We describe here a method of intravitreal injection and subsequent bacterial quantitation in mouse model of bacterial endophthalmitis. This protocol can be extended for measuring host immune responses and bacterial and host gene expression.
Intraocular bacterial infections are a danger to the vision. Researchers use animal models to investigate the host and bacterial factors and immune response pathways associated with infection to identify viable therapeutic targets and to test drugs to prevent blindness. The intravitreal injection technique is used to inject organisms, drugs, or other substances directly into the vitreous cavity in the posterior segment of the eye. Here, we demonstrated this injection technique to initiate infection in the mouse eye and the technique of quantifying intraocular bacteria. Bacillus cereus was grown in brain heart infusion liquid media for 18 hours and resuspended to a concentration 100 colony forming units (CFU)/0.5 µL. A C57BL/6J mouse was anesthetized using a combination of ketamine and xylazine. Using a picoliter microinjector and glass capillary needles, 0.5 µL of the Bacillus suspension was injected into the mid vitreous of the mouse eye. The contralateral control eye was either injected with sterile media (surgical control) or was not injected (absolute control). At 10 hours post infection, mice were euthanized, and eyes were harvested using sterile surgical tweezers and placed into a tube containing 400 µL sterile PBS and 1 mm sterile glass beads. For ELISAs or myeloperoxidase assays, proteinase inhibitor was added to the tubes. For RNA extraction, the appropriate lysis buffer was added. Eyes were homogenized in a tissue homogenizer for 1-2 minutes. Homogenates were serially diluted 10-fold in PBS and track diluted onto agar plates. The remainder of the homogenates were stored at -80 °C for additional assays. Plates were incubated for 24 hours and CFU per eye was quantified. These techniques result in reproducible infections in mouse eyes and facilitate quantitation of viable bacteria, the host immune response, and omics of host and bacterial gene expression.
Bacterial endophthalmitis is a devastating infection that causes inflammation, and, if not treated properly, can result in loss of vision or blindness. Endophthalmitis results from the entry of bacteria into the interior of the eye1,2,3,4,5. Once in the eye, bacteria replicate, produce toxins and other noxious factors, and can cause irreversible damage to delicate retinal cells and tissues. Ocular damage can also be caused by inflammation, due to the activation of inflammatory pathways leading to inflammatory cell influx into the interior of the eye1,5,6. Endophthalmitis can occur following intraocular surgery (post-operative), a penetrating injury to the eye (post-traumatic), or from metastatic spread of bacteria into the eye from a different anatomical site (endogenous)7,8,9,10. Treatments for bacterial endophthalmitis includes antibiotics, anti-inflammatory drugs, or surgical intervention3,4,11. Even with these treatments, vision or the eye itself may be lost. The visual prognosis after bacterial endophthalmitis generally varies depending upon the treatment effectiveness, the visual acuity at presentation, and the virulence of the infecting organism.
Bacillus cereus (B. cereus) is one of the major bacterial pathogens that causes post-traumatic endophthalmitis7,12. A majority of B. cereus endophthalmitis cases have a rapid course, which can result in blindness within a few days. The hallmarks of B. cereus endophthalmitis include quickly evolving intraocular inflammation, eye pain, rapid loss of visual acuity, and fever. B. cereus grows rapidly in the eye compared to other bacteria which commonly cause eye infections2,4,12 and possesses many virulence factors. Therefore, the window for successful therapeutic intervention is relatively short1, 2, 3, 4, 5, 6, 7, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25. Treatments for this infection are usually successful in treating endophthalmitis caused by other less virulent pathogens, but B. cereus endophthalmitis commonly results in greater than 70% of patients suffering from significant vision loss. About 50% of those patients undergo evisceration or enucleation of the infected eye7,16,22,23. The destructive and rapid nature of B. cereus endophthalmitis calls for immediate and proper treatment. Recent progress in discerning the underlying mechanisms of disease development have identified potential targets for intervention19,26,27. Experimental mouse models of B. cereus endophthalmitis continue to be useful in discerning the mechanisms of infection and testing potential therapeutics that may prevent vision loss.
Experimental intraocular infection of mice with B. cereus has been an instrumental model for understanding bacterial and host factors, as well as their interactions, during endophthalmitis28. This model mimics a post-traumatic or post-operative event, in which bacteria are introduced into the eye during an injury. This model is highly reproducible and has been useful for testing experimental therapies and providing data for improvements in standard of care1,6,19,29,30. Like many other infection models, this model allows for independent control of many parameters of infection and enables efficient and reproducible examination of infection outcomes. Studies in a similar model in rabbits over the past few decades have examined the effects of B. cereus virulence factors in the eye2,4,13,14,31. By injecting B. cereus mutant strains lacking individual or multiple virulence factors, the contribution of these virulence factors to disease severity can be measured by outcomes such as the concentration of bacteria at different hours of postinfection or the loss of visual function13,14,27,31,32. In addition, host factors have been examined in this model by infecting knockout mouse strains lacking specific inflammatory host factors26,29,33,34,35. The model is also useful for testing potential treatments for this disease by injecting novel compounds into the eye after infection30,36. In this manuscript, we describe a detailed protocol which includes infecting a mouse eye with B. cereus, harvesting the eye after infection, quantifying intraocular bacterial load, and preserving specimens to assay additional parameters of disease severity.
All procedures were performed following the recommendations in the Guide for the Care and Use of Laboratory Animals and the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research. The protocols were approved by the Institutional Animal Care and Use Committee of the University of Oklahoma Health Sciences Center (protocol numbers 15-103, 18-043, and 18-087).
1. Sterile glass needles
2. Bacillus cereus culture
3. Bacterial dilution for intravitreal injection
4. Mouse intravitreal injection
5. Harvesting tube preparation
6. Harvesting the eyes
7. Intraocular bacterial count
8. Preservation of samples
Generating a reproducible inoculum and accuracy of the intravitreal injection procedure are key steps in developing models of microbial endophthalmitis. Here, we demonstrated the intravitreal injection procedure using Gram-positive Bacillus cereus. We injected 100 CFU/0.5 μL of B. cereus into the mid-vitreous of five C57BL6 mice. After 10 h postinfection, we observed intraocular growth of B. cereus to approximately 1.8 x 105 CFU/eye. Figure 1 demonstrates the construction of glass needles to deliver the bacteria into the midvitreous of the mouse eyes. Bacillus cereus growing on a blood agar plate and in culture tubes is shown in Figure 2. Figure 3 shows the mouse intravitreal injection procedure using an air pressurized injection system. Figure 4 demonstrates the process of harvesting the infected eyes after the desired time postinfection. Figure 5 shows the technique for homogenizing the infected eyes. Figure 6 demonstrates the intraocular bacterial counts from five different mouse eyes at 10 h postinfection. Figure 7 depicts the overall procedure and a graphical representation of a mouse intravitreal injection.
Figure 1: Making beveled glass needles. Glass needles were made from disposable microcapillary pipettes using a needle/pipette puller and a micropipette beveler. (A) Clamped glass capillary tube in the needle/pipette puller. (B,C) Creation of glass needles using the desired voltage. (D) Beveling the glass micropipettes. (E,F) Glass micropipettes before and after beveling. (E) Scaling the glass needles. Space between two black points holds 0.5 µL. Please click here to view a larger version of this figure.
Figure 2: Bacillus cereus ATCC 14579. (A) Bacillus cereus growing on a blood agar plate. Individual colonies of B. cereus typically display clear zones of hemolysis on a blood agar plate. (B) Turbid overnight cultures of B. cereus. (C) Gram-staining of B. cereus. B. cereus are Gram-positive rod-shaped bacteria. (D) Electron micrograph of Bacillus cereus. This electron micrograph shows rod shaped Bacillus cereus with hair like structures called flagella. Please click here to view a larger version of this figure.
Figure 3: Mouse intravitreal injection. The injection procedure is performed using a pressurized air injector with viewing of the operating field using a commercial microscope (A) ketamine and xylazine drug to anesthetize the mouse. (B) Administration of ketamine and xylazine by intraperitoneal injection to anesthetize the mouse. (C) Clamping periocular skin back to proptose the eye. (D) Filling up the needles with bacteria using the air pressurized injector. (E) Intravitreal injection. (F) Monitoring infected mouse after anesthesia. Please click here to view a larger version of this figure.
Figure 4: Harvesting mouse infected eye. Post-infection, after the desired time point, harvest infected mouse eyes using sterile tweezers. (A) PBS containing sterile glass beads. (B) Harvesting of the infected eye. (C) Harvest tube containing a mouse eye. Please click here to view a larger version of this figure.
Figure 5: Processing the harvested eye for intraocular bacterial count. (A) Harvest tube with infected eye clamped tightly to a tissue homogenizer. (B) Infected eyes were homogenized twice for 1 min each. Track dilution (D) of the eye homogenates (C) and subsequent plating (E) for the bacterial quantitation. (E) Representative individual Bacillus cereus colony after track dilution. Please click here to view a larger version of this figure.
Figure 6: Intraocular bacterial counts at 10 h postinfection. M, mouse number. CFU, colony forming units. Please click here to view a larger version of this figure.
Figure 7: Mouse intravitreal injection. (A) Overall flowchart of the mouse intravitreal injection procedure. (B) Graphical presentation of the intravitreal injection. During the procedure, sterile, beveled glass needles containing a culture of B. cereus are inserted into the midvitreous of the mouse eye, and B. cereus are delivered using an air-pressurized microinjection system. Please click here to view a larger version of this figure.
Even with the availability of potent antibiotics, anti-inflammatory drugs, and vitrectomy surgery, bacterial endophthalmitis can blind a patient. Clinical studies have been useful in studying endophthalmitis; however, experimental models of endophthalmitis provide quick and reproducible results that can be translated to progress in standard of care, resulting in better visual outcome for patients.
The vitreous volume of the mouse eye is approximately 7 µL40. This small volume only allows for a limited amount of material to be injected. Volumes greater than 1.0 µL should not be injected in order to avoid ocular damage. The process requires specific equipment and practice of techniques to ensure reproducibility and accuracy. Endophthalmitis has been studied in primates, swine, rabbits, rats, guinea pigs and in the mouse28,41,42,43,44,45. Among the larger species, eyes are much closer in size to human eyes, and intravitreal injections into larger eyes can be performed without special equipment. Except for the mouse, the absence of knockout strains and reagents to study the host immune responses in other animal models constrains their usefulness in experimental endophthalmitis.
Endophthalmitis occurs most frequently as a complication of cataract surgery. This infection can also occur following penetrating ocular trauma or systemic infection. The visual outcome in this disease partly depends on the virulence of the infecting pathogen. In mice, visual outcome also depends on their immune status. Understanding the mechanisms of disease pathogenesis has been facilitated by studying the disease in experimental animal models28. The protocol for mouse intravitreal injection and the quantification of infection parameters can be adapted to study endophthalmitis initiated with almost any type of bacterial or fungal pathogen28. Furthermore, this protocol can also be applied and modified to study anatomical changes, inflammatory processes, and the gene expression profiles of both the bacteria and host during infection.
Mouse intravitreal injection and subsequent analysis of infection parameters consist of several critical steps (Figure 7). The glass needle must be accurately created, marked, and sterilized. The sharp end of beveled glass needle and proper scaling determines the adequate delivery of the bacteria and globe puncture. If the end of the needle is short and is not appropriately beveled, it could create a large hole on the globe which could cause leakage, contamination of the globe, and an inaccurate disease outcome. Therefore, it is recommended to observe the glass micropipettes while beveling under 10x optical zoom of a bright field microscope. To ensure the delivery of the correct quantity of bacteria, dilutions should be calculated beforehand, and the proper volume scale should be marked on the glass needle. Parameters for needle pulling and beveling using other types of equipment may vary.
The described intravitreal injection technique utilizes an air-pressurized microinjector system for the proper delivery of the bacteria into the mouse eye1. The appropriate use of this microinjector is crucial for the reproducible infection parameters. Since the air pressure determines the speed of delivery, excessive force might rapidly inject a larger volume into the eyes, causing excessive intraocular pressure and/or globe rupture. Therefore, the recommendation is to use a pressure of 10-13 psi to fill and inject during the procedure. Another potential issue is leakage of the bacterial solution from the glass needles after filling. Leakage could result in the injection of inaccurate volumes into the mouse eyes. Always check connections between the injection system, tubing, and glass needles before injections.
Injecting accurate quantities of bacteria is vital for the reproducibility of experimental bacterial endophthalmitis. Different experimental endophthalmitis models require inoculation of specific numbers of bacteria12,28,46,47,48,49. For B. cereus endophthalmitis, injecting 100 CFU is needed to initiate a reproducible infection1. Since bacterial growth depends on the growth medium and conditions, the growth environment, media, and dilutions must be repeated each time. Therefore, it is recommended to test the culture conditions before the experiment to reproduce the accurate inoculum in the appropriate volume for injection. As noted above, the upper limit for intravitreal injection into mouse eyes is 1.0 µL28. An excessive volume elevates the intraocular pressure which could result in glaucoma, detach the retina from the posterior segment, or rupture the globe. Intravitreal injection and the resulting infection in mice may not perfectly mimic B. cereus endophthalmitis in a human. Most animal models of human disease are limited in this way. Known quantities of B. cereus are injected into the eye after growth in nutrient-rich media. For clinical cases of B. cereus endophthalmitis, the infecting quantities are not known and the sources of contamination vary. However, the advantages of using characterized organisms and mouse strains and generating reproducible infection courses outweigh these limitations.
Proper harvesting of infected eyes is another critical step. Maintaining aseptic conditions is essential to avoid any cross-contamination which could interfere with interpretation of the data. Therefore, the recommendation is to disinfect the workbench and all instruments with 70% ethanol before harvest. Bacterial numbers increase rapidly inside the vitreous environment during infection, which may cause swelling of the eye. Therefore, care should be taken to gently remove the eye during harvesting prevent rupture of the globe. Furthermore, the cap of the harvest tube containing the infected eye must be adequately tightened before placing the tube in tissue homogenizer. A loose cap will result in leakage and contamination of the tissue homogenizer leading to inaccurate quantitation of bacteria in the leaking tubes. The eye homogenization procedure also elevates temperatures of the tubes. Therefore, it is recommended to homogenize 1 minute at a time. Elevated temperatures could impact the quantitation of some infection parameters. While this method does not provide the number of bacteria within specific locations of the eye, when combined with histological methods, we can estimate where bacteria might be localized. Localization of B. cereus in the eye during endophthalmitis has been reported in rabbits, whose eyes are larger and more easily dissected into subcompartments.2
Intravitreal injection mimics the delivery of organisms to the posterior segment of the eye, which initiates infection. This initial step facilitates the qualitative and quantitative study of infection parameters in a highly reproducible mouse model of experimental endophthalmitis. These models are also used to estimate ocular inflammation by quantifying myeloperoxidase in infiltrating neutrophils, identifying specific cell types by flow cytometry, quantifying cytokines and chemokines by real-time PCR and/or ELISA, and observing ocular architecture by histopathology1,6,19,20,26,27,34,35,38. Infected mouse eyes are harvested with different diluents depending on the type of infection parameters to be measured. For gene expression analysis, a different lysis buffer is required26. For histopathological examination, harvested eyes are placed in a fixative solution. The multitude of genetic knockout mice also facilitates the study the role of various immune factors and cells. Intravitreal injection is therefore a mainstay technique for researchers in the field of intraocular infections and therapeutics.
The authors have nothing to disclose.
The authors thank Dr. Feng Li and Mark Dittmar (OUHSC P30 Live Animal Imaging Core, Dean A. McGee Eye Institute, Oklahoma City, OK, USA) for their assistance. Our research has been supported by National Institutes of Health grants R01EY028810, R01EY028066, R01EY025947, and R01EY024140. Our research has also been supported by P30EY21725 (NIH CORE grant for Live Animal Imaging and Analysis, Molecular Biology, and Cellular Imaging). Our research has also been supported by the NEI Vision Science Pre-doctoral Trainee program 5T32EY023202, a Presbyterian Health Foundation Research Support grant, and an unrestricted grant to the Dean A. McGee Eye Institute from Research to Prevent Blindness.
2-20 µL pipette | RANIN | L0696003G | NA |
37oC Incubator | Fisher Scientific | 11-690-625D | NA |
Bacto Brain Heart Infusion | BD | 90003-032 | NA |
Cell Microinjector | MicroData Instrument, Inc. | PM2000 | NA |
Fine tip forceps | Thermo Fisher Scientific | 12-000-122 | NA |
Glass beads 1.0 mm | BioSpec | 11079110 | NA |
Incubator Shaker | New Brunswick Scientific | NB-I2400 | NA |
Microcapillary Pipets 5 Microliters | Kimble | 71900-5 | NA |
Micro-Pipette Beveler | Sutter Instrument Co. | BV-10 | NA |
Microscope Axiostar Plus | Zeiss | NA | |
Microscope OPMI Lumera | Zeiss | NA | |
Mini-Beadbeater-16 | BioSpec | Model 607 | NA |
Multichannel pipette 30-300 µL | Biohit | 15626090 | NA |
Multichannel pipette 5-100 µL | Biohit | 9143724 | NA |
Needle/Pipette Puller | Kopf | 730 | NA |
PBS | GIBCO | 1897315 | Molecular grade |
Protease Inhibitor Cocktail | Roche | 4693159001 | Molecular grade |
Reverse action forceps | Katena | K5-8228 | NA |