine formulation containing the Vibrio anguillarum bacterin and washed for 4 days in DPBS post-production was intraperitoneally implanted using the modified PIT tagging device previously described. DPBS + the Vibrio anguillarumbacterin was injected into the incision using a 1 mL syringe with 28-gauge needle and served as a negative control in five fish per duplicate tank (10 fish total) to compare acute gross pathology. Following implantation, fish were immediately returned to the respective tanks for recovery. Two fish per treatment per tank (4 fish per treatment) at each timepoint of 0, 12-, 24-, 48-, and 72-hours post-implantation (8 fish per timepoint) were sampled by euthanizing fish with a lethal dose of MS-222 supplemented with sodium bicarbonate. To verify all fish had a healthy body condition (K ≥ 1.0), sampling was performed for collecting biometric data of weights (g) and fork length (cm). Gross necropsy was performed on all Atlantic salmon to document wound and TOCNF hydrogel associated acute reactions. These were scored using the rubric in Table 15 at each time point post-implantation.

[0209] Table 15 – Rubric for scoring acute gross reactions to implanting TOCNF hydrogels Score Gross Observations 0 No significant findings 1 Hemorrhage 2 Hemorrhage, early coagulation, and swelling 3 Hemorrhage, firm coagulation, and swelling 4 Hemorrhage, coagulation, fibrinous adhesions and / or swelling 5 Hemorrhage, coagulation, firm fibrinous adhesions, and / or swelling

[0210] Calculations and Statistical Analysis

[0211] Sample sizes for the in vivo trials were determined by power analysis to detect differences in the modified Speilberg scale of tissue reaction in Atlantic salmon between the different treatment groups using G Power Version 3.1 software. Degree day was calculated by multiplying the mean water temperature during the course of the study by the number of study days (DD = ((T0 + T1 +…) / no. of days) x no. of days). Fulton’s condition factor (K) was calculated using 100WL-3where W is body weight (g) and L is fork length (cm), so that differences in condition of the fish vaccinated with the different formulations could be compared. Prevalence of adverse reactions (%) ranging from mild to severe as observed using the modified Spielberg scoring growth data were analyzed at each time point by treatment.

[0212] Differences in condition factor, Speilberg scoring, and histology scores were first examined for normality using a Shapiro-Wilk test, then an ordinary two-way ANOVA was used to determine statistical differences with a p < 0.05 considered statistically significant. Where p < 0.05, a pairwise comparison using Tukey’s post-hoc analysis was performed. All statistical analyses were performed using GraphPad Prism 9.5.1. Results were expressed as mean ± standard error of the mean (SEM) unlessotherwise noted with p < 0.05 considered statistically significant.

[0213] Results

[0214] TEMPO-Oxidized CNF hydrogel implantation in Atlantic salmon

[0215] Implantation with forceps

[0216] Initially, low retention of hydrogel in the intraperitoneal cavity was observed macroscopically when implanting with forceps. Five out of the 30 TEMPO CNF hydrogels implanted into Atlantic salmon with forceps were recovered in necropsy for an initial retrieval rate of 16.67%. Additionally, 34 out of the 35 fish implanted with forceps (97.14%) had unhealed incisions that were open and / or visible at 600- degree days post-implantation (FIG.18A). These results demonstrated the desirability of a better delivery method.

[0217] In vivo Biocompatibility Study

[0218] Observations and Biometrics

[0219] After development of the modified PIT tag implantation device, 32 out of 32 TEMPO CNF hydrogels (100%) implanted into Atlantic salmon were recovered at 600- degree days post-implantation for a significantly increased retrieval rate of 83.33% when compared to the 16.67% of hydrogels retrieved after forceps implantation. Also, 100% of the incisions were closed with only 37.5% of the total incisions remaining visible at the time of examination after method optimization (FIG.18B). However, although improved healing was attained at the delivery site with the implantation device, visible lesions on or near the incision sites developed in a large proportion (58.33%) of the fish vaccinated with the TOCNF hydrogels (FIG.18B).

[0220] No mortalities were observed during the course of this in vivo study. Fish maintained vigorous feeding and normal behavior throughout the study. An ordinary two-way ANOVA was performed to analyze the effect of TOCNF hydrogel formulation and sampling time point on Fulton’s condition factor (K) in the Atlantic salmon parr intraperitoneally implanted. Analysis revealed there was not a statistically significant interaction between the effects of TOCNF hydrogel formulation and time point (F (4, 70) = 0.2, p = 0.8969) on K. Simple main effects analysis showed the TOCNF hydrogel formulations did not have a statistically significant effect on K (F (4, 70) = 0.5, p = 0.7058) and time point did not have a statistically significant effect on K (F (1, 70) = 0.0, p = 0.8856). It is important to note mean K was greater than 1.0 across all formulations at 300- degree days (KM300= 1.2, SD = 0.02151) and at 600- degree days (KM600= 1.203, SD = 0.02598) including in the sentinel fish (KMS= 1.212, SD = 0.0146), indicating fish maintained healthy body condition as it relates to growth throughout the study (FIG.19).

[0221] External gross pathology

[0222] External examination during gross necropsy evaluated incision healing and presence of external lesions including protruding proliferative tissue at both 300 and 600 degree-days post-implantation.During the biocompatibility study, use of the implantation device improved upon hydrogel retention (100%) as well as visible wound closing by 600- degree days post-implantation (100%), however many TOCNF hydrogel incisions remained visible by the end of the study showing incomplete wound healing (FIG.20).

[0223] Analysis revealed there was not a statistically significant interaction between the effects of TOCNF hydrogel formulation and time point (F (4, 10) = 1.472, p = 0.2817) on incision healing using an ordinary two-way ANOVA. Simple main effects analysis showed the number of wash days and dehydration of the TOCNF hydrogel formulations did not have a statistically significant effect on incision healing (F (1, 10) = 3.556, p = 0.0887). By 600- degree days, significantly more incisions had healed compared to 300- degree days. (F (4, 10) = 11.09, p = 0.0011). Pairwise comparison using Tukey's post- hoc analysis further demonstrated significantly less incisional healing in the 4- day wash TOCNF compared to DPBS + Bacterin at 300-degree days post-implantation (p = 0.0346) (FIG.21, c).

[0224] Despite the significant improvement in incisional healing by 600- degree days post- implantation, a large number of fish developed external lesions that were not associated with the incision (FIG.20). Analysis revealed there was a statistically significant interaction between the TOCNF hydrogel formulation and time point (F (4, 4) = 13.8, p = 0.0130) on lesion formation using an ordinary two-way ANOVA. Simple main effects analysis showed the number of wash days and dehydration of the TOCNF hydrogel formulations did not have a statistically significant effect on lesion development (F (4, 4) = 3.18, p = 0.1439) while at 600- degree days there were statistically more lesions (F (1, 1) = 81.0, p = 0.0704). Pairwise comparison by Tukey’s post-hoc analysis demonstrates there was a statistically higher prevalence of lesions in the TOCNF hydrogels that were washed for 4- and 6- days at 600- degree days compared to at 300- degree days (p = 0.0384). Additionally, there were significantly more lesions prevalent in the 4- and 6- day washes compared to the DPBS + Bacterin control and the 2- day wash (p = 0.0114) at 600- degree days (FIG.21, b).

[0225] In addition to the high prevalence of external lesions, many fish developed proliferative external masses (FIG.20). Analysis revealed there was not a statistically significant interaction between the TOCNF hydrogel formulation and time point (F (4, 10) = 0.0, p = 0.9970) on external mass presence using an ordinary two-way ANOVA. Simple main effects analysis showed the wash day and dehydration of the TOCNF hydrogel formulations did not have a statistically significant effect on mass development (F (4, 10) = 0.6, p = 0.6123) nor did time point have a statistically significant effect on mass development (F (1, 10) = 1.1, p = 0.2999) (FIG.21, c).

[0226] To summarize, examination of the external pathology during gross necropsy showed that time had healed more incisions by 600 compared to 300 degree-days post implantation. However, a large number of fish developed external lesions near the location of the hydrogel anterior to the incision site compared to sham-injected negative controls. More of these external lesions developed by 600 degree-dayscompared to 300 degree-days post-implantation suggesting that a chronic response increased over time associated with the presence of the hydrogel. Further, many fish developed proliferative external masses with the lesions in the same region of the hydrogel location compared to sham-injected negative controls at 600 degree- days post-implantation. These external results necessitated internal investigation.

[0227] Internal Gross Pathology

[0228] Internal gross pathology results using the Speilberg rubric during the in vivo biocompatibility study are presented here. An ordinary two-way ANOVA was performed to analyze the effect of TOCNF hydrogel formulation and sampling time point on the mean Speilberg score of abdominal adhesions and visceral melanization in the Atlantic salmon parr intraperitoneally implanted with TOCNF hydrogel formulations at 300- degree and 600- degree days post-implantation. Simple main effects analysis showed the TOCNF hydrogels had statistically significant higher abdominal adhesions (F (7, 347) = 62, p < 0.0001) compared to the PBS + Bacterin group at both 300- and 600- degree days post-implantation. (FIG.22, a.) Simple main effects analysis showed the formulation of the TOCNF hydrogel did not have a statistically significant effect on visceral melanization (F (4, 66) = 1.7, p = 0.1486) (FIG.22, b). Peritoneal melanization was not observed during the course of this study.

[0229] To further assess the significance of abdominal adhesion scores in TOCNF formulations compared to the DPBS + Bacterin group, distribution of Speilberg scores (%) are presented at 300-degree days (FIG.23, a) and 600- degree days (FIG.23, b) post-implantation.

[0230] More succinctly, when examined internally, fish implanted with the TOCNF hydrogels generally displayed firm adhesions connecting some or several of the organs. In some cases, the viscera were firmly attached to the peritoneum and a tissue film was observed covering areas of the organs and surrounding the hydrogel. TOCNF hydrogels presented with greater abdominal adhesions compared to the sham-injected negative control group at both 300 and 600 degree-days post-implantation, indicating adhesions developed relatively rapidly in fish implanted with the TOCNF hydrogels. Significant localized muscle and / or organ damage was also observed in some fish in the area of the hydrogel. Minimal melanization of the viscera was seen compared to sham-injected negative controls at 600 degree- days post- implantation with no significant differences between groups nor time. Again, peritoneal melanization was not observed during the course of this study.

[0231] Histopathology

[0232] Pathology changes associated within the body wall were described, scored, and are presented by treatment group in Table 16. In summary, at 600-degree days post-treatment, only samples collected from the sham-injected DPBS + Bacterin negative control group showed normal structure with no lesions within the body wall (Score 0; FIG.24A) and minimal, focal fibrosis and occasional formation of small granulomas as a result of wound healing (Score 1; FIG.24B). Within the body wall, mild adhesions and / orfibrosis with or without mild / small granulomas were observed in both the DPBS + Bacterin treated fish and Atlantic salmon intraperitoneally implanted with TOCNF hydrogel formulations. In addition to histopathological changes associated with the TOCNF hydrogels, minimal to mild focal fibrosis was observed along a tract from the subcutis to the peritoneum with few small granulomas in five fish (3 implanted with 4 day dehydrated TOCNF hydrogel; 1 implanted with 2 day wash TOCNF hydrogel; 1 DPBS + Bacterin group) associated with implantation / injection through the body wall (Score 2; FIG.24C). In the presence of implanted TOCNF hydrogels, some tissue samples displayed moderate adhesions and focally extensive fibrosis with or without granuloma formation. (Score 3; FIG.24D). Moderate fibrosis with small granulomas and / or intramuscular inflammation with granuloma formation surrounding the hydrogel were observed in response to the presence of the TOCNF hydrogel (Score 4; FIG.24E). Further, marked focally extensive and transmural intramuscular inflammation and / or extensive fibrosis was observed with or without granuloma formation surrounding the TOCNF hydrogel (Score 5; FIG.24F). Additionally, marked transmural inflammation and / or fibrosis with granulomas surrounding the TOCNF hydrogel was observed in Atlantic salmon intraperitoneally implanted with TOCNF hydrogels (Score 6; FIG.24G). Overall, no marked differences were delineated between TOCNF hydrogel formulations microscopically.

[0233] Pathology changes associated within the coelom were described, scored, and are presented by treatment group in Table 16. In summary, only Atlantic salmon vaccinated with the DPBS + Bacterin negative control group had no lesions (Score 0; FIG.25A) or minimal adhesions in response to wound healing within the coelom. Minimal visceral adhesions were observed in the coelom of five Atlantic salmon implanted with the TOCNF hydrogels as wound healing response (Score 1; FIG.25B). Mild focal mesenteric fibrosis, few visceral adhesions, and few granulomas around hydrogel in the presence of TOCNF hydrogels were observed within the coelom (Score 2; FIG.25C.). Moderate focal nodular fibroplasia with granuloma formation surrounding the hydrogel with or without mild multifocal visceral adhesions, and fibrosis was observed in the presence of TOCNF hydrogels (Score 3; FIG.25D). Moderate, focally extensive nodular fibrosis was observed in the mesentery, liver, and / or pyloric caeca with dense granuloma formation surrounding the hydrogel, with or without mild adhesions in response to TOCNF hydrogel implantation. Further, adhesion and granuloma involvement with the splenic capsule and intestine along with hepatic necrosis and fibrosis associated with the TOCNF hydrogel was observed (Score 4; FIG. 25E). Multifocal necrosis and granuloma formation in the liver with and without necrosis, granuloma formation surrounding the hydrogel, extensive perihepatic fibrosis, and / or marked, focally extensive nodular fibrosis and inflammation extending along the splenic capsule with moderate visceral adhesions in response to the presence of the TOCNF hydrogel was observed (Score 5; FIG.25F). Focally extensive granuloma formation at the pyloric caeca, extensive fibrosis and inflammation in the liver, a large numberof granulomas, fibrosis, and inflammation throughout the mesentery with melanization and diffuse adhesions was also observed in response to the presence of the TOCNF hydrogels (Score 6; FIG.25G). Overall, no marked differences were delineated between TOCNF hydrogel formulations microscopically.

[0234] Table 16 – Observational notes of histopathology Fish # Slides Formulation Hydrogel Body wall Body wall Coelom Coelom present description score description scoreFish # Slides Formulation Hydrogel Body wall Body wall Coelom Coelom present description score description scoreFish # Slides Formulation Hydrogel Body wall Body wall Coelom Coelom present description score description scoreFish # Slides Formulation Hydrogel Body wall Body wall Coelom Coelom present description score description scoreFish # Slides Formulation Hydrogel Body wall Body wall Coelom Coelom present description score description scoreFish # Slides Formulation Hydrogel Body wall Body wall Coelom Coelom present description score description scoreFish # Slides Formulation Hydrogel Body wall Body wall Coelom Coelom present description score description scoreFish # Slides Formulation Hydrogel Body wall Body wall Coelom Coelom present description score description scoreFish # Slides Formulation Hydrogel Body wall Body wall Coelom Coelom present description score description scoreFish # Slides Formulation Hydrogel Body wall Body wall Coelom Coelom present description score description score

[0235] Histopathology scores of both the body wall and the coelom were analyzed by an ordinary two-way ANOVA to determine the effect of TOCNF hydrogel formulation and tank replicate on body wall pathology score and coelom pathology score. Analysis revealed that there was not a statistically significant interaction between the effects of TOCNF hydrogel formulation and replicate on mean body wall pathology score (F (4, 30) = 2.032, p = 0.1151). Simple main effects analysis showed that there was not a significant tank effect for the mean body wall pathology scores (F (1, 30) = 2.512, p = 0.1235) while TOCNF hydrogel formulation did have a statistically significant effect on mean body wall pathology score (F (4, 30) = 17.12, p < 0.0001). Pairwise comparison using Tukey's post-hoc analysis demonstrated significant differences in coelom scores between 2 day wash (p < 0.0001), 4 day wash (p < 0.0001), 6 day wash (p < 0.0001), and 4 day dehydrated (p < 0.0001) TOCNF hydrogels to the sham-injected negative control (FIG.26, a).

[0236] Ordinary two-way ANOVA analysis also revealed there was not a statistically significantinteraction between the effects of TOCNF hydrogel formulation and replicate tanks on mean coelom pathology score (F (4, 29) = 0.4, p = 0.7709). Simple main effects analysis showed there was not a significant tank effect for the mean coelom pathology scores (F (1, 29) = 0.0, p = 0.0.9695) while TOCNF hydrogel formulation did have a statistically significant effect on mean coelom pathology score (F (4, 29) = 5.1, p = 0.0030). Pairwise comparison using Tukey's post-hoc analysis demonstrated significant differences between 2 day wash (p = 0.0043) and 6 day wash (p = 0.0044) TOCNF hydrogels to the DPBS + Bacterin control (FIG.26, b). The distribution of body wall pathology scores (%) (FIG.27, a) and the distribution of coelom pathology scores (%) (FIG.27, b) observed are presented in FIG.27 by treatment group.

[0237] Acute reactions to TEMPO-oxidized CNF hydrogel in vivo

[0238] Assessment of acute gross pathology at 0, 12-, 24-, 48-, and 72-hours post-implantation displayed the immediate hemorrhaging of blood to the wound and surrounding the implant, coagulation of blood forming clots, swelling, and the rapid development of firm fibrinous adhesions and nodule formation in Atlantic salmon implanted with dehydrated TOCNF hydrogels compared to DPBS + Bacterin only as a negative control by 72 hours post-implantation (FIG.28).

[0239] Discussion

[0240] This example describes the development of a food safe and stable citric acid crosslinked TEMPO-oxidized CNF (TOCNF) hydrogel as a vaccine adjuvant for commercial use in Atlantic salmon. Properly assessing the feasibility of the TOCNF hydrogel for aquaculture applications begins with analyzing the in vivo biocompatibility of the biopolymer hydrogel after in vivo intraperitoneal implantation. A method for the implantation and subsequent observed foreign body response (FBR) modulated by the Atlantic salmon to our 3D biopolymer hydrogel are described.

[0241] The FBR was first documented in 1974 and has since been well-described as a highly conserved immune response. Despite the FBR being highly conserved, it has been described only rarely in teleost species in response to biomaterials, more specifically in the Common Carp (Cyprinus carpio) to PLGA microparticles and in a zebrafish (Danio rerio) model to VicrylTMimplanted surgical sutures. Historically, early developments of intraperitoneally implanted synthetic transmitters were observed to be expelled through the body wall in several fish species including in channel catfish, Ictalurus punctatus, rainbow trout, Oncorhynchus mykiss, and Atlantic salmon, (Salmo salar L.) indicative of an FBR. The documented pathological response observed and quantified in the results in this example describes macroscopically and microscopically the innate immune reaction to the cross-linked biopolymer hydrogel to inform on biopolymer vaccine development in aquaculture. The specific biological mechanisms involved in activating the FBR in Atlantic salmon in response to these materials is unknown. The FBR has been defined previously in literature as a progression of stages being injury, blood-biomaterial interaction at the interface, inflammation and adhesions, granulation of tissue, and ultimately the formation of a fibrouscapsule, all of which the macroscopic and microscopic observations presented here more than sufficiently corroborate.

[0242] TEMPO-oxidized CNF was chosen as the target nanomaterial in this example to develop a TOCNF hydrogel formulated with a Vibrio anguillarum bacterin for investigation. This global pathogen was used to produce a whole cell inactivated bacterin acting as the antigen due to being well studied, characterized, and widely used in commercial vaccines for Atlantic salmon and other aquacultured species. The bacterin was combined with TEMPO-oxidized cellulose nanofibers and crosslinked in the food safe crosslinking agent citric acid before being returned to a neutral pH through DI water and DPBS washes. This resulted in a stable hydrogel.

[0243] After hydrogel development, the TOCNF hydrogels were implanted using a modified PITT device. This method allowed for 100% retained TOCNF hydrogels in Atlantic salmon by 600- degree days. Macroscopically, implantation demonstrated visible wound healing with histopathology displaying a small tract of mild focal fibrosis from the subcutis to the peritoneum with small granulomas in five fish implanted with the hydrogels. Overall, this delivery method was feasible to implant the hydrogel within the same location commonly used for PIT tagging and vaccination in the peritoneal cavity. Noteworthy, this site is located in the vicinity of vital organs such as the pyloric caeca and others known for high immunological cell response. Adverse reactions may be reduced if the site of injected vaccination were directed posteriorly towards the pectoral fins and thus away from vital organs. Implanting the TOCNF hydrogel more posteriorly away from internal organs may therefore yield a decreased FBR.

[0244] Once the implantation method was developed, in vivo biocompatibility of the TOCNF hydrogels over 600 degree- days was examined in Atlantic salmon. Biometric data was collected to monitor growth. This was accomplished by calculating the coefficient of condition (K). Fulton’s condition factor expresses the relative robustness, or degree of well-being, of a fish. Variations in K are congruent with the degree of nourishment and general health of the fish. Condition values may also fluctuate with fish age, stage of reproductive maturity, and in some species, with sex. The implantation of TOCNF intraperitoneally into Atlantic salmon parr showed no significant impact on Fulton’s condition factor (Kmean ≥ 1.0) compared to baseline nor sham-injected controls such that growth penalties were not detected during the short duration of this study.

[0245] The purpose of testing the hydrogels after various wash days during this in vivo study was to examine if residual crosslinker (i.e., citric acid) contributed to adverse effects. However, the results do not indicate a direct effect from residual citric acid as the FBR was observed regardless of vaccine formulation. Gross external examination demonstrated that incisions took longer to heal with more washes (4- and 6- days) compared to the sham negative control and 2-day wash group. Further, at 600-degree days, significantly more external lesions were present in the groups with more washes (4- and 6- days) comparedto the sham negative control and the 2-day wash group. During implantation with the PITT device, hydrogels with more washes were less stiff and more likely to break apart, making intraperitoneal delivery more challenging. This may have caused increased wound trauma through longer time to implant, increased movement inside the peritoneal cavity during implantation, and / or more force needed to implant the hydrogel. In general, these differences did not convey an increased severity of internal gross pathologies nor histopathology, as internal adverse reactions displayed a similar severity across all TOCNF hydrogel groups regardless of washes / pH. Further, one treatment group used partial dehydration to remove water from the hydrogel to produce a stiffer, more durable formulation that was easy to implant and resulted in high incisional healing achieved by 300-degree days and maintained for the duration of the study. However, the internal gross and microscopic pathologies were scored severely in this group and observed to be equivalent to the other TOCNF hydrogel formulations. It is important to note that stiffness of the TOCNF hydrogels may be tuned for improved delivery and ultimately incisional healing. However, the results indicate that doing so may not be sufficient in thwarting activation of the FBR as all formulations showed evidence of the response.

[0246] During the in vivo biocompatibility study, external and internal gross pathology was assessed in response to hydrogel implantation with the PITT device. Internal pathology was quantified using the Speilberg rubric. External examination detected unhealed incisions with presence of external lesions including protruding proliferative tissue. Internally, visceral and peritoneal adhesions, melanization, fibrosis, granuloma formation, and hydrogel residue in response to the presence of the TOCNF hydrogels warranted further microscopic examination. Histopathology clearly demonstrated a FBR characterized by microscopic visceral proliferative lesions, granulomas, adhesions, and fibrosis surrounding the hydrogels compared to sham-injected negative controls at 600 degree- days post-implantation with no marked differences observed between examined TOCNF hydrogel formulations. This is the first time the FBR has been macroscopically and microscopically described in response to a biopolymer hydrogel in Atlantic salmon, and more specifically to a CNM hydrogel.

[0247] Finally, the acute response to implantation of the TOCNF hydrogels was defined macroscopically over 72 hours. Acutely, gross reactions to implantation of the TOCNF hydrogel compared to the sham injected negative control were observed over the first 72 hours including significant hemorrhaging, coagulation, swelling, and eventual formation of fibrinous adhesions around the wound site and the TOCNF hydrogels. These observations were consistent with early stage FBR documented in literature. In mammals, the process of implanting a biomaterial causes damage and / or irritation to the surrounding tissue, immediately initiating the pooling of blood to the site of injury. Specific surface characteristics of the implant can allow proteins within the blood, primarily albumin and fibrinogen, to become non-specifically adsorbed to the surface of the biomaterial. This initial protein layer in combinationwith blood clotting factors and mast cell activation becomes the backbone for neutrophil and macrophage population. This initial acute inflammatory response leads to a self-sustaining cyclical proliferation of macrophages which recruit further macrophages and ultimately form a layer around the foreign material in attempts to engulf, phagocytose, and break down the implant. During this process, these macrophages simultaneously mediate the inflammatory response through the release of pro-inflammatory tumor necrosis factor alpha (TNF-⍺) and multiple interleukins. If breakdown of the foreign material is successful, the response ends and the tissue slowly recovers. However, if breakdown is unsuccessful, the inflammatory response changes to a fibrotic one through foreign body giant cells and fibroblasts and the foreign material is ultimately encased in a layer of fibrous tissue or granuloma to protect surrounding organs and tissue. This physical reaction was ultimately observed by 600 degree-days post-implantation in Atlantic salmon intraperitoneally implanted with the TOCNF hydrogels through gross examination and histopathology, however specific cellular responses were not quantified.

[0248] Again, it is well known the FBR is activated by wound healing in response to trauma and persists chronically as a result of protein and cell adhesion interactions at the surface interface of the biopolymer. To avoid this chronic inflammatory immune response, a biopolymer hydrogel could have ultra-low-fouling surface properties to prevent initial adhesion of fibrinogen, albumin, and other proteins present in high concentrations during blood coagulation and wound healing; these proteins bind to the surface of the biomaterial and serve as a driver for inflammation, macrophage, and other immune cell proliferation, and the eventual development of fibrosis, granulomas, and other adverse effects. An example of an implantable (synthetic) material that successfully evades the foreign body response in Atlantic salmon is the passive integrated transponder tags coated with a biocompatible glass surface. Antifouling surfaces such as this are biocompatible because they resist nonspecific interactions with biomolecules such as proteins, cells, and others.

[0249] From these results combined with strong documentation of the FBR in previous literature, it is concluded activation of this innate inflammatory response occurred as a combined result of wound healing in response to trauma from implantation (promoting blood pooling and coagulation to the site of injury observed during acute gross internal examination) and interactions of the TOCNF hydrogel at the surface interface allowing for inflammation, protein, and immune cell adhesion, proliferation, and ultimate observed development of granulomas, adhesions, and proliferative lesions (macroscopic and microscopic internal pathologies). Biocompatibility through material surface properties can be achieved using hydrophilic materials, with charged or non-ionic moieties capable of strong non-binding interactions with water at the interface to generate a surface hydration layer to reduce nonspecific interactions with other molecules. This action is responsible for maintaining anti-fouling characteristics of the material. Cellulose nanomaterial hydrogels have demonstrated these excellent antifouling and anti-microbial properties formultiple applications, however biocompatibility is additionally dependent on other surface properties such as surface chemistry, functional groups, surface topography and roughness, size and shape of the biomaterial, purity, source of cellulose material, and preparation methods. Furthermore, evidence indicates that slow degradation of CNF scaffolding in combination with wound healing from implantation can elicit the foreign body response.

[0250] Conclusion

[0251] The highly conserved inflammatory innate immune response known as the foreign body response has been well characterized and documented in multiple species. This response occurs after wound trauma and the presence of foreign materials as an exceedingly efficacious protective defense mechanism. Strong macroscopic and microscopic evidence is presented here demonstrating that the FBR occurs in Atlantic salmon in response to the intraperitoneal implantation of a 3D biopolymer hydrogel produced from TEMPO-oxidized CNF crosslinked with citric acid as a vaccine adjuvant. This is the first documentation of this response occurring in Atlantic salmon to 3D biopolymer hydrogels, more specifically to CNM. Cellulose nanomaterials (CNM) are an abundant renewable biomaterial with unique properties including high surface area, surface tunability, biocompatibility, biodegradability, low toxicity, renewability, cost-effectiveness, shelf stability, and mass producibility. The surface properties of CNM enable chemical conjugation and electrostatic absorption for enzyme / protein immobilization resulting in high enzyme / protein loading and excellent stability. These properties make CNM a highly appealing material for vaccine development in aquaculture. From this example, it is apparent that an injectable vaccine system would reduce the chronic inflammatory responses initiated by wound healing. An injectable TEMPO CNF hydrogel with shear thinning characteristics would reduce the invasiveness of the physically crosslinked hydrogel and would be able to be injected through a small-bore needle and reduce the stiffness of the hydrogel once injected in the peritoneal cavity. Injectable shear-thinning hydrogel chemistry may therefore be an avenue to achieve immune modulation in a biocompatible manner for long-term protection against disease outbreaks in Atlantic salmon.

[0252] Example IV - Citric acid crosslinked TEMPO CNF hydrogel

[0253] pH and washing

[0254] Physically crosslinked hydrogels offer a wide range of benefits for implementation into drug diffusion devices as they offer non-toxic chemical usage and have reversible reaction that can promote the breakdown of the hydrogel over time. TEMPO CNF is able to be physically crosslinked through the use of citric acid. Citric acid works as a physical crosslinker as it lowers the pH of the solution, creating H+ ions within the solution. These free H+ ions are able to protonate the TEMPO CNF fiber’s COO- groups which forms strong hydrogen bonds between fiber chains creating the hydrogel matrix as shown in FIG.29. Citric acid has low toxicity, is inexpensive, and offers an uncomplicated procedure. Due to the low pHrequired for the physical crosslinking of the TEMPO CNF, sufficient wash steps are important in order to limit the adverse effects that would be caused to the fish due to the large concentration of citric acid.

[0255] The materials that were used for the synthesis of the TEMPO CNF citric acid crosslinked aquaculture vaccine hydrogels were TEMPO CNF, citric acid, and Vibrio anguillarum. TEMPO CNF was obtained from the Process Development Center at the University of Maine. The TEMPO CNF was received at a 1.1 wt% slurry with fiber lengths of approximately 1μm and diameter of 20 nm. Citric acid was 99.5% and was acquired from Sigma Aldrich. Inactivated Vibrio anguillarum bacteria had a concentration of 1*109CFU / mL and was cultured in a trypticase soy broth with 1.5% NaCl and was inactivated with 0.5% formalin.

[0256] For crosslinking the TEMPO CNF, a crosslinking solution of 10 wt% citric acid was utilized to allow for a loosely bound hydrogel matrix which will promote drug diffusion through the swelling of the hydrogel matrix. The TEMPO CNF wt% was chosen to be at 1.7 wt% as this allowed for hydrogel retention after sufficient wash days in comparison to lower wt% while still having a low enough wt% to allow for adequate diffusion. Creation of the vaccination utilized a 1:1 ration of TEMPO CNF to bacterin which produced a hydrogel with 2.6*108 CFU. The concentration of the bacterin after sufficient filtration and wash steps was approximately 1.0*108CFU which is equivalent to the current Vibrio anguillarum dosage given in aquaculture systems. In order to increase the wt% of the hydrogel vaccination to the desired 1.7 wt%, vacuum filtration with a 0.45 nm nylon membrane was used as the pore size and hydrophobicity of the nylon membrane allowed for retention of the bacterin endotoxins.

[0257] When forming the 1.7 wt% TEMPO CNF bacterin solution for crosslinking, a 1 mL syringe with a 0.5 mm diameter bore was filled and extruded until approximately 1 mm length of gel was released before being sliced and dropped into the 10 wt% citric acid solution. The TEMPO CNF bacterin solution was left to crosslink for 24 hours in order to allow for improved mechanical properties. After submersion for 24 hours in citric acid solution, the hydrogels had a pH of approximately 1.5. This pH is too low for immediate injection into aquatic life as the low pH can cause adverse effects and significantly damage the injection site as well as the internal organs. According to the United States Environmental Protection Agency (EPA), the minimum pH of water for aquatic life is above 6.5 pH units. If the hydrogels were to be washed in deionized (DI) water, over time the pH of the hydrogel should increase to a safe pH for aquatic life.

[0258] To assess the washing of the TEMPO CNF hydrogels after citric acid crosslinking, hydrogels with 1.1, 1.3, 1.5 and 1.7 wt% were formed and then washed in DI water. The pH of the hydrogel DI water solution was monitored every 24 hours. The supernate pH of the washed hydrogels did not change significantly depending upon the wt% of the tested hydrogels. The overall shift in pH within the hydrogels remained constant over the first 120 hours at a rate of approximately 0.02 pH hourly. After the 120 hours ofwashing, the change in supernate pH decreased significantly to just above zero. This caused the pH of the hydrogels to plateau at roughly 4.5 units as shown in FIG.30. The washed pH of the hydrogels was still too low for injection into aquatic life so it had to be increased. If the hydrogels were washed in a buffering solution such as PBS, the pH would be able to increase further.

[0259] To assess the effect the PBS solution has on the pH of the hydrogels, the 1.7 wt% hydrogels were submerged into a PBS solution after reaching the plateau that was observed in the DI water washing (240 hours). The pH of the supernate was measured every 24 hours similar to that of the DI water washing process. The PBS rapidly increased the pH past 6.5 units after 48 hours or two wash cycles as shown in FIG.31.

[0260] While washing of the hydrogel with DI water and PBS increases the pH within a safe range for injection into aquaculture, the washing parameters are important in order to reduce the diffusion of bacterin during the washing. The washing method ideal for the physically crosslinked hydrogel was determined to be washing in DI water for 2 to 6 days followed by a PBS wash day prior to being implanted into aquatic life to minimize adverse effects caused by pH.

[0261] Drying and rehydration of hydrogels

[0262] While the 1.7 wt% hydrogels had improved handling characteristics compared to lower wt% hydrogels, delivery of the hydrogel through a small incision into a salmon peritoneum cavity proved to be too challenging resulting in the hydrogels breaking. Although some hydrogels were able to be pushed into the peritoneum cavity, no hydrogels were found in any of the tested salmon. In addition to the loss of hydrogels during the preliminary tests, it was noted that the incision had not healed properly. From this test, it was determined that the hydrogels should be sized down and strengthened in order to be optimally implanted.

[0263] The method in which the hydrogels were sized down and strengthened was through a drying process. To test varying temperatures and their influence on the rate of drying for the hydrogels, a laminar flow dehydrator with an adjustable temperature was utilized. The hydrogels were rotated 180° between measurements to promote even drying between all samples. The varying temperatures of drying that were tested was 20 °C, 29.5 °C, 46 °C, and 63 °C, where the 20 °C was tested at room air. The use of the laminar flow dehydrator decreased the overall percent mass of the hydrogel significantly compared to the room air drying whereas increasing the temperature of the dehydrator had a lesser effect on the percent mass. This trend is confirmed through comparing the drying rate of the hydrogels over time as shown in FIG.32. From this figure it is shown that the ambient temperature (20 °C) had little effect on the overall drying rate of the hydrogel whereas the 63 °C laminar flow had the largest initial drying rate at 6.5 mg / min followed by 46 °C laminar flow. With the decreasing moisture content of the hydrogels, it was noted across all drying temperatures that the drying rate significantly reduces as the hydrogel approaches 0% hydration. Whencomparing the time required to dry the hydrogels, the 63 °C dried hydrogels were the fastest to a fully dehydrated hydrogel followed by the 46 °C and so on as shown in FIG.33.

[0264] When considering the ideal temperature to be used for drying of the hydrogels, the bacterin should be considered as exposing it to high temperatures will diminish the overall antigen response. The final drying temperature chosen for the application of the physically crosslinked hydrogel was 46 °C as it was within the upper temperature range in which the bacterin can withstand before antigen reduction occurs.

[0265] Mechanical properties

[0266] For testing the mechanical properties of the 1.7 wt% TEMPO CNF hydrogel, a larger hydrogel was utilized due to the mechanical limitations of the testing device. The larger hydrogels had a 1 mm diameter and were 1.5 mm in length. The compression strength and modulus of the hydrogels were tested in order to determine the overall strength and stiffness of the hydrogel matrix respectively. The hydrogels were tested at four separate hydration points, 100%, 60%, 40%, and 20% where the percentage correlates to the hydrated mass percent after being dried as shown in FIG.34.

[0267] There is a positive correlation between both the compression strength and modulus throughout the drying process of the hydrogels with the lowest hydrated hydrogels having the best mechanical performance. This trend is seen within FIG.34 as the 100% hydrated hydrogel has an average compression strength and compression modulus of 296.89 kPa and 0.48 kPa, respectively while the 20% hydrated hydrogel has an average compression strength and compression modulus of 572.71 kPa and 1.12 kPa, respectively. The process of drying the hydrogels is seen to increase the overall strength and stiffness allowing for the improved application of the hydrogels during vaccination.

[0268] Mass and volume

[0269] Rehydration of the hydrogels were analyzed after the drying of the hydrogels as they were to be submerged in PBS before the injection process. The hydrogels were dried at 46 °C for 10, 30, 50, 70, and 90 minutes which correlates to 70%, 45%, 20%, 5%, and 3% hydration in relation to its initial mass. After submersion in PBS, the hydrogels were measured over time to assess their hydration as shown in FIG.35.

[0270] Initially, the hydrogels showed a steep increase in mass after being submerged in PBS. This increase in mass was consistent across all drying times until the hour mark where the rehydration mass began to plateau. This plateau remained for 18 weeks until the collection of data was stopped. This indicates that the hydrogels are stable for a long period of time and that the physical crosslinking method is viable for long term vaccinations. Rehydration of the hydrogels should be for at least an hours before being injected to limit adverse effects that could be seen through the hydrogels swelling in the aquaculture. For future testing with the 46 °C dehydrated hydrogels, a drying time of 30 minutes was chosen as this dryingtime as it displayed the greatest increase in percent mass of a dried hydrogel that is below 50% of its initial mass.

[0271] Breakdown of hydrogel

[0272] When the physically crosslinked TEMPO CNF hydrogels were left in a PBS solution for an extended period of time, they appeared to expand and create a cloudy matrix around them. The first instance where this phenomenon was noticed was during the long-term rehydration testing. This cloud formation indicates that the hydrogel was breaking down over time. In order to confirm the breakdown of the hydrogel, the TEMPO CNF hydrogels were stained with calcofluor white. The stained hydrogels were submerged in a PBS solution for eight weeks and were imaged under UV light at the beginning and end of the time frame. The resulting images were analyzed in ImageJ through separating the color channels of the image which created a clear distinction between the cloud matrix and the remaining hydrogel (FIGS.36A- 36B).

[0273] Through separating the image into color channels based upon the concentration of calcofluor white, it is confirmed that the cloud matrix contains TEMPO CNF which indicates that the hydrogel breaks down over long-term exposure in PBS. This breakdown of the TEMPO CNF hydrogel is beneficial as it promotes the long-term diffusion of bacterin as the matrix deteriorates.

[0274] Application of hydrogel adjuvant

[0275] While drying the TEMPO CNF hydrogel with the use of the laminar flow dehydrator decreased the size and increased the strength of the hydrogel, the gels were still too large to be injected through the 26-gauge needle and syringe. Although the application of this hydrogel method could not mimic current administration of vaccination, the process of pit tagging aquatic life uses up to a 6-gauge needle. If the TEMPO CNF hydrogels were dehydrated to 40% of their initial mass, they were able to fit inside of an 8-gauge needle. After inserting the 40% hydrated hydrogel into the 8-gauge needle, a small incision was made into the peritoneal cavity allowing the needle to be inserted into the fish with minimal effort before pushing the gel into the fish. This Injection method for the TEMPO CNF hydrogels was tested in a safety study to determine its effectiveness in relation to the current injection process.

[0276] The safety study used to assess the effectiveness of the hydrogel delivery method consisted of 80 Atlantic salmon with five separate treatment groups. These treatment groups consisted of 1.7 wt% TEMPO CNF bacterin 1:1 ratio hydrogels that were washed 2 days, 4 days, and 6 days and were rehydrated in PBS for 12 hours. The final two groups were a 4 day wash hydrogel that was not rehydrated in PBS and a control group that were injected with PBS and bacterin alone. All treatment groups had a hydration of roughly 40% prior to their injection.

[0277] In order to determine the effectiveness of the TEMPO CNF bacterin hydrogel delivery method, the extent of adverse reactions were analyzed through two separate methods which were Fulton’scondition factor and adhesion score. Fulton’s condition factor is calculated based upon the fish’s weight and length ratio to determine if it is growing at a consistent rate. Adhesion score is a numerical grading system that assesses adhesions, viscera appearance, peritoneum appearance, and residues within the fish. Both the scoring rubrics used for the Fulton’s condition factor and adhesion score is included in the appendix of this thesis. Based upon Fulton’s condition factor the overall growth of the salmon was not affected by the injection method as all treatment groups had a condition factor above 1.0. This confirms that the TEMPO CNF bacterin hydrogel injection method does not significantly affect the growth and development of the salmon. While the growth and development of the salmon were not significantly affected, the delivery method showed significant adhesions when compared to the control PBS and bacterin which caused minimal adhesions as shown in Table 17. All forms of treatment groups that utilized the 8- gauge needle were found to have severe adhesion scores regardless of the time of washing or degree of rehydration.

[0278] Table 17 – Fulton’s condition factor and adhesion score from safety trial [0p treatment groups, there were often cysts that formed around the hydrogel. These cysts appeared to be pushing the hydrogel out of the peritoneum cavity and into the surrounding muscle. While the cysts were present across all treatment groups, they were most present in the TEMPO CNF bacterin 2 day wash and the TEMPO CNF bacterin 4 day dried groups which is attributed to the increased pH and the stiffness of the hydrogel upon injection.

[0280] Summary

[0281] TEMPO CNF physically crosslinked hydrogels offer a wide range of characteristics that are beneficial to drug diffusion as it is nontoxic and undergoes reversible reactions which promotes its breakdown over time. These physically crosslinked hydrogels had tunable sizing and matrix densities thatwere easily altered through the drying and reswelling which is beneficial for tuned drug delivery depending upon the sizing of the desired bacterin or viral pathogen. The mechanical compression strength and modulus showed that this form of hydrogel vaccination can withstand injection and is able to stay within the injected source for a set period of time before the reversible breakdown occurs.

[0282] Through the tested treatment groups, it was shown that the TEMPO CNF hydrogels had no effect on the tested fish’s growth based upon the results of the Fulton’s condition factor. While the hydrogel did not affect the growth of the fish, the adhesion score and presence of cysts within the tested treatment groups showed the method in which the hydrogels were injected (i.e., the 8-gauge needle method) was too invasive, and the stiff hydrogel formulation promoted a foreign body response. If the TEMPO CNF hydrogels were to be able to have shear thinning characteristics, the formulation would be able to be injected similarly to current aquaculture practices with a 26-gauge needle which would be significantly less invasive. As for reducing the foreign body response, a shear thinning hydrogel would be able to take the least invasive shape within the peritoneal cavity, further reducing the adverse reactions that have been previously seen.

[0283] Example V – In vivo immunogenicity in Atlantic salmon

[0284] The effect of an amidated TOCNF vaccine on the immune response of the vaccinated Atlantic salmon was evaluated by analyzing the relative gene expression of IgM, IgD, and IgT levels compared to the immune response to fish vaccinated with a commercial injectable vaccine. The reference gene, β-actin, was used to normalize the expression levels of the target genes. Sentinel unvaccinated fish and DPBS + bacterin vaccinated fish were not included in statistical analysis due to unequal sample sizes. However, FIGS.37A-37C show the relative gene expression of these two groups was similar to the treatment groups analyzed. Ordinary two-way ANOVA analysis results are presented in Table 18. Where p < 0.05, a pairwise comparison using Tukey’s HSD post-hoc analysis was performed where significant differences (*) were observed between the PBS negative sham control (FIG.37A).

[0285] Table 18 - Results of ordinary two-way ANOVA (α = 0.05) of mean relative quantification of gene expression from head kidney in Atlantic salmon vaccinated with test formulations.significant differences between mean relative quantification of gene expression from head kidney in Atlantic salmon vaccinated with test formulations. (p = 0.05).

[0286] Certain embodiments of the compositions and methods disclosed herein are defined in the above examples. It should be understood that these examples, while indicating particular embodiments of the invention, are given by way of illustration only. From the above discussion and these examples, one skilled in the art can ascertain the essential characteristics of this disclosure, and without departing from the spirit and scope thereof, can make various changes and modifications to adapt the compositions and methods described herein to various usages and conditions. Various changes may be made and equivalents may be substituted for elements thereof without departing from the essential scope of the disclosure. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the disclosure without departing from the essential scope thereof.

Claims

CLAIMS What is claimed is:

1. An adjuvant composition comprising a cellulose nanofibril (CNF) hydrogel crosslinked with a cation from a salt, wherein the adjuvant is an injectable solution capable of passing through a 26- gauge needle.

2. The adjuvant composition of claim 1, wherein the CNF hydrogel comprises TEMPO- oxidized cellulose nanofibrils.

3. The adjuvant composition of claim 1, wherein the cation comprises Na+or Ca2+.

4. The adjuvant composition of claim 1, wherein the salt is NaCl or CaCl2.

5. The adjuvant composition of claim 1, wherein the CNF hydrogel is present at a concentration of up to about 3 wt%.

6. A method of preparing a vaccine composition, the method comprising mixing a CNF hydrogel with an antigen and optionally one or more pharmaceutically acceptable diluents, carriers, or adjuvants, wherein the CNF hydrogel is a salt crosslinked TEMPO-oxidized CNF.

7. The method of claim 6, wherein the salt crosslinked TEMPO-oxidized CNF is crosslinked with Na+ions or Ca2+ions from the salt.

8. The method of claim 6, wherein the salt is NaCl or CaCl2.

9. The method of claim 8 wherein the NaCl is at a concentration of up to about 114 mM.

10. The method of claim 8, wherein the CaCl2 is up to a concentration of about 7.14 mM.

11. A method of vaccinating an animal, the method comprising administering to an animal a vaccine composition comprising an antigen and an adjuvant, wherein the adjuvant comprise a salt crosslinked TEMPO-oxidized CNF hydrogel.

12. The method of claim 11, wherein the salt crosslinked TEMPO-oxidized CNF hydrogel iscrosslinked with Na+or Ca2+ions.

13. The method of claim 11, wherein the animal is a fish.

14. The method of claim 11, wherein the animal is a salmon.

15. The method of claim 11, wherein the vaccine composition is an injectable solution capable of passing through a 26-gauge needle.

16. The method of claim 11, wherein the vaccine composition further comprises one or more additional adjuvants, stabilizers, preservatives, surfactants, buffering agents, or culturing substances.

17. A vaccine composition comprising: a salt crosslinked TEMPO-oxidized cellulose nanofibril (CNF) hydrogel; and an antigen or immunogen.

18. The vaccine composition of claim 17, wherein the vaccine composition is an injectable solution capable of passing through a 26-gauge needle.

19. The vaccine composition of claim 17, wherein the CNF hydrogel comprises TEMPO- oxidized CNF crosslinked with the salt.

20. The vaccine composition of claim 17, wherein the salt is NaCl or CaCl2.

21. The vaccine composition of claim 17, further comprising one or more additional adjuvants, stabilizers, preservatives, surfactants, buffering agents, or culturing substances.

22. The vaccine composition of claim 17, comprising bacterin.

23. A method of stimulating a foreign body response (FBR) in an Atlantic salmon, the method comprising administering to an Atlantic salmon a vaccine composition comprising a citric acid crosslinked TEMPO-oxidized CNF hydrogel and an immunogen.

24. A kit for making a vaccine, the kit comprising: a first container housing a salt-crosslinked TEMPO-oxidized CNF hydrogel; anda second container housing an antigen or immunogen.

25. The kit of claim 24, wherein the second container houses bacterin.

26. Use of a salt-crosslinked TEMPO-oxidized cellulose nanofibril (CNF) hydrogel as a vaccine adjuvant.

27. A vaccine composition comprising: an amidated TEMPO-oxidized CNF hydrogel; and an immunogen or antigen; wherein the vaccine composition is an injectable solution.

28. The vaccine composition of claim 27, wherein the vaccine composition has shear- thinning behavior.

Images