Doi:10.1016/j.tim.2004.07.004

Human intestinal bacteria as reservoirsfor antibiotic resistance genes Abigail A. Salyers, Anamika Gupta and Yanping Wang Department of Microbiology, University of Illinois, Urbana, IL 61801, USA Human intestinal bacteria have many roles in human through the human colon on a regular basis are pathogens health, most of which are beneficial or neutral for the such as Streptococcus pneumoniae and Staphylococcus host. In this review, we explore a more sinister side of aureus; these bacteria are normally found in the nose or intestinal bacteria; their role as traffickers in antibiotic throat but can pass through the colon if swallowed resistance genes. Evidence is accumulating to support Until recently, such bacteria were thought to be transients the hypothesis that intestinal bacteria not only that spent little time in the human colon, but some recent exchange resistance genes among themselves but reports suggest that S. aureus might transiently or might also interact with bacteria that are passing persistently colonize the human colon in low numbers, through the colon, causing these bacteria to acquire especially in hospitalized patients In addition to and transmit antibiotic resistance genes.
such pathogenic transients, there are potentially patho-genic members of the intestinal microflora itself, such as Until recently, bacterial pathogens were the primary focus Escherichia coli, Enterococcus species, Clostridium of studies of antibiotic resistance genes and their spread.
species and Bacteroides species . Is it possible Now, scientists are starting to wonder whether this focus that such diverse bacteria can and do regularly exchange is too narrow. Could the microflora of the human colon, DNA under conditions found in the human colon? normally considered innocuous or beneficial, be playing amore sinister role in human health as reservoirs for Assessing the extent to which resistance gene transfer antibiotic resistance genes? The reservoir hypothesis is depicted in According to this view, human How can the actual extent of resistance gene transfer in intestinal bacteria not only share resistance genes among the human colon be assessed? One approach would be to themselves but can also acquire from or donate resistance feed people resistant animal bacteria, then determine genes to bacteria that are just passing through the whether genes carried by these bacteria enter human intestine . The possibility that resistance gene spread colonic bacteria. This approach has not been taken for two in the human colon might be a serious threat to human obvious reasons. First, such an experiment would be health was first raised in connection with post-surgical considered unethical in most countries. Second, such a infections, which are usually caused by the normal study would be prohibitively expensive, especially in view microflora of the patient or the patient’s caretakers .
of the fact that it is not clear how long the duration of the Recently, concern about resistance gene transfers in the sample collection period should be. Such a study could human colon has expanded to include agriculture alternatively be done in laboratory animals. Notably, There is no question that feeding antibiotics to livestock to enhance an animal’s growth selects for antibiotic resistantbacteria in the animal’s intestine but to what extent are such bacteria a threat to human health? Afterall, farms are located at a considerable distance fromplaces, such as cities, where high concentrations of people are found. Nonetheless, there is a very significant link between farm and city: the food supply. It is now wellestablished that antibiotic resistant bacteria from chick- ens, pigs and cattle enter the food supply and can be foundin meat offered for sale in supermarkets Figure 1. The resistance gene reservoir hypothesis. Bacteria that normally reside inthe human colon, most of which are normally benign, transfer resistance genes If these foods are not properly cooked, the resistant among themselves. This type of transfer becomes a problem if the commensals, bacteria will enter the intestinal tracts of consumers and many of which are opportunistic pathogens, go on to cause post-surgicalinfections. Bacteria that are merely passing through the human colon will be in will have the opportunity to commingle with members of transit through the colon long enough to transfer or acquire genes by conjugation.
the resident human microflora . Also passing These bacteria might return to the sites where they are usually found (e.g. themouth and skin) by contamination of these sites with excreted bacteria. Gene Corresponding author: Abigail A. Salyers (abigails@uiuc.edu).
transfer could be occurring in the mouth, where thick biofilms are found, but here we focus on the colon for simplicity.
0966-842X/$ - see front matter Q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.tim.2004.07.004 there have been a few reports of resistance gene transfer subgroup of these strains revealed that all the tetQ genes in laboratory animals. For example, transfer of antibiotic were at least 94% identical and the ermF genes had even resistance genes among enterococci has been demon- fewer sequence differences. The increased carriage of strated in germfree rodents In another study, a these genes over the past three decades indicated that species of Lactococcus was eliminated from the normal horizontal transfer was occurring frequently enough to microflora of rodents and replaced by an antibiotic have spread these genes widely over a time period that is resistant strain of the same species . Transfer of a small in evolutionary terms . In addition, the fact that plasmid carrying an antibiotic resistance gene was the genes were found in people with no recent history of demonstrated in this model. A problem with interpreting antibiotic use (community isolates) indicates that these the results of studies that use laboratory rodents is that genes are maintained stably once they are acquired. This the normal microflora of rodents is different from that of is bad news for those who have accepted as an article of humans. Also, the use of germfree animals eliminates the faith that susceptible strains are inherently more fit than bulk of the microflora that is normally present and focuses resistant ones and that stopping use of an antibiotic would attention on normally minor populations that in the inevitably result in the disappearance of the resistant germfree animal become the predominant bacteria.
strain. Importantly, stably maintained antibiotic resist- Given the problems inherent in such prospective ance genes are fairly common and in many cases resistant studies, a second, retrospective approach has been bacteria hold their own quite well when faced with taken, where DNA sequences were determined for competition with resistant strains .
resistance genes found in different bacteria in the To determine what type of element was transferring human colon or in other sites and were compared. The tetQ, DNA from some of the strains carrying tetQ was assumption is that if genes found in two different bacterial probed with DNA from a conjugative transposon known to species are at least 95% identical, then this gene must carry tetQ, known as CTnDOT. The digest pattern on a have been transferred horizontally by one means or Southern blot was consistent with tetQ being part of a another. The 95% cutoff is somewhat arbitrary, but finding conjugative transposon in the CTnDOT family nearly identical sequences in different species rules out CTnDOT and related conjugative transposons have an the possibility of convergent evolution, where selective unusual feature. Their transfer is stimulated 100–1000 pressures on different bacteria produce proteins that are fold by tetracycline . Therefore, it is possible that virtually identical even though they evolved separately.
the extensive transfer of tetQ, which has occurred during Because selection is for a specific amino acid sequence and the past three decades, was triggered by the use of not for a specific DNA sequence, third base differences enable the DNA sequences that encode identical proteins However, there could be other stimulatory conditions. A surprising finding of this survey was that even during thepre-1970 years, when tetracycline had not yet been widely Getting the goods on intestinal anaerobesIn our studies, we chose to focus on Bacteroides species.
used, the carriage of tetQ was as high as 20–30% ().
Similar to the strains found in the 1990 s, tetQ was carried the human colon Because of their high concen- on a CTnDOT type element in the earlier isolates.
tration, they appear most likely to be involved in Consequently, horizontal gene transfer was occurring horizontal gene transfer events. Bacteroides species even before heavy use of tetracycline was begun .
harbor two types of conjugative elements: conjugative Another surprising outcome of this survey was the plasmids and conjugative transposons .
finding that two other erm genes in addition to ermF – Initially, two resistance genes were monitored in natural ermB and ermG – had moved into Bacteroides isolates of Bacteroides species: tetQ and ermF To species during the period between 1970 and the 1990 s date, tetQ has only been found on conjugative transposons (). These erm genes had previously been associated in Bacteroides spp. The ermF gene has been found on both almost exclusively with Gram-positive bacteria, not with conjugative transposons and conjugative plasmids the Gram-negative Bacteroides species Sub- . A survey of 289 strains, representing more than sequent studies have shown that ermB and ermG are 10 different Bacteroides species, revealed that in the carried on conjugative transposons that are unrelated at period before 1970, 20–30% of isolates carried tetQ, the DNA sequence level to any conjugative transposons whereas ermF was found only rarely. By contrast, in the found in Bacteroides species to date . This raises 1990 s, over 80% of strains carried tetQ and 15% carried the possibility that ermB and ermG have entered ermF () . DNA sequence analysis of a selected Bacteroides species from Gram-positive bacteria.
Table 1. Prevalence of tetQ, ermF, ermG and ermB genes in colonic Bacteroides spp.
aBacteroides isolates from the colon of people who were healthy and did not have a history of recent antibiotic use.
bIsolates from people with Bacteroides infections.
Transactions between major populations of intestinal Research of the sort described in this review has been As already mentioned, Bacteroides species account for controversial because it can be interpreted as support for w20–30% of bacteria isolated from the human colon. Most concerns about possible effects of agricultural use of of the remaining 70–80% of colonic isolates consists of antibiotics on bacteria that cause human infections.
poorly characterized Gram-positive anaerobes. The well- People in the animal agricultural field are quick to point studied facultative species, such as E. coli and the out that currently there is no ‘smoking gun’ linking the enterococci, are numerically minor, constituting less use of antibiotics on the farm with the appearance of than 1% of colonic isolates A question that needs to resistance genes in human pathogens and subsequent be answered is what types of conjugative elements are treatment failures resulting from agriculture-associated found in the Gram-positive anaerobes, and are these resistance gene transfer. It is important to remember, bacteria participating in a significant way in the exchange however, that absence of evidence is not the same as of antibiotic resistance genes among bacteria in the evidence of absence. The diversity of bacteria found in the various microfloras of the human and animal body is The original question raised in this article was whether staggering. Most of these bacteria and the resistance members of the normal flora of the human intestine could genes they carry have not been studied a great deal. The exchange genes with bacterial pathogens that might be fact that conjugative transposons, which were unknown present in low numbers or just passing through the until recently, appear to be a driving force in the transfer intestine. Evidence that such transfers can and do happen of antibiotic resistance genes in the human body is is summarized in where several genes are shown unexpected and illustrates the principle that scientists along with the genera in which they have been found. In know a lot less than they think they do about mechanisms the case of the erm genes, the sequence identity of the of horizontal gene transfer in nature . It is genes found in different species is usually 99% or higher.
important that we establish what selective or stimulatorypressures are driving the spread of antibiotic resistance The ermB gene has been found in a variety of pathogenic genes so that we can assess the relative contributions of Gram-positive bacteria, including Streptococcus pneumo- different types of antibiotic use or other human activities niae and Clostridium perfringens The ermB and to this spread. Only then will it be possible to design ermG genes were found in more than one species, there- effective strategies for preventing further increases in the fore these genes appear to have entered Bacteroides incidence of antibiotic-resistant bacteria.
species more than once in the past One of thefew reports that implicates the human colonic Gram-positive anaerobes in resistance gene transfer is a recent report of vancomycin resistance genes, such as those This article has focused on the transfer of antibiotic found in pathogenic enterococci, in the colonic anaerobe resistance genes in nature, but gene transfer among It is also worth noting that bacteria from different sites appear to be exchanging genes. For example, Porphyro-monas gingivalis is an oral anaerobe and Prevotella ruminicola is normally found in the rumen of cattle and the intestines of pigs. It is easy to imagine howP. gingivalis might be involved in gene transfers because it is constantly being swallowed, and strains from the colon could be reintroduced into the mouth by the fecal- Streptococcus spp.
Clostridia spp.
oral route. How a rumen anaerobe, which is very sensitiveto oxygen, made contact with human colonic bacteria (if that is where P. ruminicola picked up tetQ) and then made it back into the animal is harder to imagine.
A caveat is in order. The strategy of using virtually identical genes found in different genera and species to deduce that there is some genetic conduit open betweenthose species has a couple of limitations. First, it is usually not possible to ascertain the direction of the transfer. The only reason we feel comfortable about saying that ermG and ermB appear to have entered Bacteroides species from some other species of bacteria is that the pre-1970 strainsdid not carry these genes. A second limitation is that there Figure 2. Evidence that transfer of resistance genes occurs between Gram-positive is no way to ascertain how many transfers it took for a and Gram-negative bacteria in the mammalian colon and in other environmentalsites. This evidence consists of finding virtually identical resistance genes in natural gene to move between two of the species shown in .
isolates representing different bacterial species. The resistance gene is contained in Moreover, there are almost certainly unknown players, the oval that connects the Gram-positive (left boxes) and Gram-negative (right such as the Gram positive colonic anaerobes and even soil boxes) bacterial species, in which virtually identical copies of the resistance genehave been found. Examples of possible transfer events are shown here, however, this is not meant as a complete listing of all cases found in the literature.
bacteria has broader reaching consequences. Plasmids 14 Glynn, M.K. et al. (1998) Emergence of multidrug-resistance Salmo- and conjugative transposons carry genes in addition to nella enterica serotype Typhimurium DT104 infections in the UnitedStates. N. Engl. J. Med. 338, 1333–1338 antibiotic resistance genes, such as nitrogen fixation 15 Coque, T.M. et al. (1996) Vancomycin-resistant entrococci from genes that can alter the metabolic potential of a bacterial nosocomial, community and animal sources in the United States.
cell Conjugal elements can also carry virulence Antimicrob. Agents Chemother. 40, 2605–2609 factors, such as toxin genes. For example, plasmids found 16 Moubareck, C. et al. (2003) Multiple antibiotic resistance gene in Bacillus anthracis (pOX1 and pOX2) have made this transfer from animal to human enterococci in the digestive tract ofgnotobiotic mice. Antimicrob. Agents Chemother. 47, 2993–2996 species much more pathogenic than its very close relative 17 Doucet-Populaire, F. et al. (1992) Conjugal transfer of plasmid DNA Bacillus cereus . Yersinia pestis, the cause of plague, from Enterococcus faecalis to Escherichia coli in digestive tracts of and Salmonella typhimurium strain LT2 have also gnotobiotic mice. Antimicrob. Agents Chemother. 36, 502–504 acquired plasmids that make them virulent for humans.
18 Moore, W.E. and Holdeman, L.V. (1974) Human fecal flora: the normal Perhaps the most spectacular example of horizontal gene flora of 20 Japanese-Hawaiians. Appl. Microbiol. 27, 961–979 transfer to date is the 500 kbp conjugative transposon of 19 Salyers, A.A. (1993) Gene transfer in the mammalian intestinal tract.
Mesorhizobium loti strain R7A that carries genes import- 20 Tannock, G.W. et al. (1997) Effect of sodium taurocholate on the in ant for symbiosis between the rhizobia and plants .
vitro growth of lactobacilli. Microb. Ecol. 33, 163–167 There are bacteria that have chromosomes that are only 21 Moore, W.E. et al. (1978) Some current concepts in intestinal bacteriology. Am. J. Clin. Nutr. 31, 33–42 w500 bp in size therefore the transfer of this largeconjugative transposon in soil is equivalent to the transfer 22 Salyers, A.A. (1984) Bacteroides of the human lower intestinal tract.
of an entire bacterial chromosome. However, despite all of 23 Reysset, G. et al. (1992) Genetic and molecular analysis of pIP417 and this rampant ‘bacterial sex’, horizontal gene transfer does pIP419: Bacteroides plasmids encoding 5-nitroimidazole resistance.
not appear to have homogenized bacteria. Genetic diver- sity and a well-defined phylogenetic tree for bacteria are 24 Smith, C.J. et al. (1995) Nucleotide sequence determination and still the rule rather than the exception genetic analysis of the Bacteroides plasmid, pBI143. Plasmid 34,211–222 25 Trinh, S. et al. (1995) Plasmids pIP419 and pIP421 from Bacteroides: 5-nitroimidazole resistance genes and their upstream insertion sequence elements. Microbiol. 141, 927–935 Much of the work described in this article was supported by a grant 26 Novicki, T.J. and Hecht, D.W. (1995) Characterization and DNA (AI 22383) from the U.S. National Institutes of Health.
sequence of the mobilization region of pLV22a from Bacteroidesfragilis. J. Bacteriol. 177, 4466–4473 27 Trinh, S. et al. (1996) Conjugal transfer of the 5-nitroimidazole resistance plasmid pIP417 from Bacteroides vulgatus BV-17: charac-terization and nucleotide sequence analysis of the mobilization region.
1 van den Braak, N. et al. (1998) Molecular characterization of vancomycin-resistant enterococci from hospitalized patients and 28 Trinh, S. and Reysset, G. (1997) Identification and DNA sequence of poultry products in The Netherlands. J. Clin. Microbiol. 36,1927–1932 the mobilization region of the 5-nitroimidazole resistance plasmid 2 Teuber, M. et al. (1999) Acquired antibiotic resistance in lactic acid pIP421 from Bacteroides fragilis. J. Bacteriol. 179, 4071–4074 bacteria from food. Antonie Van Leeuwenhoek 76, 115–137 29 Whittle, G. et al. (2002) The role of Bacteroides conjugative 3 Simonsen, G.S. et al. (1998) Transmission of VanA-type vancomy- transposons in the dissemination of antibiotic resistance genes. Cell.
cin-resistant enterococci and vanA resistance elements between chicken and humans at avoparcin-exposed farms. Microb. Drug 30 Bedzyk, L.A. et al. (1992) Insertion and excision of Bacteroides conjugative chromosomal elements. J. Bacteriol. 174, 166–172 4 Woodford, N. (1998) Glycopeptide-resistant enterococci: a decade of 31 Halula, M. and Macrina, F.L. (1990) Tn5030: a conjugative transposon experience. J. Med. Microbiol. 47, 849–862 conferring clindamycin resistance in Bacteroides species. Rev. Infect.
5 Sullivan, A. et al. (2001) Effect of antimicrobial agents on the Dis. 12 (Suppl. 2), S235–S242(Suppl. 2) ecological balance of human microflora. Lancet Infect. Dis. 1, 101–114 32 Shoemaker, N.B. et al. (1989) Cloning and characterization of a 6 McDonald, L.C. et al. (1997) Vancomycin-resistant enterococci outside Bacteroides conjugal tetracycline- erythromycin resistance element by the health care setting: Prevalence, sources and public health using a shuttle cosmid vector. J. Bacteriol. 171, 1294–1302 implications. Emerg. Infect. Dis. 3, 311–317 33 Weisblum, B. (1995) Erythromycin resistance by ribosome modifi- 7 Ferber, D. (2003) Antibiotic resistance: WHO advises kicking the cation. Antimicrob. Agents Chemother. 39, 577–585 livestock antibiotic habit. Science 301, 1027 34 Shoemaker, N.B. et al. (1985) Evidence that the clindamycin- 8 Witte, W. (1998) Medical consequences of antibiotic use in agriculture.
erythromycin resistance gene of Bacteroides plasmid pBF4 is on a transposable element. J. Bacteriol. 162, 626–632 9 Butaye, P. et al. (2003) Antimicrobial growth promoters used in animal 35 Shoemaker, N.B. et al. (2001) Evidence for extensive resistance gene feed: effects of less well known antibiotics on gram-positive bacteria.
transfer among Bacteroides spp. and among Bacteroides and other genera in the human colon. Appl. Environ. Microbiol. 67, 561–568 10 Huyke, M.M. et al. (1998) Multiple-drug resistant enterococci: the 36 Whittle, G. et al. (2001) Characterization of the 13 kb ermF region of nature of the problem and an agenda for the future. Emerg. Infect. Dis.
Bacteroides conjugative transposon, CTnDOT. Appl. Environ. Micro- 11 Salyers, A.A. and McManus, P. (2002) Agricultural use of antibiotics: 37 Chung, W.O. et al. (1999) Host range of the ermF rRNA methylase possible impact on antibiotic resistance in human pathogens. In gene in bacteria of human and animal origin. J. Antimicrob. Che- Antibiotic Resistance and Antibiotic Development (Hughes, D. and Anderson, D. eds), Harwood Academic Publishers 38 Salyers, A.A. and Amabile-Cuevas, C.F. (1997) Minireview: why are 12 Salyers, A.A. (2002) The Ecology of Antibiotic Resistance Genes, antibiotic resistance genes so resistant to elimination? Antimicrob.
13 Aarestrup, F.M. et al. (2000) Associations between the use of 39 Shoemaker, N.B. et al. (1992) Evidence for natural transfer of a antimicrobial agents for growth promotion and the occurence of tetracycline resistance gene between bacteria from the human colon resistance among Enterococcus faecium from broilers and pigs in and bacteria from the bovine rumen. Appl. Environ. Microbiol. 58, Denmark. Microbiol. Drug Resist. 6, 63–70 40 Salyers, A.A. and Shoemaker, N.B. (1996) Resistance gene transfer in 48 Farrow, K.A. et al. (2000) The macrolide-lincosamide-streptogramin B anaerobes: new insights, new problems. Clin. Infect. Dis. 23 (Suppl. 1), resistance determinant from Clostridium difficile 630 contains two erm(B) genes. Antimicrob. Agents Chemother. 44, 411–413 41 Salyers, A.A. et al. (1995) In the driver’s seat: the Bacteroides 49 Stinear, T.P. et al. (2001) Enterococcal vanB resistance locus in conjugative transposons and the elements they mobilize. J. Bacteriol.
anaerobic bacteria in human faeces. Lancet 357, 855–856 50 Salyers, A.A. and Shoemaker, N.B. (1997) Conjugative transposons.
42 Cooper, A.J. et al. (1996) The erythromycin resistance gene from the Bacteroides conjugal transposon Tcr Emr 7853 is nearly identical to 51 de la Cruz, F. and Davies, J. (2000) Horizontal gene transfer and ermG from Bacillus sphaericus. Antimicrob. Agents Chemother. 40, the origin of the species: lessons from bacteria. Trends Microbiol.
52 Sullivan, J.T. et al. (2002) Comparative sequence analysis of the 43 Monod, M. et al. (1987) Cloning and analysis of ermG, a new symbiosis island of Mesorhizobium loti strain R7A. J. Bacteriol. 184, macrolide-lincosamide-streptogramin B resistance element from Bacillus sphaericus. J. Bacteriol. 169, 340–350 53 Martinez-Romero, E. and Caballero-Mellado, J. (1996) Rhizobium 44 Jensen, L.B. et al. (1999) Presence of erm gene classes in gram-positive phylogenies and bacterial genetic diversity. Crit. Rev. Plant Sci. 15, bacteria of animal and human origin in Denmark. FEMS Microbiol.
54 Salyers, A.A. and Whitt, D.D. (2002) Bacterial Pathogenesis, ASM 45 Nishijima, T. et al. (1999) Distribution of mefE and ermB genes in macrolide-resistant strains of Streptococcus pneumoniae and their 55 Sullivan, J.T. and Ronson, C.W. (1998) Evolution of rhizobia by variable suseptibility to various antibiotics. J. Antimicrob. Chemother.
acquisition of a 500-kb symbiosis island that intefrates into a phe- tRNA gene. Proc. Natl. Acad. Sci. U. S. A. 95, 5145–5149 46 Wang, Y. et al. (2003) A newly discovered Bacteroides conjugative 56 Freiberg, C. et al. (1997) Molecular basis of symbiosis between transposon, CTnGERM1, contains genes also found in Gram-positive Rhizobium and legumes. Nature 387, 394–401 Bacteria. Appl. Environ. Microbiol. 69, 4595–4603 57 Koonin, E.V. (2000) How many genes can make a cell: the minimal- 47 Gupta, A. et al. (2003) A new Bacteroides conjugative transpo- gene-set concept. Annu. Rev. Genomics Hum. Genet. 1, 99–116 son that carries an ermB gene. Appl. Environ. Microbiol. 69, 58 Wiener, P. et al. (1998) Evidence for transfer of antibiotic-resistance genes in soil populations of streptomycetes. Mol. Ecol. 7, 1205–1216 Elsevier.com – Dynamic New Site Links Scientists to New Research & Thinking Elsevier.com has had a makeover, inside and out. Designed for scientists’ information needs, the new site, launched in January, ispowered by the latest technology with customer-focused navigation and an intuitive architecture for an improved user experience andgreater productivity.
Elsevier.com’s easy-to-use navigational tools and structure connect scientists with vital information – all from one entry point. Users canperform rapid and precise searches with our advanced search functionality, using the FAST technology of Scirus.com, the free sciencesearch engine. For example, users can define their searches by any number of criteria to pinpoint information and resources. Search by aspecific author or editor, book publication date, subject area – life sciences, health sciences, physical sciences and social sciences – or byproduct type. Elsevier’s portfolio includes more than 1800 Elsevier journals, 2200 new books per year, and a range of innovativeelectronic products. In addition, tailored content for authors, editors and librarians provides up-to-the-minute news, updates onfunctionality and new products, e-alerts and services, as well as relevant events.
Elsevier is proud to be a partner with the scientific and medical community. Find out more about who we are in the About section: ourmission and values and how we support the STM community worldwide through partnerships with libraries and other publishers, andgrant awards from The Elsevier Foundation.
As a world-leading publisher of scientific, technical and health information, Elsevier is dedicated to linking researchers and professionalsto the best thinking in their fields. We offer the widest and deepest coverage in a range of media types to enhance cross-pollination ofinformation, breakthroughs in research and discovery, and the sharing and preservation of knowledge. Visit us at Elsevier.com.
Elsevier. Building Insights. Breaking Boundaries.

Source: http://www.haslberger.eu/images/stories/FoodBacteriaHealth/intest_res.pdf