Bacterial Culture and Inoculation of Mice (Simple Infection)
Standardization of bacterial culture is crucial for in vivo experiments addressing Helicobacter/host interaction. Here we present methods for bacteria culture and infection of mice.
Key wordsSimple infection Animal models In vitro bacterial culture
Mice were first successfully colonized with a disease-producing strain of H. pylori in 1997. Since that time, a number of different mouse models have been described utilizing different mouse and bacterial strains and inoculation protocols, as well as different methods of outcome assessment. In this chapter, we describe methods of bacterial inoculation. In subsequent chapters we describe adoptive transfer of lymphocytes and tissue collection and processing, including isolation of bacteria, DNA, RNA, and splenic and gastric immune cells for analysis. In addition, we address differences in mouse and bacterial strains, and discuss morphologic evaluation of mouse models in the context of human disease.
The laboratory mouse is currently by far the animal model most commonly used for the study of human disease. Mice are small with a rapid generation time and are genetically manipulable, resulting in dozens of available inbred strains and genetically engineered mutants designed to evaluate specific mammalian genes in the context of infectious or other diseases. Immunological reagents that react with mouse tissues are readily available, and reproducible methods are available for modeling many diseases in mice. In spite of the overall utility of mice as models of human disease, there are a number of factors that must be considered when interpreting experimental results of mouse models. Because they are mammals, mice are similar to humans physiologically and in many features of their immune responses. However, specific differences between mice and humans must be taken into account when interpreting experimental results. Also, mice, like other nonhuman mammals, are not always susceptible to infection by human pathogens, and when infection occurs, the manifestations of disease are rarely, if ever, exactly the same as they are in human patients. Infection of mice with cognate mouse pathogens can sometimes circumvent these issues, but may introduce additional interpretive difficulties because of physiologic and molecular differences between the mouse and human pathogens themselves. For example, mice are more susceptible to gastritis due to a related organism, Helicobacter felis, than they are to H. pylori, making H. felis a useful tool in investigation of host immune responses to gastric helicobacters (1) as well as preneoplastic changes (2). However, H. felis lacks many bacterial genes and functions that are of interest in the pathogenesis of disease due to H. pylori (most prominently the cag pathogenicity island, a series of genes that is epidemiologically associated with severity of disease) (3), limiting the usefulness of H. felis for evaluation of specific bacterial virulence factors. Thus, use and interpretation of mouse models of human disease must be carried out and interpreted with a clear knowledge and understanding of the differences between mice and humans and/or the differences between mouse and human pathogens. That said, mouse models of human disease have been and are increasingly central in our ability to understand basic mechanisms of disease in mammals, and can form the basis for investigation and study of disease in the natural human host.
Prior to 1997, the principal animal models used for study of disease due to H. pylori were germ-free piglets (4, 5) and nonhuman primates (6). While these models were pivotal in identifying H. pylori virulence determinants and the basis of host immune responses, they were expensive, and their use was limited to those investigators who had access to the necessary institutional resources. A few published studies reporting colonization of mice by H. pylori had been published, but colonization rates were low, disease manifestations were nonexistent, and results were generally discouraging (7, 8, 9, 10). In 1997 Lee et al. described in detail a mouse model of H. pylori in which a mouse-adapted strain, originally called HP8 and later called SS1, was used (11). The strain was isolated from a human patient and then passed in mice resulting in consistently high rates of colonization (107 cfu/g of stomach), colonization of several inbred mouse strains, and chronic progressive gastritis with many features that were similar to the most common manifestations of disease in humans: slowly progressive and subclinical chronic or chronic active gastritis. Thus, strain SS1 became the most commonly used strain for mouse models, and is the strain that is most commonly used by our laboratory (for comments on mouse-adapted H. pylori strains, see Note 1).
2 Types and Uses of Mouse Models
Current use of mouse models of H. pylori may be roughly divided into two categories, referred to here as “simple infection” and “adoptive transfer” (seeChapter 14). Simple infection involves oral inoculation of mice with live, broth-cultured H. pylori. C57BL/6J mice or congenic mutant strains are most commonly used (see Note 2), and infection results in mild, slowly progressive chronic or chronic active gastritis that is generally detectable by 8–12 weeks after inoculation, and may become severe in some mice by 6 months–1 year (12, 13, 14, 15, 16, 17, 18). This model is most useful for examination of bacterial colonization factors since gastric lesions are not present in until 8–12 weeks after inoculation, and before that time, histologic examination is unrewarding.
3 Bacterial Culture and Inoculation of Mice (Simple Infection)
Buffers and media: All solutions are made with deionized distilled water (DDW) and handled aseptically. They are autoclaved or sterilized by filtration through a 0.22 μm filter, unless otherwise indicated. Unless otherwise noted, all supplies and reagents are from Fisher Scientific or Sigma Aldrich.
3.2 Brucella broth (Difco) with 10% fetal calf serum (BB+FCS)
7 g Brucella broth powder (Difco).
225 ml DDW.
Autoclave 20 min, cool.
Add 25 ml heat-inactivated FCS.
3.3 Urease indicator broth
5 ml 0.2 M NaPO4 buffer, pH 6.5 (23 g NaH2PO4⋅2H2O + 14.2 g Na2HPO4⋅12H2O/L, adjust pH with HCl or NaOH).
6.6 ml 5 M urea (30 g/100 ml).
20 ml 0.1% Phenol Red.
1.0 ml 2% NaN3 (20 mg/100 ml).
QS to 100 ml with DDW.
3.4 Modified Skirrow’s antibiotic supplement (see Note 3)
Aseptically prepare the following solutions:
Vancomycin: 200 mg in 1 ml of DDW.
Polymyxin: 6.6 mg in 1 ml of DDW.
Trimethoprim: 50 mg in 1 ml of DDW.
Bacitracin: 400 mg in 15 ml of DDW.
Amphotericin B: 100 mg in 5 ml of 1 M NaOH.
Nalidixic Acid: 20.2 mg in 1 ml of 1 M NaOH.
Concentration of stock solution
3.5 Selective plates for marked strain
We use kanamycin and chloramphenicol for marking mutant H. pylori strains.
Both are used at 20 μg/ml final concentration.
Kanamycin (10 mg/ml) is purchased from Sigma.
Chloramphenicol stock solution is made in 100% ethanol at 10 mg/ml. Both are stored at 4°C and diluted 40 μl per plate or per 20 ml broth (see Note 4).
3.6 Other supplies
TSA sheep blood agar plates treated with modified Skirrow’s antibiotic supplement (see Note 4).
Tissue culture plates.
15 ml and 50 ml conical tubes.
4.1 Bacterial Culture and Primary Inoculation of Mice (Simple Infection)
Place 10 ml BB + FCS in a sterile 100 mm sterile plastic Petri dish.
Inoculate with 0.1 ml of thawed frozen bacterial stock or 1 loopful from a 2-day-old lawn of H. pylori SS1 (see Note 5).
Incubate overnight in a triple gas incubator (10% CO2, 5% O2, 85% N) at 37°C with gentle agitation (see Note 6).
After 18–24 h of incubation, examine the bacteria by light microscopy to evaluate motility and morphology, and perform urease and catalase tests. Bacteria for inoculation should be about 108cfu/ml, and actively motile with no clumping.
For urease and catalase tests, place two drops of culture on a glass slide and add a drop of urease broth to one and H2O2 to the other. Urease broth should turn from orange to bright cherry red in less than 5 min. H2O2 should immediately bubble.
If bacteria are present and motile, but not yet in late logarithmic growth (i.e., less than 108 cfu/ml), recheck the broth after 6–12 additional hours of incubation.
If the culture is sufficiently grown, quantify an aliquot using a standard laboratory hemocytometer.
Dilute the broth to approximately 107 cfu/ml (use NBF for dilution if bacteria are extremely motile).
Count manually by direct examination under 400× magnification. We find that this method is more accurate than optical density since optical density readings are affected by the age and morphology of the cultures. For viable plate count, seven tenfold dilutions on TSA sheep blood agar plates as described below.
Prepare the inoculum by centrifugation in a tabletop centrifuge at 2,500 × g for 20 min, and resuspend the bacterial colonies in sterile BB + FCS. Our standard protocol is to resuspend the bacteria at approximately 107 cfu/ml and orally inoculate mice with 0.1 ml (106 cfu). We have found that concentration is not critical, however, since mice are susceptible to colonization by as few as 200 cfu of H. pylori strain SS1 (unpublished data).
For inoculation, restrain the mice manually. Mice are held vertically (nose up) to prevent bending of the neck during inoculation. The inoculum is placed in a 1 cc syringe with an 18 g ball-tipped bent gavage needle, which is gently introduced into the esophagus. Care must be taken to place the tube without traumatizing the tissue or compromising the respiratory tract. Inexperienced individuals should obtain training prior to attempting this procedure. Once the gavage tube is within the esophagus (beyond the pharynx), 0.1 ml of bacterial suspension is slowly introduced and allowed to flow into the stomach. The tube is removed and the mouse held vertically briefly to allow complete swallowing. Mice should be observed for signs of difficulty during and after the procedure.
It is not necessary to fast the mice or administer acid-reducing treatment prior to bacterial inoculation. If desired, the inoculation may be repeated once or twice daily or every other day, but this is not necessary.
Mouse-adapted H. pylori. Recent studies have revealed that the early failure of mouse models noted above was most likely attributable to the natural phenotypic variability of human H. pylori isolates. Several studies have now indicated that regardless of mouse passage, bacterial virulence factors, or other considerations examined, about 20–40% of H. pylori isolates are able to colonize mice (19, 20, 21, 22, 23). Several strains have been passaged in mice, sometimes resulting in higher colonization density by the passaged isolate compared to the original clinical isolate (20, 24, 25). These strains are sometimes used in preference to SS1 because of ease of genetic manipulation, presence of specific H. pylori genes, or other factors (15, 26, 27).
It should be noted that the use of SS1 has generated controversy because of the failure of that strain to elicit IL-8 production by cultured AGS cells (22, 27, 28). This has led to the suggestion that the cag-PAI is dysfunctional in SS1, and/or that the dysfunction is associated with mouse passage and is permissive for mouse colonization. However, as noted above, other studies have failed to demonstrate an association between the cag-PAI and mouse passage or colonization (19, 20, 23, 25). As a general rule mouse-colonizing strains optimally should colonize C57BL/6J mice at about 107 cfu/g of stomach wall and be visible in Warthin–Starry or Steiner-stained histologic sections in a multifocal distribution in the gastric mucosa. Colonization may be higher in the absence of gastritis (for example, in immunodeficient mice) (12) and may differ in different mouse strains (11, 29, 30, 31).
Mouse strains. C57BL/6 mice and mutant strains on this background are the strain most commonly used for studies of H. pylori. Most studies that compared strains showed that C57BL/6 mice supported strong colonization and developed more severe gastritis than the other strains tested (11, 29, 30, 31). It should be noted, however, that both genetics and normal enteric microbiota may vary depending on the source and subline of the mice, and could result in different responses to infection by identical mouse strains (32). Our laboratory uses helicobacter-free C57BL/6J mice and mutant strains purchased from Jackson laboratories. We find that these mice respond well and produce consistent results.
Use of antibiotics. In our laboratory we use antibiotics in bacterial media only when recovering bacteria from mouse stomachs or to isolate antibiotic-marked bacterial mutants. We find that antibiotics are not necessary to prevent contamination of inocula. Therefore, we prefer to avoid antibiotic use, thereby minimizing to unnecessary selection pressure on laboratory-passaged strains. For recovery of bacteria from contaminated samples (i.e., mouse stomachs), we find the modified Skirrow’s mixture sufficient for most samples. Some strains of mouse stomach bacteria are able to grow in the presence of these antibiotics. However, complete removal of the squamous mucosa prior to homogenization and plating of the tissue usually removes enough of these organisms to permit quantification of H. pylori colonies (see necropsy technique, below). The appearance of H. pylori colonies is sufficiently characteristic to distinguish them from contaminants. Should there be any doubt, individual colonies may be rapidly identified as H. pylori by collecting a sample on a sterile cotton swab and placing a drop of urease indicator broth on the swam. If the sample is H. pylori, the swab will turn cherry red within a few minutes.
Antibiotic-treated agar plates. As an alternative to including antibiotics in the plates during pouring, antibiotics may be added to prepurchased plates. Pipet the appropriate volume on the plate, spread the solution over the entire surface, and allow the plate to dry before use.
Inoculum preparation. For mouse inoculation it is important to pass bacteria as lawns rather than select single colonies. Bacteria, particularly H. pylori, undergo rapid and often undetectable genetic variation when grown in culture. Isolation of single colonies fixes these variations and may be associated with marked changes in colonization ability and/or disease induction. We have found that isolates of either SS1 or M6, another mouse-colonizing strain, range in their colonization fitness from 0 to 107 cfu/g of gastric mucosa after cloning and isolation of single colonies (unpublished observations). For that reason, we use bacterial lawns for inoculation, and minimize in vitro passages of our inoculum strains.
Microaerobic atmosphere. If a triple gas incubator is not available, a tissue-culture incubator set at 10% CO2 in air or a CampyPak (BD diagnostic) will suffice.
- 5.Lambert JR, Borromeo M, Pinkard KJ, Turner H, Chapman CB, Smith ML (1987) Colonization of gnotobiotic piglets with Campylobacter Pyloridis—an animal model? J Infect Dis. 155:1344Google Scholar
- 7.Ehlers S, Warrelmann M, Hahn H (1988) In search of an animal model for experimental Campylobacter pylori infection: administration of Campylobacter pylori to rodents. Zentralblatt Fur Bakteriologie 268:341–346Google Scholar
- 9.Karita M, Kouchiyama T, Okita K, Nakazawa T (1991) New small animal model for human gastric Helicobacter pylori infection: success in both nude and euthymic mice. Am J Gastroentero 86:1596–1603Google Scholar
- 24.Shi Y, Liu XF, Zhuang Y, Zhang JY, Liu T, Yin Z, Wu C, Mao XH, Jia KR, Wang FJ, Guo H, Flavell RA, Zhao Z, Liu KY, Xiao B, Guo Y, Zhang WJ, Zhou WY, Guo G, Zou QM (2010) Helicobacter pylori-induced Th17 responses modulate Th1 cell responses, benefit bacterial growth, and contribute to pathology in mice. J Immunol 184:5121–5129PubMedCrossRefGoogle Scholar