Described here is a technique in which lipopolysaccharide is injected into the lactating mouse mammary gland via the nipple to simulate mastitis, a condition commonly caused by bacterial infection. Lipopolysaccharide injection results in increased nuclear factor kappa B (NF-κB) signaling, visualized through bioluminescent imaging of an NF-κB luciferase reporter mouse.
Animal models of human disease are necessary in order to rigorously study stages of disease progression and associated mechanisms, and ultimately, as pre-clinical models to test interventions. In these methods, we describe a technique in which lipopolysaccharide (LPS) is injected into the lactating mouse mammary gland via the nipple, effectively modeling mastitis, or inflammation, of the gland. This simulated infection results in increased nuclear factor kappa B (NF-κB) signaling, as visualized through bioluminescent imaging of an NF-κB luciferase reporter mouse1.
Our ultimate goal in developing these methods was to study the inflammation associated with mastitis in the lactating gland, which often includes redness, swelling, and immune cell infiltration2,3. Therefore, we were keenly aware that incision or any type of wounding of the skin, the nipple, or the gland in order to introduce the LPS could not be utilized in our methods since the approach would likely confound the read-out of inflammation. We also desired a straight-forward method that did not require specially made hand-drawn pipettes or the use of micromanipulators to hold these specialized tools in place. Thus, we determined to use a commercially available insulin syringe and to inject the agent into the mammary duct of an intact nipple. This method was successful and allowed us to study the inflammation associated with LPS injection without any additional effects overlaid by the process of injection. In addition, this method also utilized an NF-κB luciferase reporter transgenic mouse and bioluminescent imaging technology to visually and quantitatively show increased NF-κB signaling within the LPS-injected gland4.
These methods are of interest to researchers of many disciplines who wish to model disease within the lactating mammary gland, as ultimately, the technique described here could be utilized for injection of a number of substances, and is not limited to only LPS.
All animal experiments were approved by the Vanderbilt University Institutional Animal Care and Use Committee.
1. Preparing Transgenic Mice
2. Preparation for Injection
3. Intraductal Injection of LPS
4. Bioluminescent Imaging to Visualize NF-κB Activity
5. Representative Results
To obtain interpretable results from these methods, LPS (or PBS) must be infused only into the mammary ducts, leaving the adjacent tissue undisturbed. Injection into the subcutaneous tissue surrounding the nipple will result in inflammation due to injury rather than the simulated infection. Injection of trypan blue dye diluted in PBS is helpful when learning the technique. Figure 3A shows a successfully injected gland infused with 100 μl of trypan blue in PBS. As shown, only the ductal tree should be filled with the injected solution, and there should be no bolus at the injection site. A “bolus” in this context is a build up of the injected liquid at the nipple or in the tissue surrounding the nipple without infusion into the ductal tree. This is shown in Figure 3B. Please note that the practice injections shown in Figure 3 were completed in a female at day 21 of lactation. During the height of lactation (day 8-10, the stage at which LPS is injected using the current protocol), milk is abundant and it is difficult to visualize the ducts and determine if the injection has been completed correctly. Therefore, later time points in lactation are better for practice injections.
Successful injection of LPS into the mammary ducts results in increased NF-κB activity within the gland. This should be evident in the reporter mouse upon IVIS imaging, and the LPS-injected gland should have substantially higher luciferase activity (measured by photon count) than the PBS-injected gland, as shown in Figure 2 (on average, the LPS gland reads 200% of the PBS control signal). A statistically significant increase in luminescence within the LPS-treated glands as compared to the contralateral control was achieved though the completion of a group of 4 mice 4. One additional study has also utilized intraductal mammary injection of either E. coli or LPS into NF-κB reporter transgenics. Similar to our own work, these authors saw a significant increase in reporter activity 6 hr after E. coli injection2.
Please note, due to constitutive NF-κB activity in the brain and other abdominal organs, the heads and abdomens of our reporter mice consistently show a signal when imaged, even at baseline prior to any manipulation. Thus, this signal should be ignored with regard to this experiment. An obvious signal should remain within the LPS-injected mammary gland after the image read-out is adjusted to minimize background luminescence in the PBS gland. Figure 2 shows images with the background reduced.
Unsuccessful injection of the gland will most likely result in damage to the nipple. This will cause pointed, skin-level inflammation that will be evident upon imaging. This type of wounding signal will be very strong and confined to the nipple area where the damage occurred, not emanating throughout the gland, as seen after a successful LPS injection into the ducts.
Figure 1. Schematic of Procedure.
Figure 2. Representative results from IVIS bioluminescent imaging. (A) Acquired image with background reduced, revealing increased luciferase activity in the LPS-injected gland. (B) Image from A with region of interest quantifications added. Click here to view larger figure.
Figure 3. Practice injection: mammary gland injected with dye via intraductal injection. To test the accuracy of delivery via the nipple, a mouse at day 21 of lactation was injected with 100 μl of trypan blue dye diluted in PBS. Following sacrifice, the gland was exposed for visualization. (A) successful injection and (B) bolus of dye at the nipple.
Here we utilize a transgenic NF-κB reporter mouse and an optimized injection technique to model mastitis, inflammation of the breast tissue most commonly caused by bacterial infection. The most critical step in this procedure is the intraductal injection itself.
In practice, it is usually apparent whether the intraductal injection has been successful. The injection of the agents (LPS or PBS) should feel fluid and there should be little to no resistance as the syringe is emptied. The most common pitfall is damage to the nipple leading to obstructed passage into the gland. This can be visualized by looking at the nipple through the dissecting scope. Damage can be avoided by sufficient practice, delicacy when manipulating the nipple, and by ensuring that only the smallest of needles (31G or higher) is used to perform the procedure. Additionally, if the needle is not positioned correctly, or if it slips mid-injection, a build-up of LPS or PBS can form at the injection site and the liquid will not pervade the ducts. If this happens, you may either visually see the accumulation of liquid in the skin around the nipple, or the injection will not feel fluid. Although this pitfall cannot totally be eliminated, occurrences can be avoided for the most part by practice and through the use of the proper tools.
The technique used here for injection was critical for this particular experiment. Some published methods for intraductal injection into the mammary gland require cutting/removal of the nipple and/or incision through the skin 7,8. In this case, as the primary end-point of the study was to determine NF-κB related inflammation within the gland, any incisions or perturbation of the skin or nipple were undesirable and would induce an inflammatory signal due to the tissue injury that would confound the LPS-induced signal.
Additionally, our injection was completed with a commercially available insulin syringe and without the use of a drawn micropipette or micromanipulators, as in several other methodologies for injection that do not utilize nipple removal 9, 10. We believe this was only feasible because the lactating nipple is slightly enlarged and somewhat easier to inject than a young, virgin mouse nipple. To inject virgin mice, more rigorous tools may be required.
It is clear from Figure 2 that if correctly performed, injection of LPS into the ducts of the mouse mammary gland causes localized inflammation as visualized by an increase in NF-κB luciferase reporter activity. Others have analyzed more extensively this model of acute mastitis, through either E. coli or LPS injection into the mammary gland and subsequent analysis of the tissue2-3, 8, 10-14. Among many findings, these studies illustrate that the inflammatory response is modulated by interactions with CD14, and that it results in a neutrophil influx regulated by mammary alveolar macrophages. As suggested in both our previously published work 4, as well as that of others 2 this model of mastitis may be further used to study mechanisms and therapeutics relevant in humans, where as many as 3-33% of breastfeeding women suffer from this painful infection (percentage varies based on study methodology) 15. It may also prove a useful research tool for those in the dairy industry, where mastitis is a costly disease leading to significant losses in milk production each year 16.
Finally, the NGL transgenic reporter mice used here and the similarly functioning HLL reporter mice 17 have been used in numerous studies to verify, quantify, and track the up or down regulation of NF-κB activity within a variety of disease and developmental models including acute lung injury, lung cancer, airway branching morphogenesis, insulin resistance, obesity, and myocardial infarction 1, 17-21. This model provides an additional example of the utility of these reporter transgenics.
The authors have nothing to disclose.
This work was funded by NIH grant CA113734 awarded to F Yull and the Department of Veterans Affairs (TS Blackwell).
Name of the reagent | Company | Catalogue number | Comments |
Lipopolysaccharide from Escherichia coli (LPS) |
Sigma | L 2880 | serotype 055:B5 |
PBS | Mediatech, Inc. | 46-013-CM | |
31 Gauge insulin syringe | Becton Dickinson | 328438 | Short needle |
Forceps | World Precision Instruments | 15908 | |
Dissecting scope | Leica | M3Z | |
Terrell Isoflurane, USP | MINRAD INC. for Rx Elite | NDC 66794-011-25 | |
IVIS 200 imaging system | Caliper (formerly Xenogen) | N/A | |
Syringe for luciferin injection | Becton Dickinson | 309597 | 1cc; 26G5/8 |
D-Luciferin Firefly, sodium salt monohydrate (Luciferin substrate) |
Biosynth Chemistry and Biology | L-8240 | Dilute to 10 mg/ml |