Use of Micropipette-Guided Drug Administration as an Alternative Method to Oral Gavage in Rodent Models

Abstract

Oral gavage (OG) with the use of a cannula attached to a syringe is one of the most common methods used to deliver precise dosing of compounds to the stomach of research animals. Unfortunately, this method comes with difficulties for both the operator and the research animal. Studies have shown that OG may lead to complications, including esophagitis, perforation of the esophagus, and inadvertent tracheal drug administration. In addition, OG is associated with increased plasma and fecal corticosterone levels (due to stress), altered blood pressure, and increased heart rate, which could negatively influence or bias study results. A previously developed alternative method termed micropipette-guided drug administration (MDA) incentivizes the animal to consume treatments readily in a minimally invasive manner. Herein, we present examples of the use of the MDA technique with treatments reconstituted in different vehicles and demonstrate effective delivery of the varied treatments to multiple different mouse strains. We further demonstrate that MDA is a technique that decreases the timing and invasiveness of drug administration and does not affect the gut microbiome composition as assessed by quantitative analysis of core gut microbial species. Overall, MDA may offer a less stressful and effective alternative to OG.

Introduction

Drug administration to rodent models is commonly achieved via oral gavage (OG), which consists of administering a liquid preparation directly to the stomach using a cannula attached to a syringe containing the solution. This technique results in a consistent and precise dosage of the treatment to the animal, but also carries multiple disadvantages. OG has been scrutinized for not adequately modeling human dietary exposures^1,2. Furthermore, OG increases the risk of unintentional injuries to the upper digestive system (perforation of the esophagus and stomach), aspiration of the administered treatment, and respiratory tract lesions³. OG is also associated with discomfort⁴, increased blood pressure and heart rate⁵, as well as stress^6,7, and sometimes death⁸ due to decreased tolerance to gavage by the animal. These physiologic changes might interfere with or confound experimental results; thus, new procedures have been explored to avoid these side effects. Studies have utilized alternative procedures to OG, such as the use of gelatin as a drug vehicle⁹, orally dissolving strips (ODS)¹⁰, sucrose-coated gavage needles¹¹, flexible feeding tubes, wheat cookies¹², honey¹³, and peanut butter pellets¹⁴. Unfortunately, there are limitations with these modifications to the OG technique, including incompatibility with water-insoluble drugs, longer preparation time for the treatment¹⁵, drug palatability, and stability¹⁵, and familiarization of the animal with the food. Furthermore, there is potential for less precise dosing when animals feed ad lib.

Scarborough et al.⁷ previously developed an alternative oral treatment method in mice, which they termed micropipette-guided drug administration (MDA). This method of administration is based on a sweetened condensed milk solution as a vehicle for pharmacological substances, motivating the study animals to consume the prepared vehicle and/or drug solutions readily via dispensing the solution with a single channel pipette and pipette tip. To introduce this technique, rodents undergo a training session (minimum 2 days) to shorten handling times and to allow the study animal to become familiar with drinking from the pipette tip⁴. Initial validation studies by Scarborough et al.⁷ and Schalbetter et al.¹⁶ suggest that the MDA procedure is easy to implement, cost-effective, minimally invasive, and less stressful for the animals than conventional oral gavage methods. Scarborough et al. introduced the use of the MDA technique in a mouse model of maternal immune activation (MIA) of neurodevelopmental disorders⁷. This study demonstrated that the pharmacokinetic profiles of mice treated with the antipsychotic drug risperidone using MDA were comparable to the use of OG. Furthermore, MDA did not induce an increase in corticosterone (a stress hormone) levels in the mice, and chronic treatment with risperidone using the MDA technique led to a dose-dependent decrease in MIA-induced social interaction deficits and amphetamine hypersensitivity⁷. Additional studies have explored the efficacy of MDA versus OG in both mouse¹⁷ and rat¹⁸ models. MDA has also been compared to intraperitoneal injection and was shown to be as effective in delivering clozapine-N-oxide to mice¹⁶. Due to MDA's reported success in reducing animal stress and therapeutic efficacy, we now aim to further explore the MDA technique as an effective method of drug delivery using additional murine models. Here, we describe the implementation of the MDA method to treat different mouse strains, including the immune-competent FVB/NJ and C57BL/6J strains and the immunocompromised NOD.Cg-Prkdc^scid Il2rg^tm1Wjl/SzJ (NSG) strain with different oral treatments, including live bacteria, experimental compounds delivered in water-insoluble solutions (corn oil), and vehicle controls. We assessed the activity and presence of the different treatments in serum and feces, and we evaluated alterations to core gut microbiota in relation to the MDA method.

Protocol

All animal studies were performed following institutional guidelines and under Johns Hopkins University Institutional Animal Care and Use Committee (IACUC) approved protocols M021M197 and M023M195. Mouse strains used (male mice, 7 weeks old) are described in the Table of Materials. The use of male mice was due to their use in the ongoing studies of prostate cancer. The MDA method has previously been shown to be effective in female mice as well^7,17. Mice were randomized to their respective cages/treatment groups. The mice were treated in cages and in no particular order. The experimenter administering the treatments was blinded to the treatment group. The number assigned to a mouse referred to their ID and not the order in which they were treated. The data was collected throughout the timeline of the study and analyzed at the end to avoid any bias. Experimental assays were blinded to the treatment group until all data was collected.

NOTE: This protocol has been modified from Scarborough et al.⁷.

1. Preparation of treatments

1. Prepare a solution of sweetened condensed milk (ingredients: milk and sugar) and molecular biology grade water using a 3:10 ratio of milk to water. Measure the sweetened condensed milk using a 15 mL conical tube due to the viscosity of the milk.
2. Using a 50 mL conical tube, mix molecular biology-grade water with the sweetened condensed milk by slowly mixing 5 mL of water at a time with the milk.
   NOTE: This step is performed under sterile conditions using a biosafety cabinet.
3. Prepare aliquots of the milk solution (stored as 500 µL aliquots) and store the aliquots at 4 °C until combined with the treatment of interest.
4. Resuspend treatments in their respective vehicles at the correct concentration based on the animal's body weight before combining them with the sweetened condensed milk/water solution.

2. MDA training session

NOTE: This section describes mouse models; however, the MDA technique could be scaled up as needed with larger volumes/pipette tips for larger rodent models.

1. OPTIONAL: Gently restrain the study animal by holding the scruff of the neck with one hand.
2. Offer the study animal 100 µL of sweetened condensed milk/water solution only (without the addition of the experimental treatment) using a single channel pipette and a 200 µL pipette tip which is guided to the mouth.
3. Dispense the milk/water solution slowly to avoid missing the animal's mouth. Ensure the animal readily consumes the milk solution between 5-15 s.
   NOTE: This training occurs for a minimum of 2 consecutive days (one training session per day), with no changes to the volume administered. A gentle restraint by scruff was needed for the MDA training sessions. Scarborough et al.⁷ recommend full restraint of the animal on the first day of MDA training and then restraint by the tail on the second day. In contrast, Vanhecke et al.¹⁷ only held the mice slightly by the tail during training.

3. Mouse treatments using the MDA method

1. Combine the study treatment with the sweetened condensed milk/water solution (see examples with validation experiments [section 4 and section 5] below).
2. OPTIONAL: Gently restrain the study animal by holding the scruff of the neck with one hand.
   NOTE: Scarborough et al.⁷ report that mice no longer require restraint beyond the training period and will readily drink from the pipette tip. In contrast, Vanhecke et al.¹⁷ report that for some treatments (e.g., tamoxifen) that might have mild taste aversion, MDA is best performed by gently restraining the mice via scruff during all treatments. This study also determined that positioning the animal on their back with gentle scruff improves the time and efficacy of treatment consumption.
3. Administer 100 µL of treatment using a single channel pipette and a 200 µL pipette tip guided to the animal's mouth, as performed during the training sessions.
4. Treat a single animal at a time.
5. Place each animal back into their respective cage after treatment.
   NOTE: Treatments using the MDA method can be given daily or in a repeated fashion^7,17.

4. Validation experiment #1 - oral delivery of gut bacteria to mice using MDA

NOTE: This section describes the use of the MDA method to orally deliver live bacteria for gut colonization studies in mice. In this representative pilot experiment, the gram-positive bacterium Clostridium scindens is delivered to C57BL/6J mice. The mice were treated with antibiotics (cefoxitin) for two consecutive days in the drinking water prior to bacterial inoculation with MDA.

1. Bacterial culture
   1. Culture C. scindens strain ATCC 35704 in reinforced clostridial broth (RCB) using a frozen glycerol stock to inoculate 500 µL of bacteria into a 7 mL RCB tube. Grow anaerobically overnight at 37 °C.
   2. Use 500 µL of the overnight bacteria culture to inoculate a new RCB tube and incubate anaerobically for 24 h. Remove this tube from the anaerobic chamber and use it to prepare the bacterial treatment for MDA. Use this same tube to calculate the colony-forming units per milliliter (CFU/mL).
2. Calculating C. scindens CFU/mL
   1. Create a 1:3 dilution of the C. scindens culture using 3.5 mL of the culture inoculated into 7 mL of RCB media. Use a sample of this dilution to make a 20% 1 mL glycerol stock (1:1) that is stored at -80 °C.
   2. Plate 100 µL of the 1:3 dilution onto reinforced clostridial agar (RCA) plates. Use the first dilution to then make subsequent serial dilutions and plate for a total of three more times.
   3. After incubation of the RCA plates, count the colonies and calculate the CFU. Use the following equation to calculate the concentration of bacterial cells in the diluted and original samples:
      CFU in diluted sample (cells/mL) = (number of colonies counted on the RCA plate)/(amount of diluted sample added to the RCA plate in mL)
      ​CFU in original sample (cells/mL) = (CFU in diluted sample)/(dilution of the RCA plate)
3. Preparing the C. scindens treatment for MDA
   1. Combine a solution of 50 µL (1.265 x 10⁶ CFU/mL in this example) of C. scindens + 50 µL of the sweetened condensed milk/water solution.
      NOTE: If a specific CFU is needed, an alternative method of bacterial cell quantification may be used, such as creating a standard curve with optical density (OD) measurements at 600 nm.
4. Oral treatment of mice with C. scindens using MDA
   1. On days 1-3, familiarize the mice with the milk solution as described in section 2 (MDA training session).
   2. On day 4, administer 100 µL of the C. scindens + milk solution or PBS (control) + milk solution via MDA to C57BL/6J mice. Treat mice once for the duration of the study.
5. Fecal sample collection
   1. Hold the mouse by gentle scruff and wait for them to discharge a fecal pellet. Collect fecal pellets directly from the rectum into a sterile 1.5 mL microcentrifuge tube. Store the samples at -80 °C until use.
6. C. scindens quantitative PCR (qPCR)
   1. Perform fecal DNA extraction using a previously published protocol^19,20.
   2. Determine DNA concentration using a DNA fluorometer for high-sensitivity detection. Normalize DNA samples to 10 ng/µL and store at -20 °C until use.
   3. Use qPCR to investigate the abundance of: (i) C. scindens bacteria using conserved primers against the desA gene of C. scindens strain ATCC 35704, (ii) total bacteria using primers designed against the V6 hypervariable region of the bacterial 16S rRNA gene as previously described^20,21.
      NOTE: The qPCR conditions and primers used are listed in Table 1. A standard curve using C. scindens strain ATCC 35704 DNA was used to estimate the copies of desA+ bacteria.

5. Validation experiment #2 - experimental drug delivery to mice using MDA

NOTE: This section describes the use of the MDA method to deliver an experimental compound to mice. In this representative pilot experiment, the soy metabolite equol (S-equol) is delivered to NSG and FVB/NJ mice.

1. Preparation of S-equol treatment
   NOTE: Both S-equol and dimethyl sulfoxide (DMSO) are aliquoted and stored at -20 °C until use.
   1. Reconstitute S-equol in DMSO combined with corn oil (90%). Mix aliquoted S-equol in corn oil (220 µL) with 480 µL of the sweetened condensed milk/water solution. Combine and mix this solution in the animal room to avoid the separation of the oil from the milk.
      NOTE: If separation of the solution is observed, mix the solution before treatment. The dose of S-equol administered is 25 mg/kg/day. The mice used in this study were all similar in weight, so they received the same dose. The dose of the treatment may need to be adjusted and made into multiple preparations based on the study animal weights.
   2. Combine DMSO vehicle with corn oil (90%). Mix aliquoted DMSO in corn oil (220 µL) with 480 µL of the sweetened condensed milk/water solution. Combine and mix this solution in the animal room to avoid the separation of the oil from the milk.
      NOTE: If separation of the solution is observed, mix the solution before treatment.
2. Oral treatment of mice with S-equol using MDA
   1. On days 1 and 2, familiarize the mice with the milk solution as described in section 2 (MDA training session).
   2. On days 3-43, treat mice with (i) PBS control, (ii) DMSO vehicle, or (iii) S-equol, all combined with the sweetened condensed milk/water solution. Make a master mix of the solution to deliver a homogenous solution of the treatments to all study animals. Administer a total volume of 100 µL to each mouse once daily, 5 days a week, for a total of 43 days.
3. Blood collection
   1. Collect blood after euthanasia (carbon dioxide overdose) via heart puncture. Transfer blood to a microtainer serum separator tube, mix, and store at room temperature (RT) until further processing.
   2. Centrifuge blood sample to complete separation at 254 x g for 5 min. Collect serum (top layer of separation) and store at -20 °C until use.
4. S-equol liquid chromatography with tandem mass spectrometry (LC/MS/MS) assay
   1. Refer to the Table of Materials file for the chemicals and reagents used for this experiment.
   2. Calibration standards and quality control: Prepare stock solutions of S-equol and internal standard (racemic equol-d4) in DMSO at concentrations of 1 mg/mL and store at -20 °C.
   3. Dilute the stock standard solutions with acetonitrile:water (50:50) to prepare a mixed working solution with different concentrations. Prepare internal standard in ethyl acetate as extraction solution at a concentration of 100 ng/mL. Store stock and working solutions at 4 °C and use daily, diluted, or directly.
   4. Sample extraction
      1. Prepare the sample using protein precipitation by extraction solution (ethyl acetate plus internal standard, see section 5.4.3), followed by centrifugation and evaporation of supernatant under dry nitrogen (50 °C).
      2. Reconstitute the sample in 100 µL of 50% acetonitrile and transfer to autosampler vials for LC/MS/MS analysis.
   5. Sample separation
      1. Achieve separation with a C18 column, 2.1x50 mm, 1.7 µm. Use 0.1% formic acid as mobile phase A and 0.1% formic acid in acetonitrile mobile phase B.
      2. Use a gradient and hold mobile phase B at 40% with a flow rate of 0.3 mL/min for 0.5 min. Then, increase the flow rate to 100% over 1 min, hold there for 2 min, and then return to 40% B for 1 min.
      3. Monitor the column effluent using a mass-spectrometric detector using electrospray ionization, which is operating in negative mode. Program the spectrometer to monitor the following multiple reaction monitoring (MRM) transitions: 241 → 119.1 for S-equol and 245 → 123 for the internal standard, equol-d4.
      4. Compute the calibration curve for S-equol using the area ratio peak of the analysis to the internal standard by using a quadratic equation with a 1/x² weighting function using the calibration ranges of 5-500 ng/mL.

6. Measuring the effect of MDA treatment on core gut microbiota

NOTE: This section describes using qPCR to measure levels of core gut microbiota after MDA treatment.

1. Perform fecal sample collection and DNA extraction as described in steps 4.5.1 and 4.6.1.
2. Determine DNA concentration using a DNA fluorometer for high-sensitivity detection. Normalize DNA samples to 10 ng/µL and store at -20 °C until use.
3. Use qPCR to investigate the abundance of members of the gut microbiota: (i) Akkermansia muciniphila, (ii) Bifidobacterium spp., (iii) Streptococcus salivarius, (iv) Lactobacillus spp., and (v) total bacteria using primers designed against the V6 hypervariable region of the bacterial 16S rRNA gene as previously described^20,21,22.
   NOTE: The qPCR conditions and primers used are listed in Table 2.

Representative Results

MDA can be used in the oral delivery of bacterial strains in mouse models. C57BL/6J mice were treated with antibiotics (cefoxitin in the drinking water) for 2 days to clear commensal microbial communities before starting the MDA training session. The sweetened condensed milk/water solution was administered once daily consecutively for 3 days prior to treatment administration. Mice were briefly restrained by gentle scruff during MDA treatment administration. On day 4, mice were treated once with PBS (negative control, n = 4) or an anaerobic culture of C. scindens (desA+ bacteria, n = 4) that was previously resuspended in the sweetened condensed milk/water solution. All mice generally took the treatments within seconds and drank the entire dose (100 µL). To validate the delivery and presence of the anaerobic bacteria, we compared fecal samples collected after antibiotic treatment to fecal samples collected 24 h after PBS or C. scindens treatment (Figure 1). Fecal DNA was isolated, quantified, and normalized to 10 ng/µL to perform qPCR using C. scindens ATCC 35704 desA gene-specific primers (Table 1). We observed that the antibiotic-treated mice had minimal to no desA+ bacteria. As expected, we observed a substantially higher quantity of desA+ bacteria in the C. scindens-treated mice, which was significantly different from those in the PBS control group (p = 0.0286, Figure 1).

Delivery of an experimental compound (S-equol) using the MDA method. NSG (n = 15) and FVB/NJ (n = 18) mice underwent training with the sweetened condensed milk/water solution two days prior to treatment administration. NSG mice underwent antibiotic treatment (cefoxitin in the drinking water) to eliminate equol-producing gut bacteria and, hence, endogenous equol production. FVB/NJ mice were fed a soy-free diet to eliminate endogenous equol production. All mice were briefly restrained by gentle scruff in order to administer the treatments. The control group received the sweetened condensed milk/water solution in combination with corn oil, the DMSO group (vehicle control for S-equol) received the sweetened condensed milk/water solution and DMSO in corn oil, and the equol group received sweetened condensed milk/water solution and equol in corn oil. Mice were treated five days a week with the MDA technique for a total of 43 days. Mice generally took the treatments within seconds and drank the entire dose (100 µL). Overall, all the mice consumed at least half of the dose on a particular day (partial dose on a particular day, but complete dose either the day prior or the following day). None of the mice declined the treatment entirely throughout the course of the study. We observed that after scruffing, positioning the mice on their backs improved the time of treatment consumption and ensured that the animal took the entire dose.

Circulating equol levels demonstrate effective MDA drug delivery. Blood samples were taken from the NSG mice after euthanasia at the study end to validate that the MDA delivery of equol led to circulating equol levels in the study animals. We did not recover enough blood from one of the equol-treated mice to perform the LC/MS/MS assay. Serum equol levels were measured by LC/MS/MS at the Analytical Pharmacology Shared Resource at the Johns Hopkins University School of Medicine. We observed no circulating equol in the control and DMSO (equol vehicle) groups, while the equol-treated mice had detectable circulating equol (Figure 2). Importantly, we expect some variation in circulating equol levels in the equol-treated mice due to differing dates and times of animal sacrifice in this experiment.

Gut microbial communities remain unchanged after MDA treatment. The effect of the MDA milk solution (with and without equol) on the gut microbiota was assessed using the FVB/NJ mice. We quantified the microbial composition of samples taken at baseline (no treatments/no MDA) and samples taken 7 days after treatment initiation (9 days after MDA initiation, including training days). We observed no significant changes in total bacterial load or the overall microbial composition of core gut microbiota (Akkermansia muciniphila, Bifidobacterium spp., Streptococcus salivarius, Lactobacillus spp.) as assessed by species-specific qPCR analysis (Figure 3) as previously described^21,22. One cage showed a significant difference in Lactobacillus between baseline and after MDA treatment; however, this finding was observed in the DMSO vehicle control and was not consistent across all cages.

Figure 1: MDA delivers anaerobic bacteria to the mouse gut that can be quantified via fecal qPCR. C57BL/6J mice were treated with antibiotics for two consecutive days in the drinking water. Mice were then treated with the anaerobic bacterium C. scindens using the MDA method. Fecal C. scindens (desA+) bacterial load was measured by qPCR for the desA gene both after antibiotic treatment and 24 h post-MDA treatment (PBS or C. scindens). (A) C. scindens-specific desA gene quantification is shown between PBS- and C. scindens-treated mice after antibiotic treatment and 24 h post-MDA treatment. (B) The same data shown for 24 h post-MDA in (A) is shown as a column graph. Statistical significance was determined using a non-parametric Mann-Whitney test. (*) p = 0.0286, error bars = SD. Please click here to view a larger version of this figure.

Figure 2: Analysis of circulating equol levels in NSG mice treated with equol via the MDA technique. Serum equol levels were quantified via LC/MS/MS in control (n = 5), DMSO (vehicle, n = 5), and equol (n = 4) treated mice. Enough blood was not recovered from one of the equol-treated mice to perform LC/MS/MS. These treatments were delivered using the MDA method. Serum equol was only detectable in the equol-treated group. Limit of quantification < 5 ng/mL. Statistical significance was determined using a non-parametric one-way ANOVA with multiple comparisons (Dunn's correction). (**) p = 0.0053, error bars = SD. Please click here to view a larger version of this figure.

Figure 3: Core gut bacteria and total bacterial load levels are unchanged in FVB/NJ mice after MDA treatment. No significant changes were observed in the qPCR Cq values for the targeted microbial communities before and after treatment administration using the MDA method. Lactobacillus showed a significant difference (*, p = 0.0321) only in cage 3 between baseline and MDA treatment. This finding was not consistent among cages. Cage numbers are marked as C1-C4. Statistical significance was determined using a one-way ANOVA with multiple comparisons (Tukey correction). Error bars = SD. Please click here to view a larger version of this figure.

Primer name	Primer sequences	PCR conditions
DesA_CS_F	5’-GGACAGGGTGTCGGCTTTATG-3’	93°C 3 min, 34 cycles (95 °C 30 s, 54 °C 30 s)
DesA_CS_R	5’-TTACTTCACCTTCGCCAGTTTCTG-3’

Table 1: Primers and qPCR conditions for desA+ bacteria (C. scindens).

Primer name	Primer sequences	PCR conditions
V6_F V6_R	5’-CAACGCGWRGAACCTTACC-3’ 5’-CRACACGAGCTGACGAC-3’	95 °C, 3 min 36 cycles 95 °C, 30 s 53 °C, 30 s
A.muc_F A.muc_R	5’-CAGCACGTGAAGGTGGGGAC-3’ 5’-CCTTGCGGTTGGCTTCAGAT-3’	50 °C, 5 min 95 °C, 10 min 40 cycles 95 °C, 15 s 60 °C, 1 min
Bifido_F Bifido_R	5’-CGGGTGAGTAATGCGTGACC-3’ 5’-TGATAGGACGCGACCCCA-3’	95 °C, 3 min 36 cycles 95 °C, 30 s 53 °C, 30 s
S. salivarius_F S. salivarius_R	5’-CACGCCATGCTGGAAGTG-3’ 5’-GCGATGAGCCAAGCTGAAG-3’	95 °C, 10 min 44 cycles 95 °C, 15 s 60 °C, 1 min
Lacto_F Lacto_R	5’-TGGAAACAGRTGCTAATACCG-3’ 5’-GTCCATTGTGGAAGATTCCC-3’	50 °C, 2 min 95 °C, 10 min 40 cycles 95 °C, 15 s 62 °C, 1 min

Table 2: Primers and qPCR conditions used to measure the effect of MDA on core gut microbiota.

Discussion

OG can be a significant source of stress in research animals that may create a confounding variable as previously assessed in multiple studies^{7,9,11,12,13,14,15,23}. Due to the invasiveness of OG, alternate techniques have been employed to minimize the challenges associated with OG^{9,10,12,13,14,15}. Micropipette-guided drug administration (MDA) is a novel technique that allows for drug delivery in a less invasive manner using a single channel pipette and a mix of the drug of interest with a sweetened condensed milk solution⁷. A critical step for this technique is the training session (minimum 2 days) that allows the animal to become more familiar with the pipette tip and the sweetened condensed milk flavor, and overall shortens handling times and distress. This training session also allows the mice to readily consume the drugs incorporated in the sweetened condensed milk solution much faster than other techniques and with a precise dosage.

Our studies closely followed the preparations and delivery of the sweetened condensed milk/water solution presented in Scarborough et al.⁷. During the length of the study, mice were not deprived of water or food prior to dosing. They were kept in their respective housing cage, and the timing of treatments was under 1 min. One key difference in our studies was that, possibly due to the potential for visual impairment in the albino mouse strains that we used and the potential for slight taste aversion with drug treatment, all of the mice required brief restraint by gentle scruffing as part of the training and treatment protocol. This modification to the MDA treatments is recommended by Vanhecke et al.¹⁶. The gentle scruff, along with positioning the mice on their backs, encouraged the mice to eagerly consume the milk solution. We conclude that the need for gentle scruffing as part of the training and treatment steps is optional and will be study-dependent. Administration of the treatments (corrected for the animal's body weight, as applicable) was given in a standardized volume, and mice did not show changes in behavior. Other slight modifications to the initial study⁷ included a decrease in the number of training sessions and the volumes administered at each session. Some of the limitations we encountered in this study include the resistance of some mice to drink the entire dose (100 µL) of the milk solution on the first day of training and at time points during the study. This was overcome by positioning the mice on their backs while being restrained. This study did not assess the interaction of the drug with the sweetened condensed milk solution, differences in consumption between mouse sexes (although MDA has been previously shown to be effective in female mice as well^7,17), or elevation of stress markers, such as corticosterone levels^7,11,14, associated with OG.

In this study, we focused on the ability of the MDA technique to deliver a precise dose of an experimental treatment and then assessed fecal and/or circulating levels of the treatments of interest in the study animals. We likewise assessed whether MDA treatment alters gut microbiota, and we observed minimal to no alterations in core fecal microbiota or total fecal bacterial load of the study animals. We likewise observed no injuries in the animals due to MDA treatment. In conclusion, MDA can serve as an alternative method of short-term and long-term treatment delivery in rodent models. Overall, MDA stands to have a positive impact on animal welfare by reducing the risk of OG-associated injuries. MDA overcomes some of the limitations of other techniques and we propose that it can be broadly used in translational research in many different rodent models²⁴.

Materials

Name	Company	Catalog Number	Comments
200 µL pipette tips	Mettler Toledo	17005860
AB SCIEX Triple QTRAP 5500 mass-spectrometric detector	Sciex	N/A
Akkermansia muciniphila strain muc genomic DNA	American Type Culture Collection	BAA-835D-5
Ammonium acetate	Sigma–Aldrich	5.43834
C57BL/6J mice	Jackson Laboratories	Strain# 000664
C. scindens strain 35704	American Type Culture Collection	35704
Cefoxitin	Sagent	NDC25021-109-10
Corn oil	MedChemExpress	HY-Y1888
DMSO	Sigma-Aldrich	D2650
ethanol	Fisher Scientific	AC611050040
Formic acid	Sigma–Aldrich	5.33002
FVB/NJ mice	Jackson Laboratories	Strain# 001800
Glycerol	Sigma–Aldrich	G5516
Hexane	Fisher Scientific	02-002-996
LC-MS grade water	Fisher Scientific	14-650-357
Methanol	Fisher Scientific	02-003-340
Microtainer serum separator tube	Becton Dickinson	02-675-185
Molecular biology grade water	Corning	46-000-CI
NSG mice	Jackson Laboratories	Strain# 005557
PBS	Corning	21-031-CV
Qubit DNA HS kit	Invitrogen	Q32851
Racemic equol-d4	Santa Cruz Biotechnology	sc-219827
Reinforced Clostridial agar	Anaerobe Systems	AS-6061
Reinforced Clostridial broth	Anaerobe Systems	AS-606
S-equol	MedChemExpress	HY-100583
S-equol reference standard for LC-MS	Cayman Chemical	10010173
Single channel pipette	Rainin	17008652
Streptococcus salivarius genomic DNA	American Type Culture Collection	BAA-1024D-5
Sweetened condensed milk	California Farms	B09TGQ7WV8
VSL#3	VSL#3	B07WX1LVHL
β-glucuronidase from Helix pomatia	Sigma–Aldrich	G7017