Review Article
Gut Microbiome and Antibiotics

https://doi.org/10.1016/j.arcmed.2017.11.004Get rights and content

Despite that the human gastrointestinal tract is the most populated ecological niche by bacteria in the human body, much is still unknown about its characteristics. This site is highly susceptible to the effects of many external factors that may affect in the quality and the quantity of the microbiome. Specific factors such as diet, personal hygiene, pharmacological drugs and the use of antibiotics can produce a significant impact on the gut microbiota. The effect of these factors is more relevant early in life, when the gut microbiota has not yet fully established. In this review, we discussed the effect of type and doses of the antibiotics on the gut microbiota and what the major consequences in the use and abuse of these antimicrobial agents.

Introduction

More than 100 trillion microorganisms populate the human adult large intestine, with over 1000 species occupying this region and the majority of them are able to survive under anaerobic conditions 1, 2. About 90% of microbes colonizing the gut are represented by only six phyla, Firmicutes, Bacteroides, Actinobacteria, Fusobacteria, Proteobacteria, and Verrucomicrobia 3, 4. Although microbiota composition in the gut is influenced by diet, gender, geographic location, and ethnicity 5, 6, delivery mode appears to be the most critical and influential factor when acquiring our primary microbiota. Babies are coated by the maternal vaginal and gut microbiota at birth which is mainly dominated by Lactobacillus, composing more than 50% of the total microbiota 7, 8. After birth, babies acquire their secondary microbiome from their family members or the surrounding ecosystem. However, it is the initial microbiota that will influence the microbiome composition later in life, with long-lasting effects on the microbial community and host physiology (6). The period of microbiota acquisition coincides with development of a child's immune system, and both influence each other strongly, making this window in early life absolutely critical in conferring appropriate host immune responses 9, 10. Vaginally delivered babies acquire a bacterial composition resembling their mother's vaginal microbiota, dominated by Lactobacillus, Prevotella, or Sneathis (7). In contrast, babies delivered by Cesarean section acquire bacteria that resembles those present on the skin, like Staphylococcus, Corynebacterium, and Propionibacterium (7). These bacteria are not maternally derived but are delivered from hospital staff, with whom the babies have had contact (7). Several studies suggest a correlation between C section and the autoimmune system due to the essential role of the maternal microbiome on the development of the perinatal immune system 11, 12, 13, 14. However, recent publications from Swedish and Irish cohorts reported that there was no significant influence of C section on the risk of type 1 diabetes (T1D) 15, 16. Similar to T1D, as the incidence of asthma has also been increasing for the past 30–40 years in developed countries (17). The correlation between C section and the risk of asthma is a different story. Some results showed a negative correlation 18, 19; on the other hand, there were also reports of positive data 13, 14, 20, 21. Interestingly, it was critical to assess the correlation between whether C section was performed before membrane rupture or not (20). Although meta-analysis indicated that C section contributes to the risk of asthma, further studies are needed. When it comes to perturbation of microbiota in early life, C section is a good model to discuss the risk of auto-immune disease. However, T1D and asthma seem to have different correlations. This might suggest the possibility that other factors may be involved in both autoimmunity and microbiome perturbation in early life. Antibiotic treatment may be the missing link.

Many scientists have been interested in the small and large intestines for a long time, because the intestinal tract is one of the lesser known organs in terms of bacterial content, especially in regard to commensal resident bacteria. One of the major difficulties when surveying gut communities is the ability to culture the organisms. Nearly 70% of microbes colonizing the gut cannot be cultured even now (22), therefore the analysis of the microbiota has been a very difficult work for a long time. In order to assess and identify human gut microbiota, 16S rRNA sequencing has become a revolutionary and useful technique to detect diversity and abundance of the microbiome, taking advantage of curated 16S rRNA sequence databases 23, 24, 25. Purified DNA obtained from feces, rectal swabs, or other samples is amplified and barcoded for multiplex high-throughput sequencing using universal primers targeting the V (variable) regions of the bacterial 16S rRNA gene. After running PCR and adjusting the concentration, sequencing is performed using the Illumina MiSeq, HiSeq, Ion Torrent or another sequencing platform. Analysis of sequencing results using the QIIME software package is performed by clustering similar sequences (97% similarity) into operational taxonomic units (OTUs) and comparing the reference sequence against the Greengenes Core set. Finally, α-diversity (within-sample), β-diversity (between-sample), and relative abundance can be determined using this data.

After the discovery of antibiotics in the early 20th century, specifically the discovery of penicillin by Alexander Fleming in 1928, the use of antibiotic treatment have spread throughout the world, despite the fact that it took more than 10 years to purify and manufacture penicillin. The use of antimicrobial drugs to fight infectious disease in humans started when Gerhard Domak discovered sulfonamide chrysoidine in 1939. A particular event in human history, World War II, played an important role in the development and widespread use of antibiotic treatment. Since the discovery of sulfonamide chrysoidine, the last 75 years has seen a tremendous amount of people saved from previously lethal infectious diseases. Unmistakably, antibiotics are among the most important drugs used in human and animal medicine in the last 100 years. However, there are some negative aspects of antibiotic treatment that have been becoming more apparent in the last 20 years. a) Antibiotic abuse has selected for antibiotic resistance in many bacteria, and has made antibiotic treatments ineffective, claiming many lives, particularly those of compromised patients in hospital settings, like MRSA, VRE, or Clostridium difficile 26, 27. The emergence of multi-drug resistant bacteria has risen due to antibiotic abuse, with a consequent increase in mortality due to infectious disease. b) It is getting more difficult to develop new antibiotics and there aren't any indications that new antibiotics will be available in the next 5–10 years. c) A strong correlation has been observed between early antibiotic use and the development of allergic and autoimmune diseases. For example, the number of children with asthma has been steadily increasing, especially in developed countries, where antibiotics have been used for the longest period of time. In a Danish study, 78% of mothers who received antibiotic drugs from 80 weeks before pregnancy to 80 weeks post-partum showed that maternal use of antibiotic drugs was associated with an increased risk of childhood asthma (28). There was a dose-related association for the risk of the offspring acquiring asthma. Moreover, many reports indicate that prenatal antibiotic exposure promotes both weight gain in babies and an increased risk of obesity 29, 30.

T1D is an autoimmune disease that causes destruction of pancreatic beta cells and subsequent hypoinsulinemia. The incidence of T1D is increasing in developed countries (such as the Scandinavian nations, Canada, and Australia), more rapidly than in developing countries. In such developed countries, the onset of T1D is occurring earlier and earlier 31, 32. Incidence has increased 3–5% annually worldwide 33, 34, 35. The assumption is that some environmental changes have occurred in early life that is associated with T1D incidence. Early life interactions between hosts and their gut microbiota are key for the immunological development of the infant (36) and perturbations of the microbiome such as early-life antibiotic use may have an effect in the development of allergies, asthmas and T1D. Some recent evidence supports the concept of an altered microbiota in T1D (37). As mentioned above, a strong correlation between Caesarean section and the incidence of asthma has been reported (28); thus, many researchers are now focusing on the association between perturbation of the microbiome in early life, particularly due by antibiotics and T1D.

In non-obese diabetic (NOD) mice there is spontaneously develop of T1D-like and there is some evidence that microbial exposure affects diabetes incidence. In particular, it is well known that increased microbial exposures (“dirty” conditions) protect NOD mice from the development of T1D 38, 39. We have demonstrated that using antibiotics in pregnancy and early life showed enhancement and acceleration of T1D (Figure 1).

Obesity has been one of the biggest health problems for more than 20 years worldwide and it is still increasing. It is a critical situation, especially in United States. Although no one state had a prevalence of obesity higher than 15% in 1985, this has rapidly changed, and in 2015 there was no state with an obesity prevalence lower than 20%. In 2015, only 6 states had an obesity prevalence between 20 and 25%, while 4 states had an obesity prevalence of 35% or greater (Figure 2). There is a strong correlation between obesity and cardiovascular disease, type 2 diabetes, and sleep dysfunction. Recently it has been revealed that an early onset of obesity will increase the incidence of type 2 diabetes, hypertension, and chronic kidney disease (40). Therefore, it is crucial to prevent the incidence of obesity at the early stages of life.

In United States, infants have been considerably exposed to prescribed antibiotics, based on the database of 2010 (41). This report shows that antibiotic use in children, particularly in early life, has become more widespread. Based on this information, it has been estimated that children from the United States have received about three courses of antibiotic treatment before they are two years old. It is incredible that children of Unites States are exposed to such an enormous quantity of prescribed antibiotics in their early life, because this amount is about two times more than the antibiotics prescribed in children from northern European countries (42).

Antibiotic use, either orally or intravenously, influences and reduces the gut microbiome 43, 44. Each class of antibiotics has different properties and excretion systems, resulting in different patterns of alteration to the microbiome composition (Table 1). The alteration of microbiome composition depends on the antibiotic class, dose, and period of exposure. There are also differences related to their pharmacological action or target bacteria.

Macrolides are now one of the most common antibiotics used around the world in children and adults. Many patients are taking macrolides for long periods of time due to a chronic infectious disease. A Finnish report demonstrated that macrolide consumption in children led to an alteration in gut microbiota that decreased Actinobacteria and increased Bacteroides and Proteobacteria (45). Interestingly, a positive correlation was observed between children that started taking macrolides within two years of life and an increase body weight and asthma prevalence. The effect of penicillin was weaker than the effect of macrolides. Currently, clarithromycin has been the first antibiotic selected for eradication of Helicobacter pylori. A Swedish group reported a decrease in Actinobacteria and Firmicutes, with an increase in Bacteroides and Proteobacteria after H. pylori eradication (46). Even though the eradication did not succeed, alteration of the gut microbiota lasted for a long time. Another group reported the comparison between oral vancomycin and amoxicillin treatments (47). Vancomycin showed a major decrease in fecal microbiota diversity due to Firmicutes reduction and an increase in Proteobacteria. Penicillin also reduced Firmicutes but this reduction was minimal when compared to vancomycin. Furthermore, a French group reported on a large cohort of Niger children from a malnutrition project. They compared the weight gain between the placebo and the amoxicillin groups (50). Although amoxicillin decreased the risk of inpatient disease transfer, it also affected their weight gain.

Ciprofloxacin reduced Firmicutes and Actinobacteria (especially Bifidobacterium), and increased Bacteroides (48). Another group showed a comparison between Ciprofloxacin and Clindamycin, with Ciprofloxacin decreasing Bifidobacteria and Clindamycin decreasing Bifidobacteria and Lactobacilli. However, no recovery of Lactobacilli was observed in the Clindamycin group (49).

Using a mouse model, one study confirmed differences in microbiome changes caused by temporarily administering low doses of penicillin and ciprofloxacin. In mice fed with normal chow, penicillin influenced gut microbiota greater than ciprofloxacin. In addition, the perturbation of the microbiome by penicillin took five weeks to improve after five days administration. In contrast, in where the recovery of the normal microbiome was faster in ciprofloxacin (Figure 2) (51).

By the 1950's, veterinary researchers had already reported that antibiotic use in food or water promoted mammalian livestock growth 52, 53. There was no previous correlation between such growth and class of antibiotics. Tetracycline, glycopeptide, macrolides, and penicillin have been used in some livestock, such as chickens, pigs, and cows, to increase the animal's weight for more than 70 years. It is very important to mention that antibiotic use in early life has greater effect on animal growth than when used later in life. Surprisingly, researchers had already reported a similar correlation between antibiotic use and human weight growth in the 1950s 54, 55, 56. In the 1960s, a dual effect of antibiotics on weight growth was reported in a mouse model, with low dose of penicillin or oxytetracycline inducing weight gain, but high doses of the same antibiotics decreasing weight (57).

Antibiotic use in germ free animal models did not promote animal growth 58, 59, leading to the assumption that the existence of the microbiome is essential for the gain or loss of weight under antibiotic treatments. Ridaura et al. transferred twins' fecal samples where only one was obese to mice, yet only the mice who was inoculated by the fatty twin's feces had increased weight (60). They then indicated that the ratio of Firmicutes to Bacteroides was defined as the obesity index, because a proportion of increased Firmicutes to decreased Bacteroides were observed in the fatty twin's feces. Additionally, a decreased beta diversity was reported. These findings showed that the microbiome is a very important player in the gut, at the very least in terms of obesity.

Trasande et al. reported that weight gain of children depends upon periods when they took antibiotics early in life (61). Administration of antibiotics within the first 6 months of life has a strong correlation with obesity, but this correlation disappears if administration of antibiotics occurs after the first 6 months of life. The study indicated that children who were exposed to antibiotic agents until 6 months of life had a significantly higher risk of being overweight at seven years old.

A study focused on low birth weight birth infants from Taiwan (62), who were 14 d old and less than 1500 g. Erythromycin was administered at 5 mg/kg for 14 d. This study indicated that children receiving erythromycin had a significant difference of daily weight gain when compared with the placebo group.

An open label study in the U.S. and Canada that included children between 6–18 years old showed that azithromycin significantly increased the risk of weight gain when compared to the control group (63). Although it is accepted that antibiotic exposure in early life has a strong correlation with obesity, we do not know the explicit host-microbe interactions, or the necessary microbial changes that contribute to this phenotype.

A study of H. pylori eradication in adults in the U.K. included the following eradication protocol: ranitidine bismuth citrate 400 mg and clarithromycin 500 mg twice daily for two weeks. This study showed a significant increase in body mass after treatment (64).

There have been two studies about the effect of amoxicillin on weight gain 50, 65. As mentioned above, the French study could not find a significant difference in weight gain in children treated with amoxicillin in Niger (50). In another study of adults with endocarditis, patients were treated for six weeks with vancomycin + gentamycin vs amoxicillin + gentamycin by intravenous injection (65). A significant weight gain was observed in the vancomycin group, but not in the amoxicillin group.

From all these studies, we can assume that depending of the doses, there is a positive relationship between consumption of broad spectrum antibiotics and weight modification.

As mentioned earlier, there are many reports that pointed out the correlation between antibiotic treatment (especially subtherapeutic low dose treatment) and weight gain in farm animals. Cho I, et al. demonstrated that mice treated with low dose of penicillin, vancomycin, or chlortetracycline increased fat-mass and short chain fatty acid levels (66). This study suggests that microbiome alteration influences metabolic pathways. Interestingly, there were some antibiotic differences in terms of inhibiting protein synthesis.

There are many reports about the strong correlation between antibiotic treatments within the first 6 months of life and weight gain in human studies. Cox LM, et al. showed that mice treated with low dose penicillin (LDP) at birth had an increase in body mass greater than mice treated with LDP at weaning periods (36). Moreover, mice treated with LDP and high fat diet (HFD) also showed a greater impact on the perturbation of gut microbiome. Mice transplanted with feces from mice with LDP and HFD gained more weight than mice transplanted with feces from mice without antibiotics and normal chow. Mice treated with LDP gained weight and subsequently exhibited a change in hepatic metabolism of cholesterol and lipids (67).

There are also reports of some beneficial effects when antibiotics are used in low doses. Cani et al. showed that a low dose of antibiotic treatment (ampicillin and neomycin) improved metabolic endotoxemia in mice fed with HFD or in ob/ob mice (68).

There is, however, conflicting results of the effect on weight when antibiotics are administered in therapeutically doses. As we mentioned before, antibiotics in many cases have been prescribed for weight loss in humans. Muphy et al. indicated that mice treated with vancomycin decreased in weight with HFD (69). Other studies have shown a correlation between antibiotic treatment and weight loss or improvement of metabolic changes (70). In contrast, Nobel et al indicated that administration of therapeutically doses of antibiotics in mice early in life was associated with a significant increase in weight (71) (Figure 3) (Table 2).

In summary, there are many reports indicating a positive correlation between antibiotic treatment and weight gain in both human and animal studies. The conflicting results reported in some studies may be related to the type of antibiotic used as well as when the antibiotic is administrated. Therefore, physicians need to think more carefully about over what periods of life and what kind of antibiotics they should prescribe.

Section snippets

Conclusions

Antibiotics are one of the greatest discoveries in Medicine and the benefits in the reduction of the mortality of the infectious diseases is unquestionable. However, as result of their multiple benefits, antibiotics have been wildly used and abused in the number of prescriptions, particularly in pediatric population and in the success of antibiotics for growth promotion in the food industry. We are now witnessing major changes in the epidemiology of chronic and autoimmune diseases that cannot

References (71)

  • A.J. Stewardson et al.

    Collateral damage from oral ciprofloxacin versus nitrofurantoin in outpatients with urinary tract infections: a culture-free analysis of gut microbiota

    Clin Microbiol Infect

    (2015)
  • P.R. Moore et al.

    Use of sulfasuxidine, streptothricin, and streptomycin in nutritional studies with the chick

    J Biol Chem

    (1946)
  • T.H. Haight et al.

    Effect of prolonged antibiotic administration of the weight of healthy young males

    J Nutr

    (1955)
  • Y.Y. Ng et al.

    Efficacy of intermediate-dose oral erythromycin on very low birth weight infants with feeding intolerance

    Pediatr Neonatol

    (2012)
  • F. Sommer et al.

    The gut microbiota–masters of host development and physiology

    Nat Rev Microbiol

    (2013)
  • M. Arumugam et al.

    Enterotypes of the human gut microbiome

    Nature

    (2011)
  • M. Rajilic-Stojanovic et al.

    The first 1000 cultured species of the human gastrointestinal microbiota

    FEMS Microbiol Rev

    (2014)
  • T. Yatsunenko et al.

    Human gut microbiome viewed across age and geography

    Nature

    (2012)
  • J.E. Koenig et al.

    Succession of microbial consortia in the developing infant gut microbiome

    Proc Natl Acad Sci U S A

    (2011)
  • M.G. Dominguez-Bello et al.

    Delivery mode shapes the acquisition and structure of the initial microbiota across multiple body habitats in newborns

    Proc Natl Acad Sci U S A

    (2010)
  • S. Zeissig et al.

    Life at the beginning: perturbation of the microbiota by antibiotics in early life and its role in health and disease

    Nat Immunol

    (2014)
  • C.R. Cardwell et al.

    Caesarean section is associated with an increased risk of childhood-onset type 1 diabetes mellitus: a meta-analysis of observational studies

    Diabetologia

    (2008)
  • J. Phillips et al.

    History of cesarean section associated with childhood onset of T1DM in Newfoundland and Labrador, Canada

    J Environ Public Health

    (2012)
  • U. Samuelsson et al.

    Caesarean section per se does not increase the risk of offspring developing type 1 diabetes: a Swedish population-based study

    Diabetologia

    (2015)
  • A.S. Khashan et al.

    Mode of obstetrical delivery and type 1 diabetes: a sibling design study

    Pediatrics

    (2014)
  • N. Pearce et al.

    The global epidemiology of asthma in children

    Int J Tuberc Lung Dis

    (2006)
  • A. Werner et al.

    Caesarean delivery and risk of developing asthma in the offspring

    Acta Paediatr

    (2007)
  • A. Maitra et al.

    Mode of delivery is not associated with asthma or atopy in childhood

    Clin Exp Allergy

    (2004)
  • A. Sevelsted et al.

    Risk of Asthma from Cesarean Delivery Depends on Membrane Rupture

    J Pediatr

    (2016)
  • S. Thavagnanam et al.

    A meta-analysis of the association between Caesarean section and childhood asthma

    Clin Exp Allergy

    (2008)
  • A. Suau et al.

    Direct analysis of genes encoding 16S rRNA from complex communities reveals many novel molecular species within the human gut

    Appl Environ Microbiol

    (1999)
  • J.G. Caporaso et al.

    QIIME allows analysis of high-throughput community sequencing data

    Nat Methods

    (2010)
  • J.G. Caporaso et al.

    Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample

    Proc Natl Acad Sci U S A

    (2011)
  • J.G. Caporaso et al.

    Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms

    ISME J

    (2012)
  • P.J. Planet

    Life After USA 300: The Rise and Fall of a Superbug

    J Infect Dis

    (2017)
  • Cited by (130)

    • Microbiome and trauma

      2023, Current Therapy of Trauma and Surgical Critical Care
    • Effects of antimicrobials on the gastrointestinal microbiota of dogs and cats

      2023, Veterinary Journal
      Citation Excerpt :

      Profound GI microbiota shifts as a result of AT have also been reported in adults. Beta-lactams and macrolides, including azithromycin and clarithromycin, are reported to decrease Firmicutes and Actinobacteria and increase Bacteroidetes and Proteobacteria (Ianiro et al., 2016; Iizumi et al., 2017). Fluoroquinolones decrease Firmicutes and Actinobacteria and increase Bacteroidetes (Iizumi et al., 2017).

    • Antibiotics in urine from general adults in Shenzhen, China: Demographic-related difference in exposure levels

      2022, Science of the Total Environment
      Citation Excerpt :

      In China, 51.9 % of antibiotic prescriptions during 2014–2018 were inappropriate (Zhao et al., 2021). Excessive use of antibiotics may cause the development of bacterial resistance, disturbance of gut microbiota, allergies, liver disease, and even superinfection (Ferri et al., 2017; Ianiro et al., 2016; Iizumi et al., 2017). These issues are of particular concern because antibiotics have been detected in freshwater (Kovalakova et al., 2020), soil (Cerqueira et al., 2020), dust (Ding et al., 2020), and other media.

    View all citing articles on Scopus
    View full text