WO2026010969A1

WO2026010969A1 - Formulations for stabilizing lactobacillus and bifidobacterium strains

Info

Publication number: WO2026010969A1
Application number: PCT/US2025/036107
Authority: WO
Inventors: Carlo Giovanni Traverso; Kimberley A. BIGGS; Mairead HEAVEY; Christina KARAVASILI; Matt Murphy; Samantha BOZORGZADEH; Isaac J. TUCKER
Original assignee: University of Queensland UQ; Brigham and Womens Hospital Inc; Massachusetts Institute of Technology
Current assignee: University of Queensland UQ; Brigham and Womens Hospital Inc; Massachusetts Institute of Technology
Priority date: 2024-07-03
Filing date: 2025-07-01
Publication date: 2026-01-08
Anticipated expiration: 2027-01-03

Abstract

Formulations for stabilizing Lactobacillus and Bifidobacterium strains are provided, comprising one or more excipient selected from the group consisting of D-(-)-fructose, Bacto Tryptic Soy Broth (BTSB), potassium gluconate, magnesium D-gluconate, sucrose, and/or maltodextrin, and preferably also comprising L-histidine, as are lyophilized cell populations generated using the formulations, and the use of such lyophilized cell populations.

Description

MIT 25723 Formulations for stabilizing Lactobacillus and Bifidobacterium strains Background Various probiotic bacteria have been identified as critical for a healthy neonatal or vaginal microbiome. Specifically, the reduced presence of Bifidobacterium species such as Bifidobacterium infantis in infants has been linked with a higher risk of premature birth complications or allergy disorders, while women with a low abundance of Lactobacillus species such as Lactobacillus crispatus in their vaginal microbiome are at a higher risk of reduced fertility or contracting various vaginal infections. The administration of these strains has been reported to reduce the risk of subsequent comorbidities, however, they are highly sensitive to oxygen and temperature. As a result, it is difficult to maintain their viability throughout the therapeutic formulation, transportation and administration process. Summary In one aspect, the disclosure provides formulations, comprising one or more excipient selected from the group consisting of D-(-)-fructose, Bacto Tryptic Soy Broth (BTSB), potassium gluconate, magnesium D-gluconate, sucrose, and/or maltodextrin. In one embodiment, the one or more excipient comprises potassium gluconate or magnesium D- gluconate. In another embodiment, the formulation further comprises L-histidine. In a specific embodiment, the formulation comprises or consists of potassium gluconate and L- histidine. In various embodiments, the formulation comprises: (a) between about 0.4% (w/v) and about 4% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 3.65% (w/v) L-histidine; or (b) between about 0.4% (w/v) and about 2% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (c) between about 0.4% (w/v) and about 2% (w/v) potassium gluconate and between about 0.73% (w/v) and about 1.825% (w/v) L-histidine; or (d) between about 0.4% (w/v) and about 2% (w/v) potassium gluconate and between about 0.73% (w/v) and about 1.217% (w/v) L-histidine; or (e) between about 0.4% (w/v) and about 1.5% (w/v) potassium gluconate, and between about 0.73% (w/v) and about 1.825% (w/v) L-histidine; or (f) between about 0.4% (w/v) and about 1.333% (w/v) potassium gluconate, and between about 0.73% (w/v) and about 1.825% (w/v) L-histidine; or (g) between about 0.4% (w/v) and about 1.5% (w/v) potassium gluconate, and between about 0.73% (w/v) and about 1.3% (w/v) L-histidine; or (h) between about 0.4% (w/v) and about 1.333% (w/v) potassium gluconate, and between about 0.73% (w/v) and about 1.217% (w/v) L-histidine; or (i) between about 1.3% (w/v) and about 1.6% (w/v) potassium gluconate, and between about 1.2% (w/v) and about 1.3% (w/v) L-histidine; or (j) about 1.3% (w/v) potassium gluconate and about 1.2% (w/v) L-histidine; or (k) about 1.333% (w/v) potassium gluconate and about 1.217% (w/v) L-histidine. In one embodiment, the formulation further comprises a plurality of Lactobacillus cells. In various such embodiments, the formulation comprises: (a) between about 0.2% (w/v) and about 2% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 2% (w/v) L-histidine; or (b) between about 0.2% (w/v) and about 1% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 0.912% (w/v) L-histidine; or (c) between about 0.2% (w/v) and about 1% (w/v) potassium gluconate and between about 0.36% (w/v) and about 0.912% (w/v) L-histidine; or (d) between about 0.2% (w/v) and about 1% (w/v) potassium gluconate and between about 0.36% (w/v) and about 0.608% (w/v) L-histidine; or (e) between about 0.2% (w/v) and about 0.75% (w/v) potassium gluconate, and between about 0.36% (w/v) and about 0.912% (w/v) L-histidine; or (f) between about 0.2% (w/v) and about 0.666% (w/v) potassium gluconate, and between about 0.36% (w/v) and about 0.912% (w/v) L-histidine; or (g) between about 0.2% (w/v) and about 0.75% (w/v) potassium gluconate, and between about 0.36% (w/v) and about 0.65% (w/v) L-histidine; or (h) between about 0.2% (w/v) and about 0.666% (w/v) potassium gluconate, and between about 0.36% (w/v) and about 0.608% (w/v) L-histidine; or (i) between about 0.65% (w/v) and about 0.8 % (w/v) potassium gluconate, and between about 0.6% (w/v) and about 0.65% (w/v) L-histidine; or (j) about 0.65% (w/v) potassium gluconate and about 0.6 % (w/v) L-histidine; or (k) about 0.667% (w/v) potassium gluconate and about 0.609% (w/v) L-histidine. In some embodiments, the Lactobacillus cells are selected from the group consisting of Lactobacillus crispatus, Lactobacillus acidophilus, and Lactobacillus gasseri. In other embodiments, the formulation comprises: (a) between about 0.4% (w/v) and about 4% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 3.65% (w/v) L-histidine; or (b) between about 0.4% (w/v) and about 2% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (c) between about 0.4% (w/v) and about 0.8% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (d) between about 0.35% (w/v) and about 0.45% (w/v) potassium gluconate, and between about 0.3% (w/v) and about 1.3% (w/v) L-histidine; or (e) about 0.4% (w/v) potassium gluconate, and between about 0.3% (w/v) and about 1.3% (w/v) L-histidine; or (f) about 0.4% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 1.217% (w/v) L-histidine; or (g) about 0.4% (w/v) potassium gluconate and about 1.217% (w/v) L-histidine. In one such embodiment, the formulation further comprises a plurality of Lactobacillus cells. In further embodiments, the formulation comprises: (a) between about 0.2% (w/v) and about 2% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 2% (w/v) L-histidine; or (b) between about 0.2% (w/v) and about 1% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 0.917% (w/v) L-histidine; or (c) between about 0.2% (w/v) and about 0.4% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 0.917% (w/v) L-histidine; or (d) between about 0.35% (w/v) and about 0.225% (w/v) potassium gluconate, and between about 0.15% (w/v) and about 0.65% (w/v) L-histidine; or (e) about 0.2% (w/v) potassium gluconate, and between about 0.15% (w/v) and about 1.3% (w/v) L-histidine; or (f) about 0.2% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 0.608% (w/v) L-histidine; or (g) about 0.2% (w/v) potassium gluconate and about 0.608% (w/v) L-histidine. In one embodiment, the Lactobacillus cells comprise Lactobacillus jensenii. In another embodiment, the formulation comprises or consists of magnesium D- gluconate and L-histidine. In various embodiments, the formulation comprises (a) between about 1.02% (w/v) and about 10.2% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 3.650% (w/v) L-histidine; or (b) between about 2.04% (w/v) and about 10.2% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (c) between about 2.04% (w/v) and about 5.1% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (d) between about 5.0% (w/v) and about 5.2% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (e) between about 5.0% (w/v) and about 5.2% (w/v) magnesium D-gluconate, and between about 0.739% (w/v) and about 1.825% (w/v) L-histidine; or (f) about 5.1% (w/v) magnesium D-gluconate and about 0.1.217% (w/v) L- histidine; or (g) between about 2.0% (w/v) and about 2.1% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (h) between about 2.0% (w/v) and about 2.1% (w/v) magnesium D-gluconate, and between about 0.730% (w/v) and about 1.825% (w/v) L-histidine; or (i) between about 2.0% (w/v) and about 2.1% (w/v) magnesium D-gluconate, and between about 1.217% (w/v) and about 1.825% (w/v) L-histidine; or (j) about 2.04% (w/v) magnesium D-gluconate and about 1.217% (w/v) L- histidine. . In one such embodiment, the formulation further comprises a plurality of Bifidobacterium cells, including but not limited to B. infantis, B. bifidum, B. longum and B. breve. In various such embodiments, the formulation comprises: (a) between about 0.51% (w/v) and about 5.10% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 1.825% (w/v) L-histidine; or (b) between about 1.02% (w/v) and about 5.10% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 0.913% (w/v) L-histidine; or (c) between about 1.02% (w/v) and about 2.55% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 0.913% (w/v) L-histidine; or (d) between about 2.5% (w/v) and about 2.6% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 0.913% (w/v) L-histidine; or (e) between about 2.5% (w/v) and about 2.6% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 0.913% (w/v) L-histidine; or (f) about 2.55% (w/v) magnesium D-gluconate and about 0.609% (w/v) L- histidine; or (g) between about 1.0 % (w/v) and about 1.05% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 0.913% (w/v) L-histidine; or (h) between about 1.0% (w/v) and about 1.05% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 0.913% (w/v) L-histidine; or (i) between about 1.0% (w/v) and about 1.05% (w/v) magnesium D-gluconate, and between about 0.609% (w/v) and about 0.913% (w/v) L-histidine; or (j) about 1.02% (w/v) magnesium D-gluconate and about 0.609% (w/v) L- histidine. In various embodiments of all embodiments of the disclosure, the formulation does not comprise any sugars; the excipients are dissolved in water or phosphate buffered saline, or wherein the excipients are dissolved in water; and/or the formulation consists of the recited concentration of excipients, with water making up the remainder of the formulation. In another aspect, the disclosure provides a lyophilized Lactobacillus cell population, comprising a matrix of hydrogen-bonded potassium gluconate and L-histidine in which the Lactobacillus cells are embedded. In various embodiments, the Lactobacillus cells comprise or consist of one or more of L. crispatus, L. acidophilus, L. jensenii, L.rhamnosus, and L.gasseri. In a further aspect, the disclosure provides a lyophilized Bifidobacterium cell population, comprising a matrix of hydrogen-bonded magnesium D-gluconate and L-histidine in which the Bifidobacterium cells are embedded. In various embodiments, the Bifidobacterium cells comprises or consist of one or more of B. infantis, B. bifidum, B. longum and B. breve. The disclosure also provides compositions, comprising the lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population of any embodiment herein, present in or on a delivery agent selected from the group consisting of a tablet, a capsule, a gel, a film, a suspension, a baby bottle (including but not limited to coated on the nipple or inside surface of the bottle), a pacifier, or a medical device (including but not limited to an intravaginal ring or a vaginal applicator). In one embodiment, the delivery agent further comprises boric acid and/or oleic acid. In a further aspect, the disclosure provides methods for improving female fertility or limiting development of vaginal infections, comprising administering to a female in need thereof an amount effective of the lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population, or the composition, of any embodiment, to improve fertility or limit development of vaginal infections. In one aspect, the disclosure provides methods for reducing premature birth complications or allergic disorders, comprising administering to an infant in need thereof an amount effect of the lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population, or the composition, of any embodiment to reduce premature birth complications or allergic disorders in the infant. Description of the Figures Figure 1. Heat map of the viability after L. crispatus was lyophilized with different concentrations of the three best performing primary and secondary excipients and stored for 24 h at 60^oC. The concentrations that resulted in the greatest viability after lyophilization are lettered for each combinatorial formulation (N = 5). Figure 2. The long-term stability of L. crispatus after lyophilization with either the three best performing combinatorial formulations, the representative commercial formulation or no formulation and storage at 60^oC (A) and 37^oC (B) for 6 months. All data points represent mean ± SEM (N = 3). Figure 3. Visual characteristics of L. crispatus lyophilized in Formulation A. (A) Representative image of the product after L. crispatus is lyophilized with Formulation A. (B) Representative scanning electron micrograph of L. crispatus + Formulation A lyophilized product. Figure 4 The excipient concentration analysis between L. crispatus without cells (A) and with cells (B). The best performing formulation is denoted in black and referred to as Formulation ALC. Figure 5. The excipient concentration analysis between L. jensenii without cells (A) and with cells (B). The best performing formulation is denoted in black and referred to as Formulation ALJ. Figure 6. The extremophilic formulation demonstrates promising properties for translation to the clinic. Formulation A is adaptable to various therapeutically relevant Lactobacillus strains. The excipients were able to demonstrate significantly more viability directly after lyophilization and aerobic storage for 24 h at 60^oC. * P < 0.05, ** P < 0.01, *** P < 0.005, **** P < 0.001 by Tukey’s multiple comparisons test (N = 7–20). Figure 7. (A) The viability of L. crispatus when lyophilized in either Formulation A (circle), the representative commercial formulation (square) or no formulation (triangle) and exposed to sequential manufacturing processes. Each data point represents a single biological replicate. The boxed regions represent the upper and lower quartile, and the whiskers represent the upper and lower extremes. *P < 0.05, ** P < 0.01, *** P < 0.005, by Tukey’s multiple comparisons test (N = 7). (B) The viability of L. crispatus when lyophilized in either Formulation A (circle), the representative commercial formulation (square) or no formulation (triangle) and incorporated into different methods of delivery. Each data point represents a single biological replicate. The boxed regions represent the upper and lower quartile, and the whiskers represent the upper and lower extremes. *P < 0.05, ** P < 0.01, *** P < 0.005 by Tukey’s multiple comparisons test = 7–20). (C) The release of L. crispatus from the oleogel-loaded intravaginal ring (circle) into simulated vaginal fluid in comparison to the stability of the same dose suspended in simulated vaginal fluid without the intravaginal ring (square). Each data point represents a single biological replicate, and the bar represents the mean ± SD (N = 5). Figure 8. Heat map of the viability after B. infantis was lyophilized with different concentrations of the four best performing primary and secondary excipients and stored for 24 h at 60^oC. The concentrations that resulted in the greatest viability after lyophilization are lettered for each combinatorial formulation (N = 5). Figure 9. The long-term stability of B. infantis after lyophilization with either the three best performing combinatorial formulations, the representative commercial formulation or no formulation and storage at 60^oC (A) and 37^oC (B) for 6 months. All data points represent mean ± SEM (N = 3). Figure 10. Visual characteristics of B. infantis lyophilized in Formulation B. (A) Representative image of the product after B. infantis is lyophilized with Formulation B. (B) Representative scanning electron micrograph of B. infantis + Formulation B lyophilized product. Figure 11. The excipient concentration analysis between B. infantis without cells (A) and with cells (B). The best performing formulation is denoted in black and referred to as Formulation BBI. Figure 12. The viability of B. infantis when lyophilized in either Formulation B (circle), the representative commercial formulation (square) or no formulation (triangle) and exposed to manufacturing and delivery processes. Each data point represents a single biological replicate. The boxed regions represent the upper and lower quartile, and the whiskers represent the upper and lower extremes. *P < 0.05, ** P < 0.01, *** P < 0.005, by Tukey’s multiple comparisons test (N = 7). Detailed Description As used herein and unless otherwise indicated, the terms “a” and “an” are taken to mean “one”, “at least one” or “one or more”. Unless otherwise required by context, singular terms used herein shall include pluralities and plural terms shall include the singular. Unless the context clearly requires otherwise, throughout the description and the claims, the words ‘comprise’, ‘comprising’, and the like are to be construed in an inclusive sense as opposed to an exclusive or exhaustive sense; that is to say, in the sense of “including, but not limited to”. Words using the singular or plural number also include the plural or singular number, respectively. Additionally, the words “herein,” “above” and "below" and words of similar import, when used in this application, shall refer to this application as a whole and not to any particular portions of this application. All embodiments of any aspect of the invention can be used in combination, unless the context clearly dictates otherwise. As used herein in all embodiments and combinations of embodiments, “about” means +/- 5% of the recited value. In a first aspect, the disclosure provides formulations, comprising one or more excipient selected from the group consisting of D-(-)-fructose, Bacto Tryptic Soy Broth (BTSB), potassium gluconate, magnesium D-gluconate, sucrose, and/or maltodextrin. As demonstrated in the examples, the addition of one of the recited excipients to an exemplary Lactobacillus strain, followed by lyophilization, resulted in viable cells being detected in the presence of atmospheric oxygen at extremely high temperature (60°C, a temperature used for pasteurization) for more than 2 weeks, while prior art formulations only show detectable viability under the same conditions for less than 48 hours. This stability at extremely high temperatures also translates to better viability at typical storage temperatures of 25°C and 4°C. Even further stability can be achieved when one of these “primary” excipients is combined with some of the secondary excipients recited in dependent claims. As used herein, BTSB has the following components: Bacto Tryptic Soy Broth Grams/Liter Casein peptone (pancreatic) 17.0 Soya peptone (papain digest.) 3.0 Sodium chloride 5.0 Dipotassium hydrogen phosphate 2.5 Glucose 2.5 Final pH 7.3 +/- 0.2. In some embodiments: (i) the one or more excipient comprises D-(-)-Fructose, optionally wherein the D- (-)-Fructose is present at between about 1000 mg/mL (i.e. (100% (w/v)) and 5000 mg/mL (i.e.500% (w/v)) in the formulation, or about 3200 mg/mL (320% (w/v)) in the formulation; or (ii) the one or more excipient comprises BTSB, optionally wherein the BTSB is present at between about 10 mg/mL (i.e.1% (w/v)) and 50 mg/mL (i.e.5% (w/v)) in the formulation, or about 24 mg/mL (i.e.2.4% (w/v)) in the formulation; or (iii) the one or more excipient comprises potassium gluconate, optionally wherein the potassium gluconate is present at between about 10 mg/mL (i.e.1% (w/v)) and 100 mg/mL (10% (w/v)) in the formulation, or about 40 mg/mL in the formulation, or about 13 to about 20 mg/mL when combined with secondary excipients (about 1.3% to about 2%); or (iv) the one or more excipient comprises magnesium D-gluconate, optionally wherein the magnesium gluconate is present at between about 20 mg/mL (i.e.2% (w/v)) and 200 mg/mL (i.e.20% (w/v)) in the formulation, or about 51 mg/mL to about 102 mg/mL (i.e. about 5.1% to about 10.2% (w/v)) in the formulation; or (v) the one or more excipient comprises sucrose, optionally wherein the sucrose is present at between about 1000 mg/mL (i.e.100% (w/v)) and 2500 mg/mL (i.e.250% (w/v)) in the formulation, or about 1576 mg/mL (i.e.157.6% (w/v)) in the formulation; or (vi) the one or more excipient comprises maltodextrin, optionally wherein the maltodextrin is present at between about 200 mg/mL (i.e.20% (w/v)) and 1000 mg/mL (i.e. 100% (w/v)) in the formulation, or about 400 mg/mL (i.e.40% (w/v)) in the formulation. Conversions to (w/v) are provided that are then added to the resuspended bacteria before lyophilization. The reason that some are >100% is that some of the sugars are so soluble that they are still solubility at a concentration >100 g/mL and thus have a >100% (w/v) concentration. In some embodiments, the formulation further comprises one or more of: (i) a gum (including but not limited to xanthum gum and/or sodium alginate), (ii) a sugar (including but not limited to fuctooligosaccharide and maltodextrin), (iii) a salt (including but not limited to sodium citrate, and L-Potassium acid tartate), (iv) a polymer (including but not limited to poly(ethylene glycol) and poly(vinyl alcohol)), (v) an acid (including but not limited to salicylic acid and lactic acid), (vi) an antimicrobial compound (including but not limited to sulfanilamide and metronidazole), (vii) amino acids and/or protein (including but not limited to L-proline and L- serine), (viii) an antioxidant (including but not limited to calcium L-ascorbate and biotin), and/or (ix) skim milk. In one embodiment, the one or more excipient comprises potassium gluconate. In these embodiments, potassium gluconate is the primary excipient. In one embodiment, the formulation further comprises L-histidine. As shown in the examples, formulations comprising potassium gluconate and L-histidine provided were demonstrated to be particularly useful as lyophilization solutions for Lactobacillus species, resulting in lyophilized Lactobacilli having significantly superior lyoprotection and thermostability compared to representative commercial formulations and other formulations tested herein. In various embodiments, the formulation comprises: (a) between about 0.4% (w/v) and about 4% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 3.65% (w/v) L-histidine; or (b) between about 0.4% (w/v) and about 2% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (c) between about 0.4% (w/v) and about 2% (w/v) potassium gluconate and between about 0.73% (w/v) and about 1.825% (w/v) L-histidine; or (d) between about 0.4% (w/v) and about 2% (w/v) potassium gluconate and between about 0.73% (w/v) and about 1.217% (w/v) L-histidine; or (e) between about 0.4% (w/v) and about 1.5% (w/v) potassium gluconate, and between about 0.73% (w/v) and about 1.825% (w/v) L-histidine; or (f) between about 0.4% (w/v) and about 1.333% (w/v) potassium gluconate, and between about 0.73% (w/v) and about 1.825% (w/v) L-histidine; or (g) between about 0.4% (w/v) and about 1.5% (w/v) potassium gluconate, and between about 0.73% (w/v) and about 1.3% (w/v) L-histidine; or (h) between about 0.4% (w/v) and about 1.333% (w/v) potassium gluconate, and between about 0.73% (w/v) and about 1.217% (w/v) L-histidine; or (i) between about 1.3% (w/v) and about 1.6% (w/v) potassium gluconate, and between about 1.2% (w/v) and about 1.3% (w/v) L-histidine; or (j) about 1.3% (w/v) potassium gluconate and about 1.2% (w/v) L-histidine; or (k) about 1.333% (w/v) potassium gluconate and about 1.217% (w/v) L-histidine. As shown in the examples that follow, formulations according to these embodiments, are particularly useful for lyophilizing Lactobacillus crispatus (Figure 4A), as well as Lactobacillus acidophilus and Lactobacillus gasseri (Figure 6). As will be understood by those of skill in the art, the formulations recited in this embodiment do not include any cells. In a further embodiment, the formulation further comprises bacteria selected from the group consisting of Lactobacillus species, including but not limited to Lactobacillus crispatus, Lactobacillus acidophilus, and Lactobacillus gasseri. In this embodiment, the formulation including cells may be used, for example, to lyophilize the cells for therapeutic and prophylactic use as described herein. In one embodiment, a cell pellet is resuspended in the formulation. In another embodiment, the formulation is diluted 1:1 with a suspension of cells (including but not limited to a suspension in water or phosphate buffered saline (PBS)), and thus the concentration of potassium gluconate and L-histidine are 50% reduced compared to the formulation without cells (Figure 4B). Thus, in this embodiment, the formulation comprises: (a) between about 0.2% (w/v) and about 2% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 2% (w/v) L-histidine; or (b) between about 0.2% (w/v) and about 1% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 0.912% (w/v) L-histidine; or (c) between about 0.2% (w/v) and about 1% (w/v) potassium gluconate and between about 0.36% (w/v) and about 0.912% (w/v) L-histidine; or (d) between about 0.2% (w/v) and about 1% (w/v) potassium gluconate and between about 0.36% (w/v) and about 0.608% (w/v) L-histidine; or (e) between about 0.2% (w/v) and about 0.75% (w/v) potassium gluconate, and between about 0.36% (w/v) and about 0.912% (w/v) L-histidine; or (f) between about 0.2% (w/v) and about 0.666% (w/v) potassium gluconate, and between about 0.36% (w/v) and about 0.912% (w/v) L-histidine; or (g) between about 0.2% (w/v) and about 0.75% (w/v) potassium gluconate, and between about 0.36% (w/v) and about 0.65% (w/v) L-histidine; or (h) between about 0.2% (w/v) and about 0.666% (w/v) potassium gluconate, and between about 0.36% (w/v) and about 0.608% (w/v) L-histidine; or (i) between about 0.65% (w/v) and about 0.8 % (w/v) potassium gluconate, and between about 0.6% (w/v) and about 0.65% (w/v) L-histidine; or (j) about 0.65% (w/v) potassium gluconate and about 0.6 % (w/v) L-histidine; or (k) about 0.667% (w/v) potassium gluconate and about 0.609% (w/v) L-histidine. The cells may be present in the formulation in any amount appropriate for an intended purpose. In some embodiments, bacteria at an OD600 of between 1 to 20 are used; or at an OD₆₀₀ of about 15. In other embodiments, the formulation comprises at least 1×10⁹, or at least 2×10⁹ colony forming units (CFU) of the bacteria. In other embodiments, the formulation comprises (a) between about 0.4% (w/v) and about 4% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 3.65% (w/v) L-histidine; or (b) between about 0.4% (w/v) and about 2% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (c) between about 0.4% (w/v) and about 0.8% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (d) between about 0.35% (w/v) and about 0.45% (w/v) potassium gluconate, and between about 0.3% (w/v) and about 1.3% (w/v) L-histidine; or (e) about 0.4% (w/v) potassium gluconate, and between about 0.3% (w/v) and about 1.3% (w/v) L-histidine; or (f) about 0.4% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 1.217% (w/v) L-histidine; or (g) about 0.4% (w/v) potassium gluconate and about 1.217% (w/v) L-histidine. As shown in the examples that follow, formulations according to these embodiments, are particularly useful for lyophilizing Lactobacillus jensenii (Figure 5A). As will be understood by those of skill in the art, the formulations recited in this embodiment do not include any cells. In a further embodiment, the formulation further comprises bacteria selected from the group consisting of Lactobacillus species, including but not limited to Lactobacillus jensenii. In this embodiment, the formulation including cells may be used, for example, to lyophilize the cells for therapeutic and prophylactic use as described herein. In one embodiment, a cell pellet is resuspended in the formulation. In another embodiment, the formulation is diluted 1:1 with a suspension of cells (including but not limited to a suspension in water or PBS), and thus the concentration of potassium gluconate and L-histidine are 50% reduced compared to the formulation without cells. Thus, in this embodiment, the formulation comprises (Figure 5B): (a) between about 0.2% (w/v) and about 2% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 2% (w/v) L-histidine; or (b) between about 0.2% (w/v) and about 1% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 0.917% (w/v) L-histidine; or (c) between about 0.2% (w/v) and about 0.4% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 0.917% (w/v) L-histidine; or (d) between about 0.35% (w/v) and about 0.225% (w/v) potassium gluconate, and between about 0.15% (w/v) and about 0.65% (w/v) L-histidine; or (e) about 0.2% (w/v) potassium gluconate, and between about 0.15% (w/v) and about 1.3% (w/v) L-histidine; or (f) about 0.2% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 0.608% (w/v) L-histidine; or (g) about 0.2% (w/v) potassium gluconate and about 0.609% (w/v) L-histidine. The cells may be present in the formulation in any amount appropriate for an intended purpose. In some embodiments, bacteria at an OD600 of between 1 to 20 are used; or at an OD600 of about 15. In other embodiments, the formulation comprises at least 1×10⁹, or at least 2×10⁹ colony forming units (CFU) of the bacteria. In other embodiments, the formulation comprises: (a) about 1.333% (w/v) potassium gluconate and about 1.217% (w/v) L-histidine; (b) about 2.000% (w/v) potassium gluconate and about 0.112% sulfanilamide; (c) about 1.333% (w/v) potassium gluconate and about 3.200% (w/v) glycine. formulation further comprises bacteria selected from the group consisting of Lactobacillus species, including but not limited to Lactobacillus crispatus, Lactobacillus acidophilus, Lactobacillus jensenii, and Lactobacillus gasseri. In another embodiment, the formulation comprises or consists of magnesium D- gluconate and L-histidine. As shown in the examples, formulations comprising potassium gluconate and L-histidine provided were demonstrated to be particularly useful as lyophilization solutions for Bifidobacterium species, resulting in lyophilized Bifidobacteria having significantly superior lyoprotection and thermostability compared to representative commercial formulations and other formulations tested herein. In various embodiments, the formulation comprises: (a) between about 1.02% (w/v) and about 10.2% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 3.650% (w/v) L-histidine; or (b) between about 2.04% (w/v) and about 10.2% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (c) between about 2.04% (w/v) and about 5.1% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (d) between about 5.0% (w/v) and about 5.2% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (e) between about 5.0% (w/v) and about 5.2% (w/v) magnesium D-gluconate, and between about 0.739% (w/v) and about 1.825% (w/v) L-histidine; or (f) about 5.1% (w/v) magnesium D-gluconate and about 0.1.217% (w/v) L- histidine; or (g) between about 2.0% (w/v) and about 2.1% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (h) between about 2.0% (w/v) and about 2.1% (w/v) magnesium D-gluconate, and between about 0.730% (w/v) and about 1.825% (w/v) L-histidine; or (i) between about 2.0% (w/v) and about 2.1% (w/v) magnesium D-gluconate, and between about 1.217% (w/v) and about 1.825% (w/v) L-histidine; or (j) about 2.04% (w/v) magnesium D-gluconate and about 1.217% (w/v) L- histidine. As shown in the examples that follow, formulations according to these embodiments, are particularly useful for lyophilizing Bifidobacterium species, exemplified by Bifidobacterium infantis (Figure 8 and 11A). As will be understood by those of skill in the art, the formulations recited in this embodiment do not include any cells. In a further embodiment, the formulation further comprises bacteria selected from the group consisting of Bifidobacterium species, including but not limited to B. infantis, B. bifidum, B. longum and B. breve. In this embodiment, the formulation including cells may be used, for example, to lyophilize the cells for therapeutic and prophylactic use as described herein. In one embodiment, a cell pellet is resuspended in the formulation. In another embodiment, the formulation is diluted 1:1 with a suspension of cells (including but not limited to a suspension in water or phosphate buffered saline (PBS)), and thus the concentration of magnesium D- gluconate and L-histidine are 50% reduced compared to the formulation without cells (Figure 11B). Thus, in this embodiment, the formulation comprises: (a) between about 0.51% (w/v) and about 5.10% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 1.825% (w/v) L-histidine; or (b) between about 1.02% (w/v) and about 5.10% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 0.913% (w/v) L-histidine; or (c) between about 1.02% (w/v) and about 2.55% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 0.913% (w/v) L-histidine; or (d) between about 2.5% (w/v) and about 2.6% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 0.913% (w/v) L-histidine; or (e) between about 2.5% (w/v) and about 2.6% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 0.913% (w/v) L-histidine; or (f) about 2.55% (w/v) magnesium D-gluconate and about 0.609% (w/v) L- histidine; or (g) between about 1.0 % (w/v) and about 1.05% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 0.913% (w/v) L-histidine; or (h) between about 1.0% (w/v) and about 1.05% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 0.913% (w/v) L-histidine; or (i) between about 1.0% (w/v) and about 1.05% (w/v) magnesium D-gluconate, and between about 0.609% (w/v) and about 0.913% (w/v) L-histidine; or (j) about 1.02% (w/v) magnesium D-gluconate and about 0.609% (w/v) L- histidine. The cells may be present in the formulation in any amount appropriate for an intended purpose. In some embodiments, bacteria at an OD600 of between 1 to 20 are used; or at an OD600 of about 15. In other embodiments, the formulation comprises at least 1×10⁹, or at least 2×10⁹ colony forming units (CFU) of the bacteria. In other embodiments, the formulation comprises: (a) about 5.1% (w/v) magnesium D-gluconate and about 8.8% (w/v) -alanine; or (b) about 5.1% (w/v) magnesium D-gluconate and about 1.217% (w/v) L- histidine; or (c) about 5.1% (w/v) magnesium D-gluconate and about 2% (w/v) L-alanine; or (d) about 5.1% (w/v) magnesium D-gluconate and about 2% (w/v) -aminobutyric acid. In further embodiments, the formulation further comprises bacteria selected from the group consisting of Bifidobacterium species. In some embodiments, the Bifidobacterium species comprise one or more species selected from but not limited to the group consisting of B. infantis, B. bifidum, B. longum and B. breve. In one embodiment of all of the formulations of the disclosure, the formulation or bacterial formulation does not comprise any sugars. As disclosed in the examples that follow, almost all D-(-)-fructose combinatorial formulations, resulted in no detectable viability after exposure to 60^oC, suggesting the degradation of sugar-based excipients at high temperatures cannot be compensated for by the presence of a different excipient. Thus, exclusion of sugars from the formulations provides a significant benefit for thermal stability of the bacteria. Any suitable solvent can be used in all embodiments of the formulations disclosed. In various embodiments, the excipients are dissolved in water or phosphate buffered saline (PBS). In one embodiment, the excipients are dissolved in water. In a further embodiment, the formulations consist of the recited concentration of excipients, with the solvent making up the remainder of the formulation. In one embodiment, the solvent is water. By way of non-limiting example, when the concentration of excipients as noted as follows: (a) between about 0.2% (w/v) and about 1% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 1% (w/v) L-histidine; then this embodiment would result in a formulation of (a) between about 0.2% (w/v) and about 1% (w/v) potassium gluconate and about 98.8% solvent/water (w/v), and between about 0.182% (w/v) and about 1% (w/v) L-histidine and about 97.18% solvent/water (w/v). Those of skill in the art will clearly understand the amount of solvent/water in other embodiments disclosed herein in view of this example. The formulations can be used using any suitable protocol. In one non-limiting embodiment, formulations for Lactobacillus stabilization can be made as follows: Stock concentration of potassium gluconate (40 mg/mL or 4% (w/v)) and L-histidine (36.48 mg/mL or 3.65% (w/v)) were prepared. Simultaneously, the L. crispatus cultures are prepared by pelleting the cultures at 4500 rpm for 15 min at 25^oC. The supernatant is removed from the bacterial pellets, before the bacteria is resuspended in the same volume of sterile PBS and spun again using the same protocol. Once pelleted again, the bacteria pellets are resuspended in sterile PBS to a volume that resulted in a final OD₆₀₀ of 15. The same volume of formulation is then prepared by combining the potassium gluconate stock, L-histidine stock and sterile water in a 1:1:1 ratio. This would result in a concentration of 13.3 mg/mL (1.33% (w/v)) potassium gluconate and 12.16 mg/mL (1.22% (w/v)) L-histidine. The formulation is then combined in a 1:1 ratio with the L. crispatus suspended in PBS (resulting in a final concentration of 6.67 mg/mL or 0.67% (w/v) potassium gluconate and 6.09 mg/mL or 0.61% (w/v) L-histidine) before being frozen and lyophilized. In another non-limiting embodiment, formulations for B. infantis stabilization can be made as follows: Similar to L. crispatus, stock concentrations of magnesium D-gluconate, hydrate (102 mg/mL or 10.20% (w/v)) and L-histidine (36.48 mg/mL or 3.65% (w/v)) are prepared. Simultaneously, the B. infantis cultures are prepared by pelleting the cultures at 4500 rpm for 15 min at 25^oC. The supernatant is removed from the bacterial pellets, before the bacteria are resuspended in the same volume of sterile PBS and spun again using the same protocol. Once pelleted again, the bacteria pellets are resuspended in sterile PBS to a volume that resulted in a final OD₆₀₀ of 12. The same volume of formulation is then prepared by combining the magnesium D-gluconate stock, L-histidine stock and sterile water in a 5:2:3 ration. This would result in a concentration of 51 mg/mL (5.10% (w/v)) magnesium D- gluconate hydrate and 12.17 mg/mL (1.217% (w/v)) L-histidine. The formulation is then combined in a 1:1 ratio with the B. infantis suspended in PBS (resulting in a final concentration of 25.5 mg/mL (2.55% (w/v)) magnesium D-gluconate hydrate and 6.09 mg/mL (0.609% (w/v)) L-histidine) before being frozen and lyophilized. In another aspect, the disclosure provides a lyophilized Lactobacillus cell population, comprising a matrix of hydrogen-bonded potassium gluconate and L-histidine in which the Lactobacillus cells are embedded. As disclosed in the examples that follow, scanning electron microscopy (SEM) revealed that lyophilization of Lactobacillus cell populations with formulations of the disclosure resulted in a matrix structure with little surface exposed bacteria, explaining the prolonged stability observed (Fig.3A-B). Without being bound by a mechanism of action, the matrix is believed to be a hydrogen-bonding and electrostatic interaction network of potassium gluconate and L-histidine, leading to significantly improved lyoprotection of Lactobacillus cells. Then, in the absence of sugars, the formulation is able to maintain integrity at high temperatures, resulting in the novel extremophile-like stability that has never previously been achieved and can abolish the need for cold-chain practices. In one embodiment, the Lactobacillus cells comprise or consist of one or more of L. crispatus, L. acidophilus, L. jensenii, L.rhamnosus, and L.gasseri. In another embodiment, Lactobacillus cells comprise or consist of one or more of L. crispatus, L. acidophilus, L. jensenii, and L.gasseri. The disclosure further provides a lyophilized Bifidobacterium cell population, comprising a matrix of hydrogen-bonded magnesium D-gluconate and L-histidine in which the Bifidobacterium cells are embedded. As disclosed in the examples that follow, scanning electron microscopy (SEM) revealed that lyophilization of Lactobacillus cell populations with formulations of the disclosure resulted in a matrix structure with little surface exposed bacteria, explaining the prolonged stability observed (Fig.10A-B). Without being bound by a mechanism of action, the matrix is believed to be a hydrogen-bonding and electrostatic interaction network of magnesium D-gluconate and L-histidine, leading to significantly improved lyoprotection of B. infantis. Then, in the absence of sugars, the formulation is able to maintain integrity at high temperatures, resulting in the novel extremophile-like stability that has never previously been achieved and could potentially abolish the need for cold-chain practices. In some embodiments, the Bifidobacterium cells comprises or consist of one or more of B. infantis, B. bifidum, B. longum and B. breve. In other embodiments, the Bifidobacterium cells comprise or consist of B. infantis. In these aspects, the lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population may be lyophilized in the formulation of any preceding claim. The lyophilization can be carried out using any suitable technique, including but not limited to the technique disclosed in the examples that follow. In one embodiment, the lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population is a dry powder, which can then be further formulated, if needed, when preparing the compositions of the disclosure. The lyophilized cells of the disclosure are stabilized and provide exceptional viability under extreme conditions, including more 2 months of storage at 60^oC. This extremophile- like stabilization allows the cells to withstand conventional manufacturing stressors and enabled its incorporation into multiple delivery systems, including capsules, tablets, oleogels and a novel intravaginal ring. This lyophilized cells provide a robust, low-cost strategy for enhancing the stability, scalability, and accessibility of L. crispatus-based therapeutics and supports expanded distribution to populations most burdened by vaginal microbiome dysbiosis. In one embodiment of any of the above embodiments, the disclosure provides compositions comprising the lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population of any embodiment herein, present in or on a delivery agent. Any delivery agent may be used as suitable for an intended use, including but not limited to a tablet, a capsule, a gel, a film, a suspension, a baby bottle (including but not limited to coated on the nipple or inside surface of the bottle), a pacifier, or a medical device (including but not limited to an intravaginal ring or a vaginal applicator). The compositions can be used in the methods of the disclosure detailed below. In one embodiment, the delivery agent further comprises one or more of boric acid, oleic acid, and itaconic acid. As detailed in the examples, boric acid, oleic acid and itaconic acid were the only compounds that inhibited all the pathogens without effecting the growth of L. crispatus. Both boric acid and oleic acid have been previously identified as safe for vaginal applications as well as useful compounds for promoting a Lactobacillus dominated microbiome. In one embodiment, the delivery agent further comprises boric acid. The boric acid, oleic acid and/or itaconic acid may be incorporated into the delivery agent at any concentration suitable for an intended purpose. In one non-limiting embodiment, the boric acid, oleic acid and/or itaconic acid is present in the delivery agent at between about 6 mg and about 600 mg. In one embodiment, the delivery agent comprises a vaginal ring. As used herein, a vaginal ring is a flexible, annular body configured for placement within the vaginal canal. The vaginal ring may be comprised of any materials consistent with its intended use. In non- limiting embodiments, the annular body may comprise a biocompatible elastomeric material selected from the group consisting of polydimethylsiloxane (PDMS), ethylene-vinyl acetate (EVA), polyurethane (PU), thermoplastic elastomers (TPEs), poly(lactic-co-glycolic acid) (PLGA), and siloxane-urethane copolymers. In some embodiments, the elastomeric material incorporates or encapsulates the lyophilized Lactobacillus cell population, and optionally incorporates or encapsulates at least one additional therapeutic agent that can negatively select pathogens or competing microbiota. If the vaginal ring is coated in the cells, the surface of the vaginal ring may be textured in order to improve adhesion and increase the amount of bacteria present. Alternatively, the vaginal ring may be hollow or porous and then loaded with either the lyophilized cells or a tableted version of the lyophilized cells. The cells can then be administered over an extended period of time through fenestrations or pores throughout the intravaginal ring. In some embodiments, at least one additional therapeutic agent comprises boric acid, oleic acid and/or itaconic acid. The vaginal ring can provide sustained, localized delivery of the lyophilized cells and, optionally, the at least one additional therapeutic agent over a period of at least 7 days. In one embodiment, the vaginal ring serves as a delivery agent for the lyophilized Lactobacillus cell population. Women with a low abundance of Lactobacillus species such as Lactobacillus crispatus in their vaginal microbiome are at a higher risk of reduced fertility or contracting various vaginal infections. In another embodiment, the delivery agent comprises breast milk, on a baby bottle (including but not limited to coated on the nipple or inside surface of the bottle), or a pacifier. In these embodiments, the lyophilized cells are either dissolved in breast milk, or coated on the nipple or inside surface of the bottle or the pacifier. In one embodiment, the cells can be incorporated into a suspension or gel, which can then be combined with breast milk and fed to the infant. Alternatively, the lyophilized bacteria can be incorporated into a film that can be used to coat a bottle or pacifier for administration to neonates. In one embodiment, the breast milk, baby bottle, or pacifier serves as a delivery agent for the lyophilized Bifidobacterium cell population. The reduced presence of Bifidobacterium species such as Bifidobacterium infantis in infants has been linked with a higher risk of premature birth complications or allergy disorders. In all embodiments of the compositions, the lyophilized cells may be present in any concentration suitable for an intended purpose. In one non-limiting embodiment, the cells are present in a concentration of at least 1 x 10⁹ colony-forming units (CFU). In another embodiment, the cells are present in a concentration of at least 2 x 10⁹ CFU. In another aspect, the disclosure provides methods for improving female fertility or limiting development of vaginal infections, comprising administering to a female in need thereof an amount effective of the lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population, or the composition, of any embodiment herein, to improve fertility or limit development of vaginal infections. In one embodiment, the method comprises administering the lyophilized Lactobacillus cell population, or composition thereof, of any embodiment herein. Women with a low abundance of Lactobacillus species such as Lactobacillus crispatus in their vaginal microbiome are at a higher risk of reduced fertility or contracting various vaginal infections. In a further aspect, the disclosure provides methods for reducing premature birth complications or allergic disorders, comprising administering to an infant in need thereof an amount effect of the lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population, or the composition, of any embodiment herein to reduce premature birth complications or allergic disorders in the infant. In one embodiment, method comprises administering the lyophilized Bifidobacterium cell population, or composition thereof, of any embodiment herein. The reduced presence of Bifidobacterium species such as Bifidobacterium infantis in infants has been linked with a higher risk of premature birth complications or allergy disorders. Any amount of the cells or compositions of the disclosure may be administered to the subject as appropriate under all circumstances. In one embodiment, at least 1 x 10⁹ CFU of cells are administered. In another embodiment, at least 2 x 10⁹ CFU of cells are administered. In all embodiments, the female or infant may be any suitable female of infant, including but not limited to a human female or human infant. The lyophilized cells or compositions may be administered to the subject once or may be administered on multiple occasions. The disclosure also provides lyophilization methods, comprising providing the bacterial formulation of any embodiment herein, and lyophilizing the bacteria in the bacterial formulation. In non-limiting embodiments, the bacterial formulation may be frozen by storage at -20^oC to about -80^oC for any suitable time, such as about 1-8 hours or more. Lyophilization can then be carried out for between about 8 hours and about 24 hours. In some embodiments, the bacteria are Lactobacillus or Bifidobacterium. The disclosure further provides methods for drying bacteria, comprising providing the bacterial formulation of embodiment herein, and drying the bacteria in the formulation. Any suitable drying method may be used, including but not limited to spray drying and lyophilization. In some embodiments, the bacteria are Lactobacillus or Bifidobacterium. Examples Example 1. Lactobacillus Abstract Bacterial vaginosis (BV), a prevalent condition linked to serious reproductive health complications, is driven by the loss of protective Lactobacillus crispatus in the vaginal microbiome. Although L. crispatus shows therapeutic potential, its clinical translation has been hampered by poor bacterial viability during manufacturing, transport, and storage, especially in environments lacking cold-chain infrastructure. Here, we developed a two- excipient formulation that stabilizes L. crispatus through lyophilization and provides exceptional viability under extreme conditions—including >2 months of storage at 60^oC. This extremophile-like stabilization allowed L. crispatus to withstand conventional manufacturing stressors and enabled its incorporation into multiple delivery systems, including capsules, tablets, oleogels and a novel intravaginal ring. In vivo studies in an ewe model with anatomical relevance to humans demonstrated the safe and prolonged vaginal presence of L. crispatus following single-dose administration. This work establishes a robust, low-cost strategy for enhancing the stability, scalability, and accessibility of L. crispatus- based therapeutics and supports expanded distribution to populations most burdened by vaginal microbiome dysbiosis. Introduction A healthy vaginal microbiome is often dominated by a single Lactobacillus species. These Gram-positive bacteria can produce lactic acid, hydrogen peroxide, biosurfactants and bacteriocin-like compounds that acidify the vaginal microenvironment, inhibit the proliferation of pathogenic microorganisms and subsequently promote the maintenance of an eubiotic microbiome. An absence of Lactobacilli in the vaginal microbiome can often result in bacterial vaginosis (BV). BV is characterized by an increase in microbiome diversity in the vaginal tract in which an abundance of opportunistic commensal or pathogenic microbes can be detected. This condition affects between 20–60% of people with vaginal tracts globally and is the most common disorder of the vaginal tract in which clinical intervention is sought. Prolonged BV has been associated with several serious comorbidities such as: recurrent urinary tract infection, increased risk of sexually transmitted infections, pre-term labour and infertility. These serious consequences of BV combined with its prevalence secure BV as a clinical priority. The current clinical treatment for BV involves the administration of antibiotics such as metronidazole or clindamycin. However, biofilm formation can reduce the efficacy of antibiotic treatment and, even when effective, there is no guarantee of Lactobacillus recolonization. As a result, BV recurrence rates have been reported as high as 60% within 12 months of treatment. This high recurrence and subsequent concern regarding antibiotic resistance has led to alternate methods of treating BV to be investigated such as the administration of Lactobacilli to the vaginal tract. However, only viable bacteria can have a therapeutic effect and most traditional therapeutic manufacturing processes involve several stressors that can significantly reduce viability. Specifically, L. crispatus requires dehydration to be compatible with various methods of administration which exposes the bacteria to stressors such as freezing temperatures or shear forces. The remaining viability in the dried product also continues to decrease during transport and storage, especially without the use of cold-chain practices. Considering BV is prevalent in low-resource populations, room temperature stable L. crispatus-based live biotherapeutic products are essential for equal distribution. Thus, we have first evaluated the clinical and commercial landscape of Lactobacillus-based products for vaginal applications in order to understand the current state of L. crispatus products and their stability. Despite the clinical efficacy demonstrated by L. crispatus, a significant lack of formulations that were able to maintain viability was observed. To address this demand, a high-throughput screen was adapted and implemented in order to identify an inexpensive two-excipient formulation for stabilizing L. crispatus during lyophilization. This formulation provided L. crispatus with extremophile-like properties in which viability could be detected after long-term storage even at extreme temperatures such as 60^oC. The formulation was compatible and maintained long-term viability throughout various manufacturing and delivery methods as well as allowed for the investigation of new methods of delivery that have previously been unachievable. Several other therapeutically relevant Lactobacillus strains were also demonstrated to be stabilized by the formulation highlighting the potential universal application for primary or engineered strains of Lactobacilli. Finally, the developed formulations demonstrated efficacy in a novel in vivo model of BV providing the foundation for potentially establishing a new pipeline for the translation of L. crispatus to the clinic. Results Survey of the Clinical Demands for Stable, Lactobacillus-Based Products The application of L. crispatus to the vaginal microbiome to treat dysbiosis represents a promising method of reducing reoccurrence of BV. To date, 32 total clinical trials have been conducted using L. crispatus for vaginal health applications in the United States. These L. crispatus clinical trials predominantly focus on the maintenance of a healthy vaginal microbiome or treating dysbiosis. However, more recent studies have begun investigating the prevention of the subsequent comorbidities associated with BV, such as recurrent urinary tract infections, infertility, and prevention of pre-term labor and miscarriages. In order to achieve a therapeutic effect, most trials administered ~10⁹ colony forming units (CFUs) of L. crispatus multiple times a week, minimum. Additionally, when disclosed, most clinical trials relied upon patients refrigerating the probiotics or live biotherapeutic products before use, with one trial even citing owning a refrigerator as an inclusion criterion. As viability is directly related to therapeutic efficacy and clinical outcomes, this reliance on refrigeration indicated that maintaining viability during storage remained a challenge for Lactobacillus-based products. Thus, an investigation into the bacterial viability in some of the products used in the clinical trials as well as a variety of commercially available Lactobacillus-based probiotics was conducted. In an attempt to capture an insight into the Lactobacillus viability currently accepted commercially, various different formulations and methods of administration were investigated. As a result, both vaginal suppositories (capsules) as well as orally administered capsules and tablets were investigated. Despite the facultative anaerobic nature of L. crispatus, the Lactobacillus-based products performed poorly and demonstrated significantly less viable bacteria than expected or marketed. The best performing product, HappyV^TM, was measured to contain only 23% of the expected bacteria and ultimately fell below the viability of 10⁹ CFU/dose indicated to be required for a therapeutic effect by the clinical trials. Considering some of the populations most affected by L. crispatus-related comorbidities are located in areas such as sub-Saharan Africa which cannot rely on cold-chain practices, there is a clear demand for a formulation that could maintain L. crispatus viability without refrigeration. High-Throughput Screening for Synthetic Extremophile Formulations A high-throughput screen in which FDA-approved Generally Recognized as Safe (GRAS) compounds was adapted for L. crispatus to identify excipients that could maintain viability after lyophilization and eliminate reliance on cold-chain practices. Considering the minimal viability of commercial Lactobacillus products, an expanded excipient library that consisted of 479 FDA-approved GRAS compounds, food additives and compounds previously used for intravaginal administration were investigated. Three different concentrations of each excipient, resulting in 1437 unique formulations, were investigated for lyoprotective properties against L. crispatus. Upon completing the screens, 14 excipients were identified to perform equivalently or better than the common commercial lyoprotectant, trehalose. A normalized viability cut-off value of 0.8 was then applied, resulting in 51 total excipients being selected for further large- scale validation. Expectantly, sugars represented the greatest proportion of the selected excipients. Upon validation, all lyoprotective excipients identified in the high-throughput screen demonstrated greater viability after lyophilization than the absence of excipients, with 21 excipients indeed performing equivalently or better than the trehalose control. Ultimately, the three best performing excipients were D-(-)-fructose, bacto tryptic soy broth and potassium gluconate. Complexing these primary excipients with the excipient library resulted in additional lyoprotective and long-term stability. Three additional high-throughput screens were conducted with each primary excipient, resulting in an additional 1437 unique formulations examined, for a total of 2874 formulations investigated overall. These screens identified 335 combinatorial formulations that demonstrated greater viability after lyophilization than each primary excipient alone. A number of the same secondary excipients were selected for each primary excipient, with 28% of the secondary excipient hits being identified as additive to more than one of the primary excipients. A greater diversity in chemical properties amongst the secondary excipient hits was also observed in comparison to the primary excipient hits. Considering the significant number of combinatorial formulations that outperformed the primary excipients and the ultimate aim of eliminating cold-chain practices for live biotherapeutic products, a “commercial feasibility filter” was applied to the excipients in order to identify the most feasible combinatorial formulations to pursue. This filter consisted of four criteria. Firstly, due to the global demand for L. crispatus-based products, the allergies and dietary preferences of patients were considered, and all animal-derived excipients or common allergens were removed. Then, in order to simplify production costs, excipients that consisted of medias, broths and extracts that would likely differ between batches were removed. All excipients that were not stable at room temperature were also removed in an attempt to bias the final combinatorial formulations for greater stability at extreme temperatures. Finally, only excipients that were inexpensive (<1.00 USD/mL) to purchase at high purity in bulk-scale were selected in order to further reduce costs. Applying this filter resulted in 178 unique combinatorial formulations selected for validation. To ensure the lyoprotective combinatorial formulations also enabled long-term viability in the absence of cold-chain practices, the lyophilized products were aerobically stored at 60^oC for 24 h before viability was assessed. Implementing this additional selective pressure resulted in the identification of 24 combinatorial formulations that demonstrated greater L. crispatus viability after lyophilization and high temperature storage than the primary excipients alone. The top three performing combinatorial formulations were all from the same library in which potassium gluconate was complexed with either L-histidine, sulfanilamide or glycine. It was also noted that almost all D-(-)-fructose combinatorial formulations, resulted in no detectable viability after exposure to 60^oC, suggesting the burning of sugar-based excipients at high temperatures cannot be compensated for by the presence of a different excipient. An additional improvement in viability after aerobic storage at 60^oC for 24 h could also be achieved for each combinatorial formulation through further assessment of each excipient concentration (Fig.1). For example, concentrations between 0.4-4% potassium gluconate and 0.365-3.65% L-histidine were observed to be required in order to provide a lyoprotective effect to L. crispatus (Fig.4). The data were plotted in a heat map as shown in Figure 4, showing those concentrations and ratios of potassium gluconate and L-histidine that provided a surprisingly significant increase in lyoprotective and thermal stability for L. crispatus. In order to determine whether long-term stability without refrigeration could be achieved, the stability of the top three, combinatorial formulations was examined at 60^oC and 37^oC for 6 months in comparison to a representative commercial formulation. Considering pasteurization protocols begin at 60^oC, L. crispatus was not expected to maintain viability at this extreme temperature. However, all combinatorial formulations outperformed the representative commercial control, with the best performing formulation, (i.e. Formulation A), maintaining detectable viability for >2 months after storage at 60^oC (Fig.2A). In the same conditions, the representative commercial formulation had no detectable viability after 1 week. This improvement in long-term stability was also observed at 37^oC (Fig.2B). Visual examination of the lyophilized products after 6 months of storage at 60^oC highlighted a lack of burning in Formulation A in comparison to the sugar-based representative commercial formulation. Additionally, scanning electron microscopy (SEM) of the samples revealed that lyophilization with Formulation A resulted in a matrix structure with little surface exposed bacteria, explaining the prolonged stability observed (Fig.3A-B). Without being bound by a mechanism of action, the matrix is believed to be a hydrogen-bonding and electrostatic interaction network of potassium gluconate and L-histidine, leading to significantly improved lyoprotection of L. crispatus. Then, in the absence of sugars, the formulation is able to maintain integrity at high temperatures, resulting in the novel extremophile-like stability that has never previously been achieved and can abolish the need for cold-chain practices. One final high-throughput screen was conducted to determine if a tertiary excipient achieved further lyoprotection, however all tertiary excipient hits that improved initial viability after lyophilization reduced the long-term stability of the lyophilized product at high temperatures. Extremophilic Formulation Provided an Opportunity to Investigate Alternative Delivery Methods Another challenge faced by L. crispatus therapeutics is maintaining viability after administration. Since opportunistic commensal and pathogenic microbes dominate the dysbiotic vaginal epithelial, the pH of the vaginal tract generally increases to more neutral values and there is little surface area for L. crispatus to colonize. L. crispatus colonization is directly correlated to a reduction in BV recurrence. Therefore, to combat the opportunistic commensal or pathogenic microbes, a course of antibiotics is generally administered before L. crispatus is applied. However, this continues to amplify the concern of antibiotic resistance. Therefore, in an attempt to identify excipients that could be incorporated into the extremophilic formulation to inhibit commensal opportunistic or pathogenic microbes and promote L. crispatus colonization, the high-throughput pipeline with the excipient library was adapted to screen for inhibition. To assess the inhibitory potential of the excipient library a screen against L. crispatus and four different pathogenic or opportunistic commensal microbes was conducted. Each microbe was selected for either its direct association with BV (Gardnerella vaginalis) or its roles in adjacent conditions such as vaginal candidiasis (Candida albicans) or urinary tract infections (Escherichia coli)(9, 23, 24). Additionally, Lactobacillus iners was examined as it is a Lactobacillus sp. that is often observed to colonize the vaginal tract after antibiotic treatment alone (25). However, since L. iners does not express hydrogen peroxide and produces minimal lactic acid, its presence often results in greater rates of BV recurrence (25). Of the 479 compounds that were tested against each microbe in the high-throughput screen, 20 were found to inhibit all pathogenic microbes. However, only seven of these compounds were selective for all vaginal pathogens without inhibiting L. crispatus. Of the 17 excipients selected for further validation, boric acid, oleic acid and itaconic acid were the only compounds that inhibited all the pathogens without effecting the growth of L. crispatus (Table 1). Both boric acid and oleic acid have been previously identified as safe for vaginal applications as well as useful compounds for promoting a Lactobacillus dominated microbiome (25, 26). Therefore, both were selected as excipients to complement the extremophilic formulation for L. crispatus. Table 1. The MIC values for the excipient hits identified in the high-throughput pathogen screens Since Formulation A consists of only two excipients, it is significantly less dense than commercial formulations resulting in far greater viability per gram of lyophilized product (Fig.7A). The formulation also demonstrated compatibility with traditional manufacturing processes. Many manufacturing methods are often considered too harsh for bacteria as milling involves shear force and wet granulation requires organic solvent exposure. However, the unique stability provided by Formulation A translated to improved viability during these processes. Milling and wet granulation of Formulation A resulted in uniform powders with minimal surface exposed bacteria that could be easily loaded and ejected from commercially available vaginal applicators. The Formulation A product demonstrated and maintained a ~1.5-log improvement in viability in comparison to the representative commercial formulation throughout the milling process (Fig.7A). A significant decrease in viability could only be detected after wet granulation, which was expected considering the antimicrobial nature of isopropanol. However, whilst a 2.5-log reduction in viability was observed for Formulation A, a 3.4-log reduction was observed for the representative commercial formulation highlighting the novel protective capabilities of Formulation A (Fig. 7A). The extremophilic formulation was then examined in traditional methods of delivery that have previously been investigated for L. crispatus. L. crispatus-loaded capsules, tablets, gels and films have all been investigated clinically but also involve additional stressors such as compression, hydrophobicity or baking. However, the lyophilized Formulation A product could successfully be incorporated into each delivery method with little detectable impact on viability (Fig.7B). To evaluate whether the long-term stability provided by Formulation A could be maintained in these current methods of delivery, three representative commercial products were created: a tablet, capsule and gel. Boric acid was incorporated into the tablet and capsules, while oleic acid was utilized to form the oleogel in order to produce representative products with the additional capacity for pathogen inhibition. After 1 month of aerobic storage at 37^oC, each method of delivery demonstrated significant stability with a relevant therapeutic dose (~2×10⁹ CFU/g) of viable bacteria still present. Currently, all commercially available and clinically investigated L. crispatus products require multiple doses a week in order to achieve recolonization of the vaginal tract. This vaginal dosing schedule has often been identified as inconvenient and uncomfortable for patients leading to low patient adherence. However, the significant loading capacity and novel long-term stability of Formulation A, combined with the pathogen inhibitory properties of the oleic acid oleogel provides the unique opportunity to investigate new methods of prolonged L. crispatus release. A common method of prolonged delivery to the vaginal tract is via an intravaginal ring (IVR). Thus, a hollow IVR with the dimensions of the clinically available Femring was designed (Table 2). Eight 2 mm diameter openings were evenly distributed throughout the IVR as this was determined to provide the best in vitro release of L. crispatus. After loading the IVR with 500 mg/mL of L. crispatus lyophilized in Formulation A and suspended in oleic acid oleogel, prolonged release of L. crispatus could be detected for a week in synthetic vaginal fluid (SVF) in comparison to the same dose suspended in SVF without the IVR (Fig.7C). Therefore, Formulation A not only provides unique long-term stability, but allows for novel methods of delivery that potentially provide a more comfortable experience for patients, to be investigated. Table 2.The dimensions and physical properties of the oleogel-loaded intravaginal ring in comparison to other developed or commercially available intravaginal rings. Clinical Translatability of the L. crispatus Extremophilic Formulation It is well recognized that L. crispatus is not the only beneficial Lactobacillus sp. for vaginal health. Other strains such as L. gasseri and L. jensenii have been observed to dominate the vaginal microbiome and combat BV development. Additionally, whilst L. acidophilus is originally derived from the gastrointestinal microbiome, it is a strain that is commonly present in commercial or clinical products for vaginal health. Therefore, the compatibility of Formulation A with these other important Lactobacillus strains was investigated. Formulation A provided L. acidophilus and L. gasseri with significantly more lyoprotection and thermostability than the representative commercial formulation (Fig.6A- D). Additionally, after further analysis of the potassium gluconate and L-histidine concentrations, Formulation A could also demonstrate equivalent lyoprotective and stabilizing effects for L. jensenii in comparison to the representative commercial control (Fig. 5-6A-B). However, potassium gluconate and L-histidine have not been previously examined for vaginal administration. Therefore, in order to ensure the presence of Formulation A does not adversely affect beneficial bacteria present in the dysbiotic vaginal microbiome, its inhibitory potential was examined against L. crispatus, L. gasseri and L. jensenii. No inhibitory effect was observed for any of the strains at the highest concentration for each excipient tested suggesting the presence of Formulation A should not hinder the shift to a Lactobacillus- dominated vaginal microbiome. Additionally, the impact of both potassium gluconate and L- histidine on the viability of two vaginal epithelial cell lines, End1 and Vk2, was examined. Assuming a vaginal vault volume of 50 mL, no significant impact on cell viability was determined for either the excipients alone or combined at the concentration required to administer 2×10⁹ CFU. Considering these positive indicators in the translatability of Formulation A, its preliminary safety and efficacy was examined in vivo. Due to the physiological and anatomical similarities between the ewe and human vaginal tract, a single dose of 2×10⁹ CFU of L. crispatus in various formulations was delivered via an applicator to the vaginal tract of nulliparous ewes, and the presence of L. crispatus was monitored over a 2-week period. During this time, no inflammation of the vaginal vault could be detected visually. The L. crispatus lyophilized in the absence of formulation was >100-fold greater in mass than the Formulation A doses, yet the L. crispatus could only be detected for 48 h post-administration. Alternatively, both powder and oleogel doses using L. crispatus lyophilized in Formulation A could be detected for >1 week in the ewe vaginal tract, with one ewe dosed with the oleic acid oleogel formulation still demonstrating detectable L. crispatus two weeks post- administration. Since the L. crispatus-loaded oleic acid oleogel demonstrated the greatest resonance of all the formulations examined, a daily dosing study was also conducted. This dosing was directly compared with the oleogel-loaded IVR, which aimed to administer the same 2x10⁹ CFU dose each day with only one initial administration. The presence of the L. crispatus in the ewe vaginal tract was identical between the single dosing and the daily dosing studies, suggesting the rapid decay in the first 24 h after administration occurs each day and no L. crispatus is accumulating throughout this time. For the IVR group however, a greater presence of L. crispatus could be detected for the first 5 days after insertion (Table 3). However, a rapid decline was then observed after day 5, resulting in no detectable L. crispatus for the remainder of the study. The difference in L. crispatus detection from the IVR in comparison to the daily dosing suggests the bacteria may have been released from the IVR at a faster rate than observed in vitro. This could be explained by the additional contractive forces that act upon the IVRs when placed in the vaginal vault, which cannot be captured in vitro. Removal of the IVRs on day 7 highlighted that the L. crispatus-loaded oleogel had indeed been completely administered but discharge had begun to fill the IVRs in its place. Additionally, SEM of the IVRs also demonstrated mucus and potential biofilm development on their surface potentially further contributing to the rapid decline in detectable L. crispatus after day 5 of the study. Table 3. Area under the curve of the qPCR data for each in vivo study conducted. Discussion In this study, we have successfully adapted and expanded upon a high-throughput screen in order to identify a novel stabilizing formulation for L. crispatus. Lactic acid expressing bacteria have many industrial, agricultural and pharmaceutical applications. However, very limited investigation into stabilizing L. crispatus for vaginal health applications has been conducted. Of the formulations that have been developed for Lactobacillus sp. all contain a sugar-based excipient that will inevitably burn or ferment at higher temperatures. As a result, the stability of these formulations is often only examined at a maximum of 25^oC for 9 months. Sugars are also generally hygroscopic leading to the rehydration of lyophilized bacteria in high humidity and a significant loss in viability. Without being bound by a mechanism of action, the matrix is believed to be a hydrogen- bonding and electrostatic interaction network of potassium gluconate and L-histidine, leading to significantly improved lyoprotection of L. crispatus. In the absence of sugars, the formulation is able to maintain integrity at high temperatures, resulting in the novel extremophile-like stability that has never previously been achieved and could potentially abolish the need for cold-chain practices. When considering how to best administer this extremophilic formulation, there is significant discourse and ambiguity regarding the optimal method of delivering Lactobacillus to the vaginal tract. The most established methods include capsules and tablets, which are simple and inexpensive to produce. However, both require frequent administration and often result in increased discharge, impacting patient adherence and therapeutic outcome. Alternatively, films represent another method of vaginal delivery that have been investigated to circumvent these disadvantages, but they are often excluded for L. crispatus delivery, due to their smaller loading capacity. The viability provided to L. crispatus by the formulations disclosed herein, establishes a new gel-based method of delivery. Additionally, the ability for oleic acid to inhibit opportunistic commensal or pathogenic bacteria and promote L. crispatus growth highlights a potential pathway to reducing the dependence upon antibiotic treatments in the clinic. This compatibility of the present formulations with all current methods of delivery ensures patient comfort and convenience can be prioritized, as more investigation into the best method of delivery is conducted. The IVR developed in this study represents a novel method of prolonged delivery to the vaginal tract. IVRs have been determined as well tolerated by patients for various gynaecological applications, with two IVRs available commercially (i.e. Nuvaring and Femring). The IVR designed in this study has equivalent dimensions to the Femring and demonstrated retention in vivo with minimal discomfort or irritation. Thus, it represents the first clinically feasible IVR developed and tested in vivo for the release of viable L. crispatus. Additionally, the greater presence of L. crispatus in the IVR groups for the first 5 days of the study suggest that with further optimization, a L. crispatus-loaded IVR could potentially provide equivalent or better clinical outcomes to daily dosing with only a single administration. The novel thermostability the formulations of the disclosure provide L. crispatus, specifically at 60^oC, provide a solution to challenges of biofilm formation and backfilling of the IVR with discharge. For example, the extrusion of an IVR with the stabilized L. crispatus embedded throughout could be investigated. This would ideally result in a solid IVR that could release L. crispatus into the vaginal tract without the requirement for a hollow interior, ultimately removing concerns of backfilling with discharge and potentially minimizing biofilm formation. Extrusion techniques for L. crispatus-loaded IVRs have previously been unachievable with all current sugar-based formulations because they would ultimately caramelize. Considering the stabilizing effects of the formulations of the disclosure were observed for a variety of Lactobacillus strains, the formulation will be amenable to primary and genetically modified strains of Lactobacillus, such as L. crispatus. Spray drying is another common large-scale method to dry bacteria that is utilized in the manufacturing of live biotherapeutic products. While spray drying involves different stressors in comparison to lyophilization, the formulations of the disclosure demonstrated protective capabilities against various stressors. As a result, the formulation’s protective properties translate to spray drying as well. Finally, we have provided the foundation for a new in vivo method of examining the therapeutic efficacy of L. crispatus. Usually, human clinical trials have to be conducted in order to assess the efficacy of L. crispatus products. However, this is time consuming and creates a significant ravine between preliminary research and translation to the clinic. In vivo analysis of L. crispatus-based therapeutics in ewes could provide an intermediate step that bridges this gap allowing for more rapid translation to the clinic. However, the vaginal microbiome of ewes is not innately dominated by Lactobacillus sp.. Instead, the microbiome of the ewe vaginal tract is highly diverse and has a more neutral pH of ~6.7. As a result, colonizing the ewe vaginal microbiome with L. crispatus is highly challenging. However, the ewe vaginal microbiome shares similar properties to the human vaginal microbiome diagnosed with BV. Therefore, long-term detection of L. crispatus would likely suggest significant efficacy in humans. Additionally, the anatomic dimensions and epithelial similarities between the ewe and human vaginal tract provides a foundation for the safety of devices and therapeutics. For example, whilst both potassium gluconate and L-histidine have not been examined for vaginal administration they are both well established as safe for oral consumption in humans as well as showed no cytotoxicity in the ewe studies. Materials and Methods Study Design The ultimate objective of the study was to develop a formulation for L. crispatus that could improve long-term bacterial viability in the presence of various stressors to reduce reliance upon cold-chain practices and improve distribution of live biotherapeutic products to low-resource populations. This formulation needed to also discourage pathogen and opportunistic commensal microbe growth, promote L. crispatus colonization in the vaginal microbiome and be compatible with various manufacturing and delivery methods. To assess the translational potential of these formulations, in vivo studies in sheep were implemented. For all studies, three sheep were used in each group. Experiments were not blinded, and two technical replicates for each sheep were conducted at each sampling point. All animal experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals (National Research Council), and all procedures were approved by the Tufts University Research Institution Animal Care and Use Committee (protocol number G2024-45) and performed with the support of the Solvandria Foundation. High-Throughput Screen for Lyoprotective Formulation L. crispatus was cultured and the compound library was prepared at described in Supplementary Methods and Materials S2.1. and S2.2., respectively. For the initial high- throughput screen, L. crispatus suspended in PBS to an OD600 of 1.5 was combined with the arrayed library in a 1:1 ratio to a final volume of 200 μL in a U-bottom 96-well plate using a liquid handler robot (OpentronFlex^TM, Opentron). In each plate, a positive control well of trehalose as well as a negative control well of water was included. For the combinatorial screen, the primary excipients (D-(-)-fructose, bacto tryptic soy broth or potassium gluconate) were prepared at 80% their solubility in water and complexed in a 1:1 ratio with the excipient library to a final volume of 100 μL in a U-bottom 96-well plate using a liquid handler robot (OpentronFlex^TM, Opentron). The L. crispatus culture suspended in PBS was added to the complexed library to a final volume of 200 μL/well. All high-throughput screens were conducted in triplicate. Once prepared, the arrayed L. crispatus+material plates were covered with parafilm and immediately placed at -25^oC to freeze for 2–4 h. The frozen plates were transferred to a tray freeze dryer (FreeZone^TM Stoppering Tray Dryer, Labconco) and dried at 25^oC and 0.1 mBar for 15 h total. The dried wells for each plate were rehydrated in 200 μL of PBS using the liquid handler robot (OpentronFlex^TM, Opentron). Three 10-fold serial dilutions for the initial high-throughput screen or four 10-fold serial dilutions for the combinatorial high-throughput screen were conducted before 4 μL of the most dilute samples were spotted onto MRS agar in a 1-well rectangular plate. The rectangular agar plates were incubated anaerobically (Vinyl Type A Anaerobic Chamber, Coy Lab Products) at 37^oC for 48 h before the plates were imaged in an iBright^TM Imaging System (ThermoFisher Scientific) using a standard protein gel light protocol. An automated colony counting protocol was implemented as described in Supplementary Methods and Materials S3 to identify lyoprotective hits (20). Vial-Based Validation of Lyoprotective Hits The best performing excipients from the initial lyoprotective high-throughput screen were validated at a larger scale to determine a more accurate relative viability value (CFU/mL). Results are shown in Figure 4A (without cells) and 4B (with cells). L. crispatus suspension in PBS was prepared as described in Supplementary Methods and Materials S2.1. before being combined in a 1:1 ratio with 1 mL of each excipient hit prepared at the lyoprotective concentration identified in the high-throughput screen. The final L. crispatus- compound vials were vortexed briefly before immediately placed at -25^oC for 2–4 h. Once frozen, the vials were transferred to a tray freeze dryer (FreeZone^TM Stoppering Tray Dryer, Labconco) and dried at 25^oC and 0.1 mBar for 15 h total. The viability after lyophilization for each excipient was assessed by rehydrating the samples with 2 mL of PBS. Five 10-fold serial dilutions were prepared before 20 μL of each dilution and the vial sample were plated onto MRS agar in two 100 mm three-segment circle petri dishes. The agar plates were incubated anaerobically (Vinyl Type A Anaerobic Chamber, Coy Lab Products) for 48 h at 37^oC. Colonies were counted at the dilution that provided ~10-200 distinct colonies and the CFU/mL for each excipient was calculated. The lyoprotective capacity of all excipient hits were evaluated using five biological replicates. Each two-compound combinatorial formulation was prepared in 20 mL glass scintillation vials to a final volume of 1 mL. The L. crispatus was prepared as described in Supplementary Methods and Materials S2.1. before 1 mL was added to all combinatorial formulation vials and they were immediately stored at -25^oC for 2–4 h. Once frozen, the vials were transferred to a tray freeze dryer (Labconco, FreeZone^TM Stoppering Tray Dryer) and dried at 25^oC and 0.1 mBar for 15 h total. The dried material vials were packed into zip-loc bags with a desiccant pack and stored aerobically in the dark at 60^oC. After 24 h, the viability in the vials was assessed as described for the initial lyoprotective excipient hits. To determine the optimal concentration for each of the excipients in the three best performing combinatorial formulations, sterile stocks of potassium gluconate and each secondary excipient (i.e. L-histidine, sulfanilamide or glycine) were prepared at 80% their solubility in water. Different ratios of the potassium gluconate stock and each secondary excipient stock were prepared into 20 mL glass scintillation vials. Sterile water was added for a final volume of 1 mL in each vial. L. crispatus was then prepared as described in Supplementary Methods and Materials S2.1. before 1 mL of the culture was added to each vial and all samples were immediately stored at -25^oC for 2–4 h. Once frozen, the vials we lyophilized, stored at 60^oC and their viability was evaluated after 24 h as previously described for the lyoprotective combinatorial formulation hits. Impact of Excipients on the Vaginal Microbiome High-Throughput Screen for Pathogen Inhibitors All microbial strains used in this study were obtained from the American Type Culture Collection (ATCC). Microbial strains were prepared from glycerol stocks stored at - 80°C. The same compound library as the lyoprotective screen was used and the high- throughput inhibitory screen was performed in triplicate in clear, sterile, flat-bottom 96-well plates. Using a liquid handler (OpentronFlex^TM, Opentron), 20 L of each compound stock, 130 L of broth, and 50 L of diluted microbial culture were combined for a final microbial concentration of 5×10⁵ CFU/mL in each well. All plates were incubated and visually scored 24–48 h after the screen. Quantification of the Minimum Inhibitory Concentration of Excipient Hits Inhibitory excipient hits were selected based upon the commercial feasibility filter, inhibitory activity in the high-throughput screen and chemical properties (e.g. safe dosage range, current applications as antimicrobials, etc.). Excipients from the top formulations for L. crispatus were also included to examine their potential impact upon the microbiome. For all microbes, including L. gasseri and L. jensenii, a minimum inhibitory concentration (MIC) assay was performed using modified CLSI guidelines, as described in Supplementary Methods and Materials S5. Compatibility of the Formulation with Various Manufacturing and Delivery Processes Viability Throughout Different Manufacturing Processes An L. acidophilus, L. crispatus, L. gasseri or L. jensenii suspension in PBS was prepared as described in Supplementary Methods and Materials S2.1. except the bacteria was suspended at an OD600 of 15. The suspensions were then combined in a 1:1 ratio with either Formulation A, the representative commercial formulation or no formulation, before being immediately stored at -25^oC for 2–4 h. Once frozen, the vials were lyophilized as previously described in “Vial-Based Validation of Lyoprotective Hits”. A proportion of the lyophilized L. crispatus powders were then mixed with the following excipients: 1% (w/w) magnesium stearate, 4% (w/w) polyvinylpyrrolidone, and 65% (w/w) -lactose, for a final loading of 30% (w/w) of the different lyophilized L. crispatus samples. This milled powder proceeded to wet granulation. Wet granulation was conducted by first adding 300 μL of isopropanol per g of milled product. The mixture was vortexed for 5 min to evenly combine before the resulting paste was passed through a bench top oscillating granulator (FGS II, ERWEKA) with a 1 mm mesh screen. The collected granules were dried at 60^oC for 30 min in 20 mL glass scintillation vials. The lyophilized samples of L. acidophilus, L. crispatus, L. gasseri, and L. jensenii before and after aerobic storage at 60^oC for 24 h as well as the milled and dried granule samples were weighed, before 2 mL of PBS was added to all samples. Five 10-fold dilutions were prepared for each sample before 20 μL of each dilution as well as the undiluted samples were plated on MRS agar in two 100 mm three-segment petri dishes. The agar plates for all samples were incubated anaerobically (Vinyl Type A Anaerobic Chamber, Coy Lab Products) for 48 h at 37^oC and colonies were counted at the dilution that provided ~10-200 distinct colonies before the CFU/g for each sample was calculated by dividing the mass of only the bacterial fraction. Viability in Traditional Methods of Delivery Lyophilized L. crispatus samples in Formulation A, the representative commercial formulation or no formulation were prepared and milled as described in the above subsection. Each milled sample was tableted in the open air on a tablet press (NP-RD10A, Natoli) with a 2.5 x 2.5 mm circular punch and die, set to 5 mm depth, and pressed with 17 kN of force (3.4 kN per tablet equal to a pressure of 173 μPA). Capsules were prepared by filling pre- made gelatine capsules with the same milled powder as the tablets. The capsules and tablets were then weighed, and their viability (CFU/g) was evaluated as described in “Viability Throughout Different Manufacturing Processes”. L. crispatus-loaded oleogels were prepared by heating coconut oil to 37^oC before adding oleic acid to a final concentration of 1% (v/v). The lyophilized products were then added to the warm oleogel at a concentration of 100 mg/mL and thoroughly vortexed to combine. The resultant oleogel was allowed to cool and solidify before it was stored aerobically at 37^oC for 1 h. Five 10-fold dilutions were prepared for each oleogel sample before 20 μL of each dilution as well as the undiluted oleogel samples were plated on MRS agar in 100 mm three-segment petri dishes. The agar plates were incubated and the viability (CFU/g) was evaluated as described in “Viability Throughout Different Manufacturing Processes”. Viability and Release from Oleogel-Loaded IVR The final oleogel-loaded IVRs were designed in Fusion 360 and printed on a Form 4 DLP Formlabs printer with the biocompatible Formlabs BioMed Flex^TM 80A resin. The holes of the IVRs were covered in parafilm before 5 mL of L. crispatus lyophilized in Formulation A and suspended in the oleic acid oleogel at a concentration of 500 mg/mL was loaded into each IVR. The loaded IVRs were set at 4^oC for 20 min before being placed in a 100 mm circle petri dish and stored submerged in 20 mL of SVF composed of 3.510 g/L sodium chloride, 1.400 g/L potassium hydroxide, 0.222 g/L calcium hydroxide, 0.018 g/L bovine serum albumin, 2.000 g/L lactic acid, 1.000 g/L acetic acid, 0.160 g/L glycerol, 0.400 g/L urea and 5.000 g/L glucose adjusted to pH 4.6 (43). Simultaneously, the same volume of oleic acid oleogel loaded with 500 mg/mL of L. crispatus lyophilized in Formulation A was suspended in 20 mL of SVF in a glass scintillation vial. At each timepoint, the IVRs were transferred to a new petri dish containing 20 mL of SVF and aerobically stored at 37^oC. Additionally, a 20 μL sample was taken from each glass scintillation vial. The L. crispatus viability present in the SVF of all samples was evaluated as described in “Vial-Based Validation of Lyoprotective Hits”. In vivo Evaluation of Safety and Efficacy of Formulation A Doses of L. crispatus were prepared as described in Methods and Materials S12.1. Two vaginal swabs (BD Eswabs, Becton Dickinson) and a blood sample for each sheep was collected prior to the commencement of each study. Following baseline sampling, 5 mL of metronidazole was applied daily via an applicator to the vaginal tract of each sheep for two days. On the third day of the study, two vaginal swabs (BD Eswabs^TM Becton Dickinson) and a blood sample for each sheep was conducted. Three sheep were administered with the dried Formulation A as a powder and three sheep were administered with the Formulation A- loaded oleogel. Vaginal swabs were then collected 1 h, 2 h, 4 h, 8 h, 24 h, 48 h, 5 days, 7 days, 10 days, and 14 days post-administration. A final blood sample was also collected from all sheep 14 days post-L. crispatus administration. After a 2-week washout period, this study was repeated except three sheep were administered lyophilized L. crispatus with no formulation, three sheep were administered the L. crispatus lyophilized in Formulation A, and three sheep were administered the L. crispatus lyophilized in Formulation A and suspended in the oleic acid oleogel. For the daily dosing study, the same antibiotic dosing and blood collection schedule was followed. After the antibiotic course, one group of three sheep were administered 2x10⁹ CFU of L. crispatus lyophilized in Formulation A and suspended in the oleic acid oleogel daily for seven days. Another group of three sheep were administered one L. crispatus-loaded IVR. Swabs for this study were taken at the same time for both groups. For the daily dosing group, with the exception of the 1 h and 8 h samples, swabs were taken immediately prior to administration of the next L. crispatus dose. After seven days the IVRs were removed. Throughout all studies, the vaginal swabs were stored at 4^oC until the completion of the study when DNA extraction and qPCR analysis for all samples could be conducted as described in Supplementary Methods and Materials S12.2 and 12.3. Blood samples were stored at 4^oC for <24 h before being sent for analysis. Statistical Analysis GraphPad Prism^TM 10 software (GraphPad Software) was used for statistical analysis. Statistical significant was assessed by Turkey’s multiple comparison test. All data was expressed at mean ± SD with a minimum of five replicates conducted unless otherwise stated in figure legends. Example 2. Bifidobacterium Introduction Although B. infantis shows therapeutic potential for various neonatal and paediatric disorders, its clinical translation has been hampered by poor bacterial viability during manufacturing, transport, and storage, especially in environments lacking cold-chain infrastructure. Here, we developed a two-excipient formulation that stabilizes B. infantis through lyophilization and provides exceptional viability under extreme conditions— including >5 months of storage at 60^oC. This extremophile-like stabilization allowed B. infantis to withstand conventional manufacturing stressors and enabled its incorporation into multiple delivery systems, including capsules, tablets, oleogels and a novel oleogels. This work establishes a robust, low-cost strategy for enhancing the stability, scalability, and accessibility of B. infantis-based therapeutics and supports expanded distribution to low- resource populations. Results High-Throughput Screen for Lyoprotective Formulation A high-throughput screen in which FDA-approved Generally Recognized as Safe (GRAS) compounds was adapted for B. infantis to identify excipients that could maintain viability after lyophilization and eliminate reliance on cold-chain practices. Considering the minimal viability of commercial Bifiobacterium products, an expanded excipient library that consisted of 479 FDA-approved GRAS compounds, food additives and compounds previously used for oral administration were investigated. Three different concentrations of each excipient, resulting in 1437 unique formulations, were investigated for lyoprotective properties against B. infantis. Upon completing the screens, multiple excipients were identified to perform equivalently or better than the common commercial lyoprotectant, trehalose. A normalized viability cut-off value of 0.8 was then applied, resulting in 60 total excipients being selected for further large-scale validation. Expectantly, sugars represented the greatest proportion of the selected excipients. Upon validation, all lyoprotective excipients identified in the high- throughput screen demonstrated greater viability after lyophilization than the absence of excipients, with 32 excipients indeed performing equivalently or better than the trehalose control. Ultimately, the three best performing excipients were sucrose, magnesium D- gluconate and maltodextrin. Complexing these primary excipients with the excipient library resulted in additional lyoprotective and long-term stability. Three additional high-throughput screens were conducted with each primary excipient, resulting in an additional 1437 unique formulations examined, for a total of 2874 formulations investigated overall. These screens identified 207 combinatorial formulations that demonstrated greater viability after lyophilization than each primary excipient alone. A greater diversity in chemical properties amongst the secondary excipient hits was observed in comparison to the primary excipient hits. Considering the significant number of combinatorial formulations that outperformed the primary excipients and the ultimate aim of eliminating cold-chain practices for live biotherapeutic products, a “commercial feasibility filter” was applied to the excipients in order to identify the most feasible combinatorial formulations to pursue. This filter consisted of four criteria. Firstly, due to the global demand for B. infantis-based products, the allergies and dietary preferences of patients were considered, and all animal-derived excipients or common allergens were removed. Then, in order to simplify production costs, excipients that consisted of medias, broths and extracts that would likely differ between batches were removed. All excipients that were not stable at room temperature were also removed in an attempt to bias the final combinatorial formulations for greater stability at extreme temperatures. Finally, only excipients that were inexpensive (<1.00 USD/mL) to purchase at high purity in bulk-scale were selected in order to further reduce costs. Applying this filter resulted in 138 unique combinatorial formulations selected for validation. To ensure the lyoprotective combinatorial formulations also enabled long-term viability in the absence of cold-chain practices, the lyophilized products were aerobically stored at 60^oC for 24 h before viability was assessed. Implementing this additional selective pressure resulted in the identification of 17 combinatorial formulations that demonstrated greater B. infantis viability after lyophilization and high temperature storage than the primary excipients alone. The top four performing combinatorial formulations were all from the same library in which magnesium D-gluconate was complexed with either -alanine, L-histidine, L-alanine or -aminobutyric acid. An additional improvement in viability after aerobic storage at 60^oC for 24 h could also be achieved for each combinatorial formulation through further assessment of each excipient concentration (Fig.8). For example, concentrations between 1.020-10.2% magnesium D-gluconate and 0.365-3.65% L-histidine were observed to be required in order to provide a lyoprotective effect to B. infantis (Fig.11). The data were plotted in a heat map as shown in Figure 11, showing those concentrations and ratios of magnesium D-gluconate and L-histidine that provided a surprisingly significant increase in lyoprotective and thermal stability for B. infantis. In order to determine whether long-term stability without refrigeration could be achieved, the stability of the top three, combinatorial formulations was examined at 60^oC and 37^oC for 6 months in comparison to a representative commercial formulation. Considering pasteurization protocols begin at 60^oC, B. infantis was not expected to maintain viability at this extreme temperature. However, all combinatorial formulations outperformed the representative commercial control, with the best performing formulation, (i.e. Formulation B), maintaining detectable viability for >5 months after storage at 60^oC (Fig.9A). In the same conditions, the representative commercial formulation had no detectable viability after 1 week. This improvement in long-term stability was also observed at 37^oC (Fig.9B). Visual examination of the lyophilized products after 6 months of storage at 60^oC highlighted a lack of burning in Formulation B in comparison to the sugar-based representative commercial formulation. Additionally, scanning electron microscopy (SEM) of the samples revealed that lyophilization with Formulation A resulted in a matrix structure with little surface exposed bacteria, explaining the prolonged stability observed (Fig.10A & B). Without being bound by a mechanism of action, the matrix is believed to be a hydrogen-bonding and electrostatic interaction network of magnesium D-gluconate and L-histidine, leading to significantly improved lyoprotection of B. infantis. Then, in the absence of sugars, the formulation is able to maintain integrity at high temperatures, resulting in the novel extremophile-like stability that has never previously been achieved and could potentially abolish the need for cold-chain practices. Extremophilic Formulation Provided an Opportunity to Investigate Alternative Delivery Methods Since Formulation B consists of only two excipients, it is significantly less dense than commercial formulations resulting in far greater viability per gram of lyophilized product (Fig.11 & 12). The formulation also demonstrated compatibility with traditional manufacturing processes. Many manufacturing methods are often considered too harsh for bacteria as milling involves shear force and wet granulation requires organic solvent exposure. However, the unique stability provided by Formulation B translated to improved viability during these processes. Milling and wet granulation of Formulation B resulted in uniform powders with minimal surface exposed bacteria. The Formulation B product demonstrated and maintained a ~1.5-log improvement in viability in comparison to the representative commercial formulation throughout the milling and wet granulation process (Fig.12). The extremophilic formulation was then examined in traditional and novel methods of delivery that have previously been investigated for B. infantis. B. infantis-loaded capsules, tablets and gels have all been investigated clinically but also involve additional stressors such as compression, hydrophobicity or baking. However, the lyophilized Formulation B product could successfully be incorporated into each delivery method with little detectable impact on viability (Fig.12). Discussion In this study, we have successfully adapted and expanded upon a high-throughput screen in order to identify a novel stabilizing formulation for B. infantis. Lactic acid expressing bacteria have many industrial, agricultural and pharmaceutical applications. However, very limited investigation into stabilizing B. infantis for neonatal health applications has been conducted. Of the formulations that have been developed for Bifidobacterium sp. all contain a sugar-based excipient that will inevitably burn or ferment at higher temperatures. As a result, the stability of these formulations is often only examined at a maximum of 25^oC for 9 months. Sugars are also generally hygroscopic leading to the rehydration of lyophilized bacteria in high humidity and a significant loss in viability. In the absence of sugars, the formulation is able to maintain integrity at high temperatures, resulting in the novel extremophile-like stability that has never previously been achieved and can abolish the need for cold-chain practices. When considering how to best administer this extremophilic formulation, there is significant discourse and ambiguity regarding the optimal method of delivering Bifidobacterium to neonates. However, the formulation disclosed herein can be observed to demonstrate stability and compatibility with all traditional and new methods of delivery. Methods and Materials High-Throughput Screen for Lyoprotective Formulation B. infantis was cultured and the compound library was prepared at described in Supplementary Methods and Materials S2.1. and S2.2., respectively. For the initial high- throughput screen, B. infantis suspended in PBS to an OD600 of 3 was combined with the arrayed library in a 1:1 ratio to a final volume of 200 μL in a U-bottom 96-well plate using a liquid handler robot (OpentronFlex^TM, Opentron). In each plate, a positive control well of trehalose as well as a negative control well of water was included. For the combinatorial screen, the primary excipients (sucrose, magnesium D-gluconate or maltodextrin) were prepared at 80% their solubility in water and complexed in a 1:1 ratio with the excipient library to a final volume of 100 μL in a U-bottom 96-well plate using a liquid handler robot (OpentronFlex^TM, Opentron). The B. infantis culture suspended in PBS was added to the complexed library to a final volume of 200 μL/well. All high-throughput screens were conducted in triplicate. Once prepared, the arrayed B. infantis+material plates were covered with parafilm and immediately placed at -25^oC to freeze for 2–4 h. The frozen plates were transferred to a tray freeze dryer (FreeZone^TM Stoppering Tray Dryer, Labconco) and dried at 25^oC and 0.1 mBar for 15 h total. The dried wells for each plate were rehydrated in 200 μL of PBS using the liquid handler robot (OpentronFlex^TM, Opentron). Three 10-fold serial dilutions for the initial high-throughput screen or four 10-fold serial dilutions for the combinatorial high-throughput screen were conducted before 4 μL of the most dilute samples were spotted onto MRS agar in a 1-well rectangular plate. The rectangular agar plates were incubated anaerobically (Vinyl Type A Anaerobic Chamber, Coy Lab Products) at 37^oC for 48 h before the plates were imaged in an iBright^TM Imaging System (ThermoFisher Scientific) using a standard protein gel light protocol. An automated colony counting protocol was implemented as described in Supplementary Methods and Materials S3 to identify lyoprotective hits (20). Vial-Based Validation of Lyoprotective Hits The best performing excipients from the initial lyoprotective high-throughput screen were validated at a larger scale to determine a more accurate relative viability value (CFU/mL). Results are shown in Figure 11A (without cells) and 11B (with cells). B. infantis suspension in PBS was prepared as described in Supplementary Methods and Materials S2.1. before being combined in a 1:1 ratio with 1 mL of each excipient hit prepared at the lyoprotective concentration identified in the high-throughput screen. The final B. infantis- compound vials were vortexed briefly before immediately placed at -25^oC for 2–4 h. Once frozen, the vials were transferred to a tray freeze dryer (FreeZone^TM Stoppering Tray Dryer, Labconco) and dried at 25^oC and 0.1 mBar for 15 h total. The viability after lyophilization for each excipient was assessed by rehydrating the samples with 2 mL of PBS. Five 10-fold serial dilutions were prepared before 20 μL of each dilution and the vial sample were plated onto MRS agar in two 100 mm three-segment circle petri dishes. The agar plates were incubated anaerobically (Vinyl Type A Anaerobic Chamber, Coy Lab Products) for 48 h at 37^oC. Colonies were counted at the dilution that provided ~10-200 distinct colonies and the CFU/mL for each excipient was calculated. The lyoprotective capacity of all excipient hits were evaluated using five biological replicates. Each two-compound combinatorial formulation was prepared in 20 mL glass scintillation vials to a final volume of 1 mL. The B. infantis was prepared as described in Supplementary Methods and Materials S2.1. before 1 mL was added to all combinatorial formulation vials and they were immediately stored at -25^oC for 2–4 h. Once frozen, the vials were transferred to a tray freeze dryer (Labconco, FreeZone^TM Stoppering Tray Dryer) and dried at 25^oC and 0.1 mBar for 15 h total. The dried material vials were packed into zip-loc bags with a desiccant pack and stored aerobically in the dark at 60^oC. After 24 h, the viability in the vials was assessed as described for the initial lyoprotective excipient hits. To determine the optimal concentration for each of the excipients in the three best performing combinatorial formulations, sterile stocks of magnesium D-gluconate and each secondary excipient (i.e. -alanine, L-histidine, L-alanine or -aminobutyric acid) were prepared at 80% their solubility in water. Different ratios of the magnesium D-gluconate stock and each secondary excipient stock were prepared into 20 mL glass scintillation vials. Sterile water was added for a final volume of 1 mL in each vial. B. infantis was then prepared as described in Supplementary Methods and Materials S2.1. before 1 mL of the culture was added to each vial and all samples were immediately stored at -25^oC for 2–4 h. Once frozen, the vials we lyophilized, stored at 60^oC and their viability was evaluated after 24 h as previously described for the lyoprotective combinatorial formulation hits. Compatibility of the Formulation with Various Manufacturing and Delivery Processes Viability Throughout Different Manufacturing Processes An B. infantis suspension in PBS was prepared as described in Supplementary Methods and Materials S2.1. except the bacteria was suspended at an OD₆₀₀ of 12. The suspensions were then combined in a 1:1 ratio with either Formulation B, the representative commercial formulation or no formulation, before being immediately stored at -25^oC for 2– 4 h. Once frozen, the vials were lyophilized as previously described in “Vial-Based Validation of Lyoprotective Hits”. A proportion of the lyophilized B. infantis powder were then mixed with the following excipients: 1% (w/w) magnesium stearate, 4% (w/w) polyvinylpyrrolidone, and 65% (w/w) lactose, for a final loading of 30% (w/w) of the different lyophilized B. infantis samples. This milled powder proceeded to wet granulation. Wet granulation was conducted by first adding 300 μL of isopropanol per g of milled product. The mixture was vortexed for 5 min to evenly combine before the resulting paste was passed through a bench top oscillating granulator (FGS II, ERWEKA) with a 1 mm mesh screen. The collected granules were dried at 60^oC for 30 min in 20 mL glass scintillation vials. The lyophilized samples before and after aerobic storage at 60^oC for 24 h as well as the milled and dried granule samples were weighed, before 2 mL of PBS was added to all samples. Five 10-fold dilutions were prepared for each sample before 20 μL of each dilution as well as the undiluted samples were plated on MRS agar in two 100 mm three-segment petri dishes. The agar plates for all samples were incubated anaerobically (Vinyl Type A Anaerobic Chamber, Coy Lab Products) for 48 h at 37^oC and colonies were counted at the dilution that provided ~10-200 distinct colonies before the CFU/g for each sample was calculated by dividing the mass of only the bacterial fraction. Viability in Traditional Methods of Delivery Lyophilized B. infantis samples in Formulation B, the representative commercial formulation or no formulation were prepared and milled as described in the above subsection. Each milled sample was tableted in the open air on a tablet press (NP-RD10A, Natoli) with a 2.5 x 2.5 mm circular punch and die, set to 5 mm depth, and pressed with 17 kN of force (3.4 kN per tablet equal to a pressure of 173 μPA). Capsules were prepared by filling pre- made gelatine capsules with the same milled powder as the tablets. The capsules and tablets were then weighed, and their viability (CFU/g) was evaluated as described in “Viability Throughout Different Manufacturing Processes”. B. infantis-loaded oleogels were prepared by heating either sesame or safflower oil to to ~60^oC before either beeswax or carnauba wax to a final concentration of 3% (v/v). The lyophilized products were then added to the warm oleogel at a concentration of 100 mg/mL and thoroughly vortexed to combine. The resultant oleogel was allowed to cool and solidify before it was stored aerobically at 37^oC for 1 h. Five 10-fold dilutions were prepared for each oleogel sample before 20 μL of each dilution as well as the undiluted oleogel samples were plated on MRS agar in 100 mm three-segment petri dishes. The agar plates were incubated and the viability (CFU/g) was evaluated as described in “Viability Throughout Different Manufacturing Processes”. References and Notes 1. J. Ravel et al., Vaginal microbiome of reproductive-age women. Proceedings of the National Academy of Sciences 108, 4680-4687 (2011). 2. J. M. Fettweis et al., Differences in vaginal microbiome in African American women versus women of European ancestry. Microbiology 160, 2272-2282 (2014). 3. N. Fredricks David, L. Fiedler Tina, M. Marrazzo Jeanne, Molecular Identification of Bacteria Associated with Bacterial Vaginosis. New England Journal of Medicine 353, 1899-1911. 4. J. M. Marrazzo, Interpreting the epidemiology and natural history of bacterial vaginosis: Are we still confused? Anaerobe 17, 186-190 (2011). 5. E. H. Koumans et al., The Prevalence of Bacterial Vaginosis in the United States, 2001–2004; Associations With Symptoms, Sexual Behaviors, and Reproductive Health. Sexually Transmitted Diseases 34, (2007). 6. J. Atashili, C. Poole, P. M. Ndumbe, A. A. Adimora, J. S. Smith, Bacterial vaginosis and HIV acquisition: a meta-analysis of published studies. AIDS 22, (2008). 7. M. Shimaoka et al., Association between preterm delivery and bacterial vaginosis with or without treatment. Sci Rep 9, 509 (2019). 8. J. Ravel, I. Moreno, C. Simón, Bacterial vaginosis and its association with infertility, endometritis, and pelvic inflammatory disease. American Journal of Obstetrics & Gynecology 224, 251-257 (2021). 9. A. E. Stapleton et al., Randomized, Placebo-Controlled Phase 2 Trial of a Lactobacillus crispatus Probiotic Given Intravaginally for Prevention of Recurrent Urinary Tract Infection. Clinical Infectious Diseases 52, 1212-1217 (2011). 10. R. Chen et al., Probiotics are a good choice for the treatment of bacterial vaginosis: a meta-analysis of randomized controlled trial. Reproductive Health 19, 137 (2022). 11. Y. Gu et al., Efficacy and safety of different drugs for the treatment of bacterial vaginosis: a systematic review and network meta-analysis. Frontiers in Cellular and Infection Microbiology 14, (2024). 12. C. S. Bradshaw et al., High Recurrence Rates of Bacterial Vaginosis over the Course of 12 Months after Oral Metronidazole Therapy and Factors Associated with Recurrence. The Journal of Infectious Diseases 193, 1478-1486 (2006). 13. C. A. Muzny, J. D. Sobel, Understanding and Preventing Recurring Bacterial Vaginosis: Important Considerations for Clinicians. Int J Womens Health 15, 1317- 1325 (2023). 14. F. Wu et al., Improvement of vaginal probiotics Lactobacillus crispatus on intrauterine adhesion in mice model and in clinical practice. BMC Microbiology 23, 78 (2023). _{15. M. Strus, M. Brzychczy-W och, T. Gosiewski, P. Kochan, P. B. Heczko, The in vitro} effect of hydrogen peroxide onvaginal microbial communities. FEMS Immunology & Medical Microbiology 48, 56-63 (2006). 16. J. Kaewsrichan, K. Peeyananjarassri, J. Kongprasertkit, Selection and identification of anaerobic lactobacilli producing inhibitory compounds against vaginal pathogens. FEMS Immunology & Medical Microbiology 48, 75-83 (2006). 17. Y. Kong et al., Protective Effects of Engineered Lactobacillus crispatus on Intrauterine Adhesions in Mice via Delivering CXCL12. Frontiers in Immunology Volume 13 - 2022, (2022). 18. E. Armstrong et al., Sustained effect of LACTIN-V (<em>Lactobacillus crispatus</em> CTV-05) on genital immunology following standard bacterial vaginosis treatment: results from a randomised, placebo-controlled trial. The Lancet Microbe 3, e435-e442 (2022). 19. J. Lyu et al., From whole genomes to probiotic candidates: A study of potential lactobacilli strains selection for vaginitis treatment. Heliyon 10, e30495 (2024). 20. M. Jimenez et al., Synthetic extremophiles: Species-specific formulations for microbial therapeutics and beyond. bioRxiv, 2022.2011.2030.518573 (2022). 21. N. C. A. Juliana et al., The Association Between Vaginal Microbiota Dysbiosis, Bacterial Vaginosis, and Aerobic Vaginitis, and Adverse Pregnancy Outcomes of Women Living in Sub-Saharan Africa: A Systematic Review. Front Public Health 8, 567885 (2020). 22. T. P. Parks, A. Macrobal. (OSel, Inc., United States, 2021). 23. J.-P. Menard, F. Fenollar, M. Henry, F. Bretelle, D. Raoult, Molecular Quantification of Gardnerella vaginalis and Atopobium vaginae Loads to Predict Bacterial Vaginosis. Clinical Infectious Diseases 47, 33-43 (2008). 24. L. Lehtoranta et al., Characterization of vaginal fungal communities in healthy women and women with bacterial vaginosis (BV); a pilot study. Microbial Pathogenesis 161, 105055 (2021). 25. M. Zhu et al., Vaginal Lactobacillus fatty acid response mechanisms reveal a metabolite-targeted strategy for bacterial vaginosis treatment. Cell 187, 5413- 5430.e5429 (2024). 26. O. Reichman, R. Akins, J. D. Sobel, Boric Acid Addition to Suppressive Antimicrobial Therapy for Recurrent Bacterial Vaginosis. Sexually Transmitted Diseases 36, (2009). 27. A. van der Straten et al., Evaluation of 3 approaches for assessing adherence to vaginal gel application in clinical trials. Sex Transm Dis 40, 950-956 (2013). 28. R. Sánchez-Borrego et al., Satisfaction and medication adherence in women with vulvovaginal atrophy: the CRETA. Climacteric 26, 437-444 (2023). 29. K. Oddsson et al., Efficacy and safety of a contraceptive vaginal ring (NuvaRing) compared with a combined oral contraceptive: a 1-year randomized trial. Contraception 71, 176-182 (2005). 30. A. R. Thurman, C. M. R., H. J. A., G. F. and Doncel, Intravaginal rings as delivery systems for microbicides and multipurpose prevention technologies. International Journal of Women's Health 5, 695-708 (2013). 31. K. T. Barnhart et al., Baseline dimensions of the human vagina. Hum Reprod 21, 1618-1622 (2006). 32. J. M. McCracken et al., Animal Models and Alternatives in Vaginal Research: a Comparative Review. Reproductive Sciences 28, 1759-1773 (2021). 33. S. Davis, K. Parthun, D. R. Friend, A nine-month repeat-dose intravaginal ring (Ovaprene) irritation study in sheep. Contraception 133, 110387 (2024). 34. R. F. Stefanello et al., Survival and stability of Lactobacillus fermentum and Wickerhamomyces anomalus strains upon lyophilisation with different cryoprotectant agents. Food Research International 115, 90-94 (2019). 35. A. Bodzen et al., Design of a new lyoprotectant increasing freeze-dried Lactobacillus strain survival to long-term storage. BMC Biotechnology 21, 66 (2021). 36. K. Izutsu et al., Freeze-drying of proteins in glass solids formed by basic amino acids and dicarboxylic acids. Chem Pharm Bull (Tokyo) 57, 43-48 (2009). 37. J. M. Bohbot et al., Efficacy and safety of vaginally administered lyophilized Lactobacillus crispatus IP 174178 in the prevention of bacterial vaginosis recurrence. Journal of Gynecology Obstetrics and Human Reproduction 47, 81-86 (2018). 38. E. Armstrong et al., Vaginal Lactobacillus crispatus persistence following application of a live biotherapeutic product: colonization phenotypes and genital immune impact. Microbiome 12, 110 (2024). 39. M. Barba et al., Description of the vaginal microbiota in nulliparous ewes during natural mating and pregnancy: preliminary signs of the male preputial microbiota modulation. Front Microbiol 14, 1224910 (2023). 40. J. J. Quereda et al., Vaginal Microbiota Changes During Estrous Cycle in Dairy Heifers. Frontiers in Veterinary Science Volume 7 - 2020, (2020). 41. K. Vomstein et al., Uterine microbiota plasticity during the menstrual cycle: Differences between healthy controls and patients with recurrent miscarriage or implantation failure. Journal of Reproductive Immunology 151, 103634 (2022). 42. K. L. Nunn, L. J. Forney, Unraveling the Dynamics of the Human Vaginal Microbiome. Yale J Biol Med 89, 331-337 (2016). 43. M. Falavigna et al., The Vaginal-PVPA: A Vaginal Mucosa-Mimicking In Vitro Permeation Tool for Evaluation of Mucoadhesive Formulations. Pharmaceutics 12, (2020). 44. A. R. Kirtane et al., Development of oil-based gels as versatile drug delivery systems for pediatric applications. Science Advances 8, eabm8478. 45. C. F. McCoy et al., Mechanical testing methods for drug-releasing vaginal rings. International Journal of Pharmaceutics 559, 182-191 (2019). 46. K. DeLong et al., Conceptual Design of a Universal Donor Screening Approach for Vaginal Microbiota Transplant. Frontiers in Cellular and Infection Microbiology Volume 9 - 2019, (2019). Supplementary Materials Supplementary Methods and Materials S1. Viability of Commercial Probiotics Within one week of arrival, 2 mL of sterile phosphate-buffer saline (PBS) was added to the recommended dosage of each Lactobacillus-based product. The dose was thoroughly suspended before eight 10-fold serial dilutions were performed. A 20 μL sample of each dilution and the undiluted sample were plated on MRS agar across three 100 mm three- segment circle petri dishes. The agar plates were incubated anaerobically (Vinyl Type A Anaerobic Chamber, Coy Lab Products) for 48 h at 37^oC and colonies were counted at the dilution that provided ~10-200 distinct colonies in order to calculate the total CFU present in each dose. This process was repeated with five different doses over two different batches of each commercial product. S.2. L. crispatus and Excipient Handling S2.1. Bacterial Culturing and Preparation L. crispatus (33820) was obtained from ATCC and cultured anaerobically at 37^oC in MRS broth. All cultures were prepared from L. crispatus glycerol stocks stored at -80^oC. Before each experiment, L. crispatus was precultured anaerobically in 10 mL of MRS broth at 37^oC for 48 h. The overnight culture was then diluted into 90 mL of fresh MRS broth in a flask and cultured anaerobically for another 24 h at 37^oC. The optical density (OD₆₀₀) of each culture was recorded (typically ~6–8) and the cells were pelleted at 4,500 rpm for 15 min. The cell pellets were washed in PBS before the final cell pellets were then resuspended in PBS to a final OD600 of 1.5 and used immediately. S2.2. Excipient Library Preparation The excipient library was prepared by solubilizing each material in sterile, ultrapure water to a concentration of 80% their solubility in water. For those materials that are not soluble in water a suspension of 0.8% (w/v or v/v) was prepared. The same concentration was also used for gums such as pectin, which were too viscous to pipette at 80% their solubility in water. The material solutions were aliquoted into sealed 96-well flat-bottom plates, leaving the border wells vacant, and stored at -25^oC until needed. On the day of use, the material plates were thawed at room temperature for 1 h prior to conducting the high-throughput screen. S3. Data Analysis for Identification of Lyoprotective Formulations An automated colony counting protocol was implemented as previously described by Jimenez et al (20). The viability scores were then calculated by subtracting any colony counts in the negative control from the colony counts of each excipient as well as the positive control. The adjusted colony count for each excipient was then divided by the adjusted colony count of the positive control to determine the normalized viability score for each combinatorial formulation or excipient (Viability Score = Countexcipient- negative control/Countpositive control-negative control). An arbitrary cut-off value of 0.8 for the initial high- throughput screen and >1.0 for the combinatorial screen was used for determining hits, before applying the commercial feasibility filter. S4. Long-Term Stability of Bacteria Lyophilized in the Best Formulations For each best combinatorial formulation, 1 mL was added to triplicate vials for each timepoint. Simultaneously, this was repeated with water (i.e. no formulation) and a representative commercial formulation that consisted of 15% (w/v) trehalose, 6% (w/v) xylitol, 1% (w/v) sodium ascorbate, 10 mM of NaPO₄ pH 7.8, and 5 mM monosodium glutamate (22). The L. crispatus culture was then prepared as described in Supplementary Methods and Materials S2.1. before 1 mL was added to all vials and the samples were immediately stored at -25^oC for 2–4 h. Once frozen, the vials were lyophilized, stored at either 60^oC or 37^oC and, at each timepoint, the viability of the corresponding vials was evaluated as previously described in “Vial-Based Validation of Lyoprotective Hits.” S5. Data Analysis for the Identification of Suitable Pathogen Inhibitory Compounds Each microbial culture was added to five 2-fold dilutions of selected compounds in triplicate as described in the “High-Throughput Screen for Pathogen Inhibitors”. Results were scored by measuring absorbance of all samples at 600 nm using an Infinite 200 Pro plate reader (Tecan Life Sciences) after incubation for 24–48 h and were confirmed visually. All experiments were conducted in triplicate where the positive controls contained an applicable antimicrobial and the negative controls contained broth with no additional compounds. A control plate containing the broth and compounds without the addition of the microbes was also prepared simultaneously during each high-throughput screen and MIC assay. S6. Compatibility of Combinatorial Formulations with Large-Scale Concentrations of L. crispatus To determine the concentration of L. crispatus that could be loaded into the combinatorial formulation and maintain the novel long-term stability at high temperatures an L. crispatus suspension was prepared as previously described in “Bacterial Culturing and Preparation” with some small changes. After the bacteria was washed with PBS, the cell pellet was resuspended in PBS to a OD600 of 30. This suspension of L. crispatus was then used in a variety of serial dilutions to produce L. crispatus suspensions at an OD600 of 30, 25, 20, 15, 12, 9, 6, 3, and 1.5. In 20 mL glass scintillation vials, 1 mL of Formulation A was added. Twelve vials were prepared for each OD600 of L. crispatus and 1 mL of the respective L. crispatus suspension in PBS was added to each vial before all samples were stored immediately at -25^oC for 2–4 h. Once frozen, the vials were transferred to a tray freeze dryer (Labconco, FreeZone Stoppering Tray Dryer) and dried at 25^oC and 0.1 mBar for 15 h total. The initial viability and the viability after 24 h storage at 60^oC was evaluated as described previously in “Vial-Based Validation of Lyoprotective Hits”. S7. L. crispatus Tablet Formulation L. crispatus lyophilized in Formulation A, was prepared as described in “Bacterial Culturing and Preparation” except the bacteria was suspended in PBS at an OD600 of 15. For tableting, the lyophilized products were milled with each of the different tablet formulations described in Table 4, for a final loading of 30% (w/w) of lyophilized L. crispatus. Each milled sample was then tableted in the open air on a tablet press (Natoli, NP-RD10A) with a 2.5 x 2.5 mm circular punch and die, set to 5 mm depth, and pressed with 17 kN of force (3.4 kN per tablet equal to a pressure of 173 μPA). To determine the initial viability, three of each tablet were weighed before they were suspended in 2 mL of PBS. Five 10-fold serial dilutions were prepared before 20 μL of each dilution and the vial samples were plated on MRS agar in two 100 mm three-segment circle petri dishes per tablet. The remaining tablets for each formulation were all weighed and divided evenly. The first group of tablets were suspended in 2 mL of SVF, while the other group remained dry. All groups were then stored at 37^oC for 24 h. Once the incubation was complete the dry tablets were suspended in 2 mL of PBS before the viability of all samples was evaluated as described in “Viability in Traditional Methods of Delivery”. Table 4. The different table formulations examined for improving L. crispatus viability. S8. L. crispatus Oleogel Formulation Several gelling agents were selected to examine their capacity to form oleogels and maintain L. crispatus viability. Oleic acid and linoleic acid were selected because they are both fatty acids that have been identified to inhibit vaginal pathogen growth and promote L. crispatus colonization (25). The same properties were identified for itaconic acid and octanoic acid in the vaginal pathogen high-throughput screen, thus they were included as potential gelling agents as well. Finally, whilst they both do not meet the commercial feasibility filter, beeswax and carnauba were also included as gelling agents to investigate as they have previously been identified as the best excipients for oleogel formulation (44). To measure the gelling capacity of these gelling agents they were mixed with coconut oil at 37^oC in increasingly higher concentration. The mixtures were then heated to 5–10^oC above each gelling agent’s melting point, before being cooled to room temperature. Gel formulation was determined using a vial inversion test (44). Combinations that did not flow to the bottom of the vial were considered gels. The viability of L. crispatus was investigated using the concentration of each gelling agents that resulted in a gel with coconut oil at room temperature but lost its gelling capacity at physiological temperature (37^oC). L. crispatus-loaded oleogels were prepared by heating coconut oil to 37^oC before adding 1% (v/v) oleic acid. The lyophilized products were then added to the warm oleogel at a concentration of 100 mg/mL and thoroughly vortexed to combine. The resultant oleogel was allowed to cool and solidify before it was stored aerobically at 37^oC for 1 h. Five 10-fold dilutions were prepared for each oleogel sample before 20 μL of each dilution as well as the undiluted oleogel samples were plated on MRS agar in 100 mm three-segment petri dishes. The agar plates for all samples were incubated anaerobically for 48 h at 37^oC and colonies were counted at the dilution that provided 10-200 distinct colonies before the CFU/g for each oleogel was calculated by dividing the mass of only the bacterial fraction. S9. Long-Term Stability of Representative L. crispatus Products Dried L. crispatus in Formulation A was produced as described previously in “Bacterial Culturing and Preparation”, except the bacteria was suspended at an OD600 of 15. The lyophilized product was then mixed with 1% (w/w) magnesium stearate, 4% (w/w) hydroxypropylmethylcellulose, 10% (w/w) boric acid and 55% (w/w) cellulose, for a final loading of 30% (w/w) of lyophilized L. crispatus. The resultant milled sample was prepared into tablets or capsules as described in “Viability in Traditional Delivery Methods”. L. crispatus-loaded oleic acid oleogel was prepared as previously described by heating coconut oil to 37^oC before adding oleic acid to a final concentration of 1% (v/v). The lyophilized products were then added to the warm oleogel at a concentration of 100 mg/mL and thoroughly vortexed to combine. The final tablets, capsules and oleogel samples were stored aerobically at 37^oC. At each timepoint the viability of the products was assessed as described previously in “Viability in Traditional Methods of Delivery”. S10. Development of the Oleogel-Loaded Intravaginal Ring S10.1. Manufacturing of Oleogel-Release Test Devices The oleogel-release test devices consisted of three parts: a tube with evenly distributed holes, and two caps at the end of the tube. McMaster EVA tubing was used for the devices due to its established use in commercially available intravaginal rings. The tubes were 5 cm long in order to be completely submerged in 20 mL of SVF in a glass scintillation vial, and had inner and outer diameters of in and ¼ in, respectively. Release holes were introduced to the test devices by drilling with varying sizes of YG-1 Gold-P drill bits on a Sherline Model 4400 desktop lathe. Any excess lathed material was cleared with compressed air. The caps were designed using Fusion 360 and printed on a Stratasys Objet260 Connex3 Polyjet printer with MED610 resin. Once printed, the caps were thoroughly cleaned with 2% (v/v) sodium hydroxide solution and rinsed in water for 10 min in a VIVOHOME^TM Ultrasonic Cleaner. This removed any residual SUP706 material. S10.2. Examination of Release Kinetics from the Oleogel-Release Test Devices The oleogel-release test devices were wrapped in parafilm and filled with 300 μL of 1% oleic acid oleogel loaded with 500 mg/mL of L. crispatus lyophilized in Formulation A. Once filled, the test devices were capped and stored at 4^oC for 20 min to set the oleogel. Each test device was submerged in a glass scintillation vial filled with 20 mL of SVF and stored aerobically at 37^oC on an orbital shaken set at 100 rpm. At each timepoint, the test devices were removed from the scintillation vials and placed into new, sterile SVF. The SVF from the previous timepoint is then diluted and plated as described in “Vial-Based Validation of Excipient Hits” and the total CFU of L. crispatus released from the test devices at the associated time point was calculated. Control test devices in which no holes were present in the device were used with all experiments to monitor any bacterial release from the caps. S10.3. Construction of Intravaginal Ring Prototype The devices were designed in Fusion 360, and printed on a Form 4 DLP Formlabs printer with Formlabs BioMed Flex^TM 80A resin. This resin was chosen for its biocompatibility, flexible properties, and its capacity for iterative design. S10.4. Tensile Test of Final Intravaginal Ring Design The tensile properties of the intravaginal ring (N = 3) was determined using an Admet eXpert^TM 5952 fatigue testing system (Admet, Norwood, MA, USA) equipped with a 1000 lbf load cell, as previously described by McCoy, et al. (45). Custom-designed pin grips were employed to minimize localized stress concentrations within the intravaginal rings during testing. Each intravaginal ring was elongated at a constant crosshead speed of 500 ± 50 mm/min until fracture. Maximum tensile load (N) and extension at maximum load (mm) were recorded using the integrated MTESTQuattro software (Admet). Percent elongation at break (%) was calculated as follows: Percent elongation at break = Extension at break (mm)/Original grip separation (mm)*100 S10.5. Fatigue Test of Final Intravaginal Ring Design Fatigue properties were assessed by subjecting vaginal rings (N = 10) to cyclic compression testing using an Admet eXpert^TM 5952 fatigue testing system (Admet, Norwood, MA, USA), equipped with a 1000 lbf load cell and controlled via MTESTQuattro^TM software (45). Custom ring holders featuring rectangular grooves were 3D-printed from polylactic acid (PLA) using a Bambu Labs X1C printer and designed to simultaneously accommodate 10 intravaginal rings. The intravaginal rings were positioned vertically in grooves on the lower platform, and an upper plate with matching grooves aligned with the lower plate ensured consistent intravaginal ring placement with slight pre-compression. Rings underwent cyclic compression from 100% to approximately 25 ± 5% of their original outer diameter (OD). Each ring was compressed and released 1000 times at a test speed of 15 mm/s, with a compression distance of 30 mm. Following the 1000-cycle test, the intravaginal rings were visually inspected for visible deterioration. Outer diameters were measured within 15 min post-testing using ring gauges and expressed as percentage recovery of the original OD. S11. Cytotoxic Potential of Formulation A S11.1. Mammalian Cell Culturing and Preparation Both Vk2/E6E7 (CRL-2616) and End1/E6E7 (CRL-2615) were obtained from ATCC and cultured in Keratinocyte-Serum Free (KSF) medium supplemented with 0.1 ng/mL human recombinant epidermal growth factor (EGF), 0.05 mg/mL bovine pituitary extract, and 0.4 mM calcium chloride. The cells were cultured for no more than 20 passages before being seeded at a concentration of 1x10⁶ cells/well in a clear, sterile polystyrene flat-bottom 96-well plate for 24 h. S11.2. MTT Assay The MTT assay was conducted according to manufacturer’s instructions. Briefly, a stock of potassium gluconate, L-histidine or Formulation A was prepared in the supplemented KSF medium. The potassium gluconate and L-histidine stocks were prepared at 80% the excipients solubility in water. These stocks were also used to prepare Formulation A (i.e.1:1:1 ratio of potassium gluconate:L-histidine:complete medium). Nine 2-fold serial dilutions were prepared in the complete medium before 200 μL of each serial dilution was applied in triplicate to the cells seeded in the 96-well plates. Three high-mortality replicates (cells + 1% Triton X-100), three low-mortality replicates (cells + complete medium) and three background replicates (complete media) were included on all plates. The plates were incubated with the excipients for 24 h at 37^oC in 5% CO2 before the excipients were removed and all wells were washed with 200 μL of PBS. To examine the cytotoxic effect of the excipients, 10 μL of 0.5 mg/mL MTT solubilized in PBS was added to each well. Each plate was incubated for 4 h at 37^oC in 5% CO2 before 100 μL of acidified isopropanol was added to the wells and all plates were wrapped in foil and shaken at 100 rpm for 15 min to solubilize the MTT formazan product. The absorbance of all samples was measured at 600 nm using an Infinite^TM 200 Pro plate reader (Tecan Life Sciences). S11.3. Data Analysis Cell viability (%) was calculated by subtracting the average absorbance values of background replicates from the average absorbance value of treatment wells and low- mortality wells before dividing the adjusted treatment value from the adjusted low-mortality samples and multiplying by 100 (A_{treatment-background}/A_{average low-mortality control-background}x100). Five biological replicates were conducted for each cell line and three technical replicates for each were averaged. S12. Dosing and Quantification of L. crispatus in vivo S12.1. Sample Preparation Vehicle samples were prepared by freezing 400 mL of Formulation A for 2–4 h at -25^oC. Once frozen, the formulation was transferred to a tray freeze dryer (FreeZone Stoppering Tray Dryer, Labconco) and dried at 25^oC and 0.1 mBar for 15 h total. The dry formulation was weighed and manually disrupted to produce a uniform powder before 375 mg of powder was loaded into three vaginal applicators. The remaining lyophilized combinatorial formulation was suspended in 1% oleic acid oleogel at a concentration of 500 mg/mL. Three additional vaginal applicators were then filled with 750 μL of the Formulation A-loaded oleic acid oleogel. For L. crispatus doses, the lyophilized bacteria was prepared as described in “Viability Throughout Different Manufacturing Processes”. The dried bacteria was weighed and manually disrupted to produce a uniform powder. For the bacteria lyophilized in the absence of formulation, the equivalent mass for 2×10⁹ CFU was loaded into each intravaginal applicator. For the bacteria lyophilized in Formulation A, 375 mg of the powder was loaded into each vaginal applicator. The remaining L. crispatus-formulation powder was suspended in the 1% oleic acid oleogel at a concentration of 500 mg/mL. To create a dose of 2×10⁹ CFU, a vaginal applicator was filled with 750 μL of the L. crispatus-loaded oleic acid oleogel. Alternatively, 5 mL of the L. crispatus-loaded oleic acid oleogel was loaded into the IVRs as described previously in “Viability and Release from Oleogel-Loaded IVR”. All samples were prepared the day of administration and stored at 4^oC for 1 h during transit, before being administered to the respective group of sheep. S12.2. DNA Extraction from Vaginal Swabs The DNA from all vaginal samples was extracted within <1 week of the completion of the trial. To extract the microbial DNA, the BD Eswab fluid (360 μL) was combined with 432 μL of lysis buffer (1% Triton X-100, 20 mM Tris-HCl pH 8.0, 2 mM EDTA) with 20 mg/mL of lysozyme and incubated for 1 h at 37^oC. Following lysozyme treatment, DNA was extracted using the DNeasy® Blood and Tissue kit (QIAGEN®), including the addition of 48 μL of proteinase K and incubation for 10 min at 56^oC for pretreatment of Gram-positive bacteria. The final DNA for each sample was eluted in a final volume of 100 μL before 5 μL of 3 M sodium acetate pH 5.4 and 1 mL of 100% ethanol was added to all samples. The DNA was then precipitated overnight at -25^oC. After 16 h, all samples were spun for 20 min at 17,000 g and 4^oC. The ethanol was decanted from the DNA pellets of each sample before 1 mL of 70% ethanol was added to the tubes and centrifuged for a further 15 min. The supernatant for each sample, was once again removed and the final DNA pellets were air dried before being resuspended in 50 μL of ultrapure water. DNA concentration was measured using a NanoDrop^TM. To prepare an L. crispatus standard, a 10 mL liquid culture of the bacteria was prepared in MRS broth from a glycerol stock and incubated anaerobically at 37^oC for 48 h. The OD600 of the culture was measured before the culture was pelleted at 4500 rpm for 15 min. The supernatant was then removed before the cell pellet was resuspended in 10 mL of PBS and spun again at 4500 rpm for 15 min. The supernatant was removed again from the cell pellet before it was resuspended in 10 mL of PBS and ten 5-fold serial dilutions of the culture into PBS were prepared. For both the stock L. crispatus and its serial dilutions, 20 μL of each sample and dilution was plated onto MRS agar in 100 mm three-segment circle petri dishes. The agar plates were incubated anaerobically for 48 h at 37^oC and colonies were counted at the dilution that provided ~10-200 distinct colonies and the CFU/mL for each sample was calculated. The DNA from each L. crispatus stock was extracted following the same protocol as the vaginal samples, allowing DNA concentration to be correlated to relative CFU of L. crispatus. S12.3. Quantitative PCR Quantitative PCR (qPCR) was performed for both DNA extractions from vaginal swabs and cultured L. crispatus on Applied Biosystems, QuantStudio6 in accordance with the procedure outlined by DeLong et al (46). In brief, bacterial species-specific primers (Integrated DNA Technologies) for the 16S rRNA gene were used at a concentration of 100 nM. To standardize the qPCR, a fixed volume (2 μL) of DNA was used for cultured standards and swab samples. The protocol started with an initial incubation at 95^oC for 15 min, followed by 40 cycles with denaturing at 95^oC for 1 s, annealing at 60^oC for 20 s and extension at 72^oC for 20 s. Following qPCR, the cycle threshold (Ct) of the L. crispatus culture extraction dilutions were plotted against their CFU concentration. Linear fitting provided and equation that was then used to predict the CFU concentration in the sheep vaginal swab sample extraction.

Claims

We claim 1. A formulation, comprising one or more excipient selected from the group consisting of D-(-)-fructose, Bacto Tryptic Soy Broth (BTSB), potassium gluconate, magnesium D- gluconate, sucrose, and/or maltodextrin. The formulation of claim 1, wherein the one or more excipient comprises potassium gluconate or magnesium D-gluconate.

3. The formulation of claim 1 or 1A, further comprising L-histidine.

4. The formulation of any one of claim 1-3, wherein the formulation comprises or consists of potassium gluconate and L-histidine.

5. The formulation of claim 4, wherein the formulation comprises: (a) between about 0.4% (w/v) and about 4% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 3.65% (w/v) L-histidine; or (b) between about 0.4% (w/v) and about 2% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (c) between about 0.4% (w/v) and about 2% (w/v) potassium gluconate and between about 0.73% (w/v) and about 1.825% (w/v) L-histidine; or (d) between about 0.4% (w/v) and about 2% (w/v) potassium gluconate and between about 0.73% (w/v) and about 1.217% (w/v) L-histidine; or (e) between about 0.4% (w/v) and about 1.5% (w/v) potassium gluconate, and between about 0.73% (w/v) and about 1.825% (w/v) L-histidine; or (f) between about 0.4% (w/v) and about 1.333% (w/v) potassium gluconate, and between about 0.73% (w/v) and about 1.825% (w/v) L-histidine; or (g) between about 0.4% (w/v) and about 1.5% (w/v) potassium gluconate, and between about 0.73% (w/v) and about 1.3% (w/v) L-histidine; or (h) between about 0.4% (w/v) and about 1.333% (w/v) potassium gluconate, and between about 0.73% (w/v) and about 1.217% (w/v) L-histidine; or (i) between about 1.3% (w/v) and about 1.6% (w/v) potassium gluconate, and between about 1.2% (w/v) and about 1.3% (w/v) L-histidine; or (j) about 1.3% (w/v) potassium gluconate and about 1.2% (w/v) L-histidine; or (k) about 1.333% (w/v) potassium gluconate and about 1.217% (w/v) L-histidine.

6. The formulation of claim 5, wherein the formulation further comprises a plurality of Lactobacillus cells.

7. The formulation of claim 6. wherein the formulation comprises: (a) between about 0.2% (w/v) and about 2% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 2% (w/v) L-histidine; or (b) between about 0.2% (w/v) and about 1% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 0.912% (w/v) L-histidine; or (c) between about 0.2% (w/v) and about 1% (w/v) potassium gluconate and between about 0.36% (w/v) and about 0.912% (w/v) L-histidine; or (d) between about 0.2% (w/v) and about 1% (w/v) potassium gluconate and between about 0.36% (w/v) and about 0.608% (w/v) L-histidine; or (e) between about 0.2% (w/v) and about 0.75% (w/v) potassium gluconate, and between about 0.36% (w/v) and about 0.912% (w/v) L-histidine; or (f) between about 0.2% (w/v) and about 0.666% (w/v) potassium gluconate, and between about 0.36% (w/v) and about 0.912% (w/v) L-histidine; or (g) between about 0.2% (w/v) and about 0.75% (w/v) potassium gluconate, and between about 0.36% (w/v) and about 0.65% (w/v) L-histidine; or (h) between about 0.2% (w/v) and about 0.666% (w/v) potassium gluconate, and between about 0.36% (w/v) and about 0.608% (w/v) L-histidine; or (i) between about 0.65% (w/v) and about 0.8 % (w/v) potassium gluconate, and between about 0.6% (w/v) and about 0.65% (w/v) L-histidine; or (j) about 0.65% (w/v) potassium gluconate and about 0.6 % (w/v) L-histidine; or (k) about 0.667% (w/v) potassium gluconate and about 0.609% (w/v) L-histidine.

8. The formulation of claim 6 or 7, wherein the Lactobacillus cells are selected from the group consisting of Lactobacillus crispatus, Lactobacillus acidophilus, and Lactobacillus gasseri.

9. The formulation of any one of claims 6-8, wherein the formulation comprises at least 1×10⁹ colony forming units (CFU) or at least 2×10⁹ CFU of Lactobacillus cells.

10. The formulation of claim 4, wherein the formulation comprises: (a) between about 0.4% (w/v) and about 4% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 3.65% (w/v) L-histidine; or (b) between about 0.4% (w/v) and about 2% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (c) between about 0.4% (w/v) and about 0.8% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (d) between about 0.35% (w/v) and about 0.45% (w/v) potassium gluconate, and between about 0.3% (w/v) and about 1.3% (w/v) L-histidine; or (e) about 0.4% (w/v) potassium gluconate, and between about 0.3% (w/v) and about 1.3% (w/v) L-histidine; or (f) about 0.4% (w/v) potassium gluconate, and between about 0.365% (w/v) and about 1.217% (w/v) L-histidine; or (g) about 0.4% (w/v) potassium gluconate and about 1.217% (w/v) L-histidine.

11. The formulation of claim 10, wherein the formulation further comprises a plurality of Lactobacillus cells.

12. The formulation of claim11, wherein the formulation comprises: (a) between about 0.2% (w/v) and about 2% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 2% (w/v) L-histidine; or (b) between about 0.2% (w/v) and about 1% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 0.917% (w/v) L-histidine; or (c) between about 0.2% (w/v) and about 0.4% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 0.917% (w/v) L-histidine; or (d) between about 0.35% (w/v) and about 0.225% (w/v) potassium gluconate, and between about 0.15% (w/v) and about 0.65% (w/v) L-histidine; or (e) about 0.2% (w/v) potassium gluconate, and between about 0.15% (w/v) and about 1.3% (w/v) L-histidine; or (f) about 0.2% (w/v) potassium gluconate, and between about 0.182% (w/v) and about 0.608% (w/v) L-histidine; or (g) about 0.2% (w/v) potassium gluconate and about 0.609% (w/v) L-histidine.

13. The formulation of claim 11 or 12, wherein the Lactobacillus cells comprise Lactobacillus jensenii.

14. The formulation of any one of claims 11-13, wherein the formulation comprises at least 1×10⁹ colony forming units (CFU) of Lactobacillus cells, or at least 2×10⁹ CFU of Lactobacillus cells.

15. The formulation of claim 2, wherein the formulation comprises: (a) about 1.333% (w/v) potassium gluconate and about 1.217% (w/v) L-histidine; (b) about 2% (w/v) potassium gluconate and about 0.112% sulfanilamide; or (c) about 1.333% (w/v) potassium gluconate and about 3.2% (w/v) glycine.

16. A bacterial formulation, comprising: (a) the formulation of any one of claims 5 or 7; and (b) bacteria selected from the group consisting of Lactobacillus species.

17. The bacterial formulation of claim 15, wherein the Lactobacillus species comprise one or more species selected from L. crispatus, L. acidophilus, L. jensenii, L.rhamnosus and L.gasseri.

18. The bacterial formulation of claim 16 or 17, wherein (a) the formulation is the formulation of claim 5, and wherein the Lactobacillus species comprise one or more species selected from L. crispatus, L. acidophilus, and L.gasseri; or (b) the formulation is the formulation of claim 10, and wherein the Lactobacillus species comprise L. jensenii.

19. The formulation of claim 2 or 3, wherein the formulation comprises or consists of magnesium D-gluconate and L-histidine.

20. The formulation of claim 19, comprising: between about 1.02% (w/v) and about 10.2% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 3.650% (w/v) L-histidine; or (b) between about 2.04% (w/v) and about 10.2% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (c) between about 2.04% (w/v) and about 5.1% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (d) between about 5.0% (w/v) and about 5.2% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (e) between about 5.0% (w/v) and about 5.2% (w/v) magnesium D-gluconate, and between about 0.739% (w/v) and about 1.825% (w/v) L-histidine; or (f) about 5.1% (w/v) magnesium D-gluconate and about 0.1.217% (w/v) L- histidine; or (g) between about 2.0% (w/v) and about 2.1% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 1.825% (w/v) L-histidine; or (h) between about 2.0% (w/v) and about 2.1% (w/v) magnesium D-gluconate, and between about 0.730% (w/v) and about 1.825% (w/v) L-histidine; or (i) between about 2.0% (w/v) and about 2.1% (w/v) magnesium D-gluconate, and between about 1.217% (w/v) and about 1.825% (w/v) L-histidine; or (j) about 2.04% (w/v) magnesium D-gluconate and about 1.217% (w/v) L- histidine.

21. The formulation of claim 20 or 21, wherein the formulation further comprises a _{plurality of Bifidobacterium cells.}

22. The formulation of claim 21, wherein the plurality of Bifidobacterium cells are selected from the group consisting of B. infantis, B. bifidum, B. longum and B. breve.

23. The formulation of claim 21, wherein the plurality of Bifidobacterium cells comprise B. infantis.

24. The formulation of any one of claims 21-23, wherein the formulation comprises at least 1×10⁹ colony forming units (CFU) of Bifidobacterium cells.

25. The formulation of any one of claims 21-24, wherein the formulation comprises: (a) between about 0.51% (w/v) and about 5.10% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 1.825% (w/v) L-histidine; or (b) between about 1.02% (w/v) and about 5.10% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 0.913% (w/v) L-histidine; or (c) between about 1.02% (w/v) and about 2.55% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 0.913% (w/v) L-histidine; or (d) between about 2.5% (w/v) and about 2.6% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 0.913% (w/v) L-histidine; or (e) between about 2.5% (w/v) and about 2.6% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 0.913% (w/v) L-histidine; or (f) about 2.55% (w/v) magnesium D-gluconate and about 0.609% (w/v) L- histidine; or (g) between about 1.0 % (w/v) and about 1.05% (w/v) magnesium D-gluconate, and between about 0.183% (w/v) and about 0.913% (w/v) L-histidine; or (h) between about 1.0% (w/v) and about 1.05% (w/v) magnesium D-gluconate, and between about 0.365% (w/v) and about 0.913% (w/v) L-histidine; or (i) between about 1.0% (w/v) and about 1.05% (w/v) magnesium D-gluconate, and between about 0.609% (w/v) and about 0.913% (w/v) L-histidine; or (j) about 1.02% (w/v) magnesium D-gluconate and about 0.609% (w/v) L- histidine.

26. The formulation of claims 1 or 2, wherein the formulation comprises: (a) about 5.1% (w/v) magnesium D-gluconate and about 8.8% (w/v) -alanine; or (b) about 5.1% (w/v) magnesium D-gluconate and about 0.73% (w/v) L-histidine; or (c) about 5.1% (w/v) magnesium D-gluconate and about 2% (w/v) L-alanine; or (d) about 5.1% (w/v) magnesium D-gluconate and about 2% (w/v) -aminobutyric acid.

27. A bacterial formulation, comprising: (a) the formulation of claim 26; and (b) bacteria selected from the group consisting of Bifidobacterium species.

28. The bacterial formulation of claim 27, wherein the Bifidobacterium species comprise one or more species selected from but not limited to the group consisting of B. infantis, B. bifidum, B. longum and B. breve.

29. The formulation or bacterial formulation of any one of claims 1-28, wherein the formulation does not comprise any sugars.

30. The formulation or bacterial formulation of any one of claims 1-29, wherein the excipients are dissolved in water or phosphate buffered saline, or wherein the excipients are dissolved in water.

31. The formulation or bacterial formulation of any one of claims 1-30, wherein the formulation consists of the recited concentration of excipients, with water making up the remainder of the formulation.

32. A lyophilized Lactobacillus cell population, comprising a matrix of hydrogen-bonded potassium gluconate and L-histidine in which the Lactobacillus cells are embedded.

33. The lyophilized Lactobacillus cell population of clam 32, wherein the Lactobacillus cells comprise or consist of one or more of L. crispatus, L. acidophilus, L. jensenii, L.rhamnosus, and L.gasseri.

34. The lyophilized Lactobacillus cell population of clam 32, wherein the Lactobacillus cells comprise or consist of one or more of L. crispatus, L. acidophilus, L. jensenii, and L.gasseri.

35. A lyophilized Bifidobacterium cell population, comprising a matrix of hydrogen- bonded magnesium D-gluconate and L-histidine in which the Bifidobacterium cells are embedded.

36. The lyophilized Bifidobacterium cell population of claim 35, wherein the Bifidobacterium cells comprises or consist of one or more of B. infantis, B. bifidum, B. longum and B. breve.

37. The lyophilized Bifidobacterium cell population of claim 35, wherein the Bifidobacterium cells comprise or consist of B. infantis.

38. The lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population of any one of claims 32-37, wherein the Lactobacillus cells or Bifidobacterium cells was lyophilized in the formulation of any preceding claim.

39. The lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population of any one of claims 32-38, wherein the lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population is a dry powder.

40. A composition, comprising the lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population of any one of claims 32-39, wherein the lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population is present in or on a delivery agent selected from the group consisting of a tablet, a capsule, a gel, a film, a suspension, a baby bottle (including but not limited to coated on the nipple or inside surface of the bottle), a pacifier, or a medical device (including but not limited to an intravaginal ring or a vaginal applicator).

41. The composition of claim 40, wherein delivery agent further comprises boric acid.

42. The composition of claim 41, wherein the boric acid is present in the delivery agent at between about 6 mg and about 600 mg.

43. The composition of claim 40, wherein the delivery agent comprises a vaginal ring.

44. The composition of claim 43, wherein the vaginal ring comprises a flexible, annular body configured for placement within the vaginal canal, wherein the annular body comprise a biocompatible elastomeric material selected from the group consisting of polydimethylsiloxane (PDMS), ethylene-vinyl acetate (EVA), polyurethane (PU), thermoplastic elastomers (TPEs), poly(lactic-co-glycolic acid) (PLGA), and siloxane- urethane copolymers.

45. The composition of claim 44, comprising the lyophilized Lactobacillus cell population disposed on or in the vaginal ring. .

46 The composition of claim 40, wherein the delivery agent comprises breast milk, on a baby bottle (including but not limited to coated on the nipple or inside surface of the bottle), or a pacifier.

47. The lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population, or the composition, of any one of claims 32-46, wherein the Lactobacillus cells or the Bifidobacterium cells are present in a concentration of at least 1 x 10⁹ CFU, or at least 2 x 10⁹ CFU.

48. A method for improving female fertility or limiting development of vaginal infections, comprising administering to a female in need thereof an amount effective of the lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population, or the composition, of any one of claims 32-47, to improve fertility or limit development of vaginal infections.

49. The method of claim 48, comprising administering the lyophilized Lactobacillus cell population of any one of claims 32-34, or 38-39 when dependent from claims 32-34, or the composition of any one of claims 40-47 when depending from claims 32-34.

50. A method for reducing premature birth complications or allergic disorders, comprising administering to an infant in need thereof an amount effect of the lyophilized Lactobacillus cell population or the lyophilized Bifidobacterium cell population, or the composition, of any one of claims 32-47 to reduce premature birth complications or allergic disorders in the infant.

51. The method of claim 50, comprising administering the lyophilized Bifidobacterium cell population of any one of claims 35-37, or 38-38 when depending from claims 35-37, or the composition of any one of claims 40-47 when depending from claims 35-37.

52. A lyophilization method, comprising providing the bacterial formulation of any preceding claim, and lyophilizing the bacteria in the bacterial formulation.

53. A method for drying bacteria, comprising providing the bacterial formulation of any preceding claim, and drying the bacteria in the bacterial formulation.