The dark side of antibiotics

Martin J Blaser

Departments of Medicine and MicrobiologyNew York University School of Medicine

Department of Biology, NYU

Reflux Disease (GERD)

White men

Nonwhite men

400

350

300

250

200

150

100

50

0

1970

-74

1975

-79

1980

-84

1985

-89

1990

-95

Diseasesincreasing in

recent decades

Juvenile (type 1) diabetes

Asthma

Sources:Ann NY Acad Sci

2008 12:1150N Engl J Med

2006;355:2226Gut 1997;41:594

Obesity trends in US adults: changing physiology

No Data

Percent of adults who are obese, in 10 countries

Perc

ent (

%)

NHANES I1971-1974

NHANES II1976-1980

NHANES III1988-1994

NHANES 2003-2006

Obesity trends among U.S. children and adolescents

Sex- and age-specific BMI > 95th %ile, based on CDC growth charts

Chart1

546.1

56.55

7.211.310.5

12.41717.6

2-5y

6-11y

12-19y

Sheet1

2-5y6-11y12-19y

546.1

56.55

7.211.310.5

12.41717.6

To resize chart data range, drag lower right corner of range.

0

2

4

6

8

10

12

14

16

1990 1995 2000 2005 2010 2015 2020

Comparison of time trends of overweight & obese children globally

% Overweight& Obese

Developed

Developing

Prevalence: (>2SD above weight-for-height median) in children 0-5 years old

M de Onis et al. Am J Clin Nutrition 2010; 92; 1257-64.

0

10

20

30

40

50

60

70

80

1990 1995 2000 2005 2010 2015 2020

Global number of overweight & obese children 2SD from weight-for-height median) in children 0-5 years old

M de Onis et al. Am J Clin Nutrition 2010; 92; 1257-64.

Global

Developing

Developed

Number ofoverweightand obesechildren

(millions)

GG Kaplan, SC Ng. Gastroenterology 2017; 152: 313-21.

Worldwide trends in IBD incidence over 150 years

Host Fecal Microbiota

H. Ochman et al. PLoS Biology 2010Evolutionary relationships of wild hominids

Concept 1: Evolution

10

Schematic of interactions between a co-evolved colonizing microbe and host

Signals Signals•Robust•Resilient

Nature 2007; 449:843-9

Concept 2: Equilibrium

11

What happens to the physiology of the host when the co-evolved microbe is lost (becomes extinct)?

•Robust•Resilient

Loss?

EMBO Reports 2006; Nature Rev Microbiol 2009

When does the adult gut microbiome become established?

T Yatsunenko et al. Nature 2012;486, 222-7

Concept 3: Age Window

Mother Child Transfer of Microbes (Ancient)

Nature Reviews Genetics 2012;13:260-70

Oral

Mammary

Skin

Vaginal

Mother Child Transfer of Microbes (Modern)

Nature Reviews Genetics 2012;13:260-70

Bottle feeding

Caesarean sectionExtensivebathing

Antibiotics

Maternalexposures

• Antiseptics• Antibiotics• Dietary

anti-bacterials

Oral

Mammary

Skin

Vaginal

Changed human ecology has altered transmission and maintenance of ancestral microbes, which affects the composition of the microbiota.Especially important are microbes usually acquired early in life, since they affect a developmentally critical stage.

Lancet 1997; Gut 1998; Perspect Biol Med 2002; Scientific American 2005; EMBO Reports 2006; Nature Rev Microbiol 2009; Nature 2011; Nature Rev Immunol 2017

Theory of Disappearing Microbiota

Skin

Vaginal

Mammary

Oral

Maternalexposures

• Antiseptics• Antibiotics• Dietary

anti-bacterials

Bottle feeding

100%

0 20 40 60 80Time (years)

Rep

rese

ntat

ion

of a

co

nser

ved

mic

robi

ota

Vertical acquisition

Potentialhorizontal acquisition

The effect of maternal status on the resident microbiota

of the next generation

Nature Rev Microbiol 2009;7:887

Adapted from Y. Urita et al.J Ped Child Health 2013; 49:394-8

Prevalence of Helicobacter pylori

in Japanese families

%decline in

prevalence

-

37

7112.5

43.4

68.7

1940 1970 2000

% positive

Grandmothers (244/355)

Mothers (251/578)

Children (101/808)

Approximate year of birth

100%

0 20 40 60 80Time (years)

Rep

rese

ntat

ion

of a

co

nser

ved

mic

robi

ota

Antibiotics: one of the greatest discoveries of the 20th century

• Saved lives• Revolutionized medicine

St. Mary’s Hospital, Paddington

• >73 billion antibiotic doses worldwide yearly• USA (2010): 258 million courses (833/1000)• Children: 2.7 courses by 2 years; 10.9 by 10 years• Pregnancy: >50% treated or given prophylaxis

+ Exposures from use of antibiotics on the farm (scale unknown)

Variation in antibiotic use

Per capita antibiotic use in 31 countries in Europe

o

Country

Median number of

illnessepisodes

% of episodes

treated withantibiotics

Prescription rate per

infant peryear

% of infantsreceiving

≥ 1 courseof antibiotics

Italy 3 18.4 1.3 55Netherlands 3 9.8 0.6 37Austria 4 6.7 0.5 33Switzerland 4 3.9 0.2 18Germany 4 5.1 0.5 33

Variation in antibiotic prescribing in the first year in life in 839 children in five European countries

J Stam et al. Acta Pediatr 2012; 101:929-34.

ET Rogawski et al. Bull WHO 2017; 95: 49-61.

Incidence of antibiotic use in the first 2 years of life by study site among 2134 children in the MAL-ED birth cohort, 2009-14

Excluding: preventive visits, CCC (Chronic Complex Conditions)Standardized by: age, sex, age-sex, race, Medicaid status

Rates of antibiotic prescribing for sick children

JS Gerber et al. J Pediatric Infect Dis 2015

Excluding: preventive visits, CCC, antibiotic allergy, prior antibioticsStandardized by: age, sex, age-sex, race, Medicaid status

Rates of prescribing broad-spectrum antibiotics

Trends in overall antibiotic prescribing for acute respiratory infections among USA Veterans, 2005-2012

BE Jones et al. Ann Intern Med 2015

Left. Variation among providers. The histogram shows the distribution of observed proportions of visits with an antibiotic prescription across 2594 providers with at least 100 ARI visits each (n= 480 875). The curve depicts the modeled distribution of antibiotic Rx’s across providers, after controls were set for the measured patient, provider, and setting.

Right. Sources of variation. The lines depict modeled distributions describing variation in proportion of antibiotic prescriptions attributable specifically to VAMCs, clinics, and providers, respectively, after controls were set for the measured patient, provider, and setting variables. The dashed-and-dotted line corresponds to the curve in the left panel and depicts overall modeled variation in antibiotic prescription across providers, including differences between providers at different clinics and VAMCs.

Variation in antibiotic prescribing for acute respiratory infections among USA Veterans, 2005-2012

Outpatient antibiotic usage rates by region, 2010

Northeast830

Midwest868

West638

South936

National rate833/1000 population(258 million courses)

L Hicks et al. N Engl J Med 2013, 368:1461.

Antibiotic prescribing per capita in USA, 2011

LA Hicks et al. Clin Infect Dis 2015;60:1308-16

ALL AGES AGE ≤2

AGE 3–64 AGE ≥65

Sex-specific rates of antibiotic prescribing, 2011

Prescriptions/1000 persons

SexChildren

(Age ≤ 19)Adults

(Age ≥ 20)

Female 941 990

Male 841 596

LA Hicks et al. Clin Infect Dis 2015;60:1308-16

Associations between county-level educational and income characteristics and high antibiotic prescribinga

Characteristic Adjusted Odds Ratio(95% CI)Four-year college (%)

Lowest tertile (

Associations of antibiotic use with long term health consequences

OR

Number of purchases

Association between number of antibiotic purchases from birth to diagnosis of cow’s milk allergy in Finland

J Metsala et al. Epidemiology 2013; 24:303-9.

Adjusted model includes maternal age, smoking, prior deliveries, mode of delivery, and child’s birth weight.

UnadjustedAdjusted

OR

Before pregnancy During pregnancy

Number of purchases Number of purchases

Association between number of maternal antibiotic uses and risk of cow’s milk allergy in the offspring

J Metsala et al. Epidemiology 2013; 24:303-9.

KH Mikkelson et al. J Clin Endocrinol Metab 2015

Adjusted Odds Ratio (95% CI)a

Type of Antibiotics 2-4 ≥ 5Any antibiotic 1.21 (1.19-1.23) 1.53 (1.50-1.55)Narrow-spectrum 1.22 (1.20-1.23) 1.55 (1.53-1.57)Broad-spectrum 1.18 (1.16-1.20) 1.31 (1.29-1.34)Bactericidal 1.18 (1.17-1.20) 1.48 (1.46-1.50)Bacteriostatic 1.20 (1.19-1.22) 1.39 (1.36-1.41)

Adjusted ORs for type 2 diabetes according to antibiotic exposure before initiation of diabetes treatment

aOR for type 2 diabetes with redemption of 2-4 or ≥ 5 antibiotic prescriptions,compared with 0-1 redemptions .

KH Mikkelson et al. J Clin Endocrinol Metab 2015

Ratio of antibiotic use in cases versus controls in the 15 years before initiation of treatment for T2D

Results are only for cases with an index between 2010 and 2012.

KH Mikkelson et al. J Clin Endocrinol Metab 2015

Outpatient antibiotic usage rates by region, 2010

Northeast830

Midwest868

West638

South936

National rate833/1000 population(258 million courses)

L Hicks et al. N Engl J Med 2013, 368:1461.

Comparisons between the geography of obesity and antibiotic use, 2010

Antibiotic prescriptions per 1000 persons, 2010

L Segal & MJ Blaser, Ann Am Thor Soc 2014 Observational data

2011—2016: 16 studies published – 14/16 show parallel findings.

Exposed to antibiotics during the first 6 months of life

Is early life antibiotic exposure associated with increased weight gain?

L Trasande et al. Int J Obesity 2013; 37: 16-23.

*

*

*

* Significant difference in Z-score; p

Antibiotics used in farm animals to promote their growth

Antibiotic Class Target

Bambermycin Glycolipid Cell wall

Virginiamycin Streptogrammin Protein synthesis

Avilamycin Orthosomycin Protein synthesis

Bacitracin Cyclic peptide Cell wall synthesis

Monensin Ionophore Cell membrane

Carbadox Quinoxaline DNA Synthesis

Using mice to examine the effects of antibiotics

Antibiotic Class Target

Bambermycin Glycolipid Cell wall

Virginiamycin Streptogrammin Protein synthesis

Avilamycin Orthosomycin Protein synthesis

Bacitracin Cyclic peptide Cell wall synthesis

Monensin Ionophore Cell membrane

Carbadox Quinoxaline DNA Synthesis

Antibiotic Control

Measure properties

Analyze microbiome

Find relationships

44

Body fat in antibiotic-exposed and control 10-week old mice

% b

ody

fat

*p

Effects of combining high fat diet and antibiotics

Control

STAT

Control Normal ChowControl Normal Chow High fat dietSTAT Normal ChowSTAT Normal Chow High fat diet

4 17 30 weeks0

Laurie CoxCell 2014;158:705-21

STAT = low-dose antibiotic exposure,as used on the farm

HFD and antibiotic both contribute to body fat

Control

STAT

Control Normal ChowControl Normal Chow High fat dietSTAT Normal ChowSTAT Normal Chow High fat diet

Amount of body fat

Nursing0-4 weeks 4-6w

High Fat Diet6-28 weeks

DuraSTAT: Are changes durable with limited antibiotic exposure?

4 weeks

No antibioticsControl

4-STAT

8 weeks8-STAT

28 weeks sub-therapeutic antibiotic treatment (STAT)28-STAT

No antibiotics

No antibiotics

Diet:Chow

Laurie CoxCell 2014;158:705-21

Nursing0-4 weeks 4-6w

High Fat Diet6-28 weeks

Morphometric changes with limited antibiotic exposure

4 weeks

No antibioticsControl

4-STAT

8 weeks8-STAT

28 weeks sub-therapeutic antibiotic treatment (STAT)28-STAT

No antibiotics

No antibiotics

* P < 0.05, t-testX2: sacrificed 4 control and 4 8-STATX3: sacrificed 3 control and 3 28-STAT

Diet:Chow

Female C57

28-STAT n = 8

Control STAT

IL-17A

IL-2

2

Control STAT

IL-17A

IL-2

2Small intestine Large intestine

* *

* p

Fecal community structure at 3 weeks

Cont

rol

4-ST

AT8-

STAT

28-S

TAT

Die

t

Milk

NC

HFD

Wee

k

0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

3

Control STAT

Fecal community structure at 8 weeks

Cont

rol

4-ST

AT8-

STAT

28-S

TAT

Die

t

Milk

NC

HFD

Wee

k

0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

Control STAT Post 4-STAT

8

Is microbe-induced obesity transferable?

No abx

Antibiotics

Donors Germ-Free recipientsMicrobiota

Body composition - Days post-transfer

Control

No abx

5-weeks

Expression of genes involved in intestinal defenses in the microbiota of donor and recipient mice

p-values, by t-test

Th17 Antimicrobial peptides

Effects of a single antibiotic pulse on microbial populations

Tylosin

Victoria Ruiz et al.Nature Communications 2017

Effects of a single antibiotic pulse on host gene expression

How does the antibiotic pulse affect ileal gene expressionin the mother and in her pups?

Pups(1287) Dams

(106)

1238 49 57

Ileal gene expression in PAT or control pups and dams

RNAseqanalysis

A Schulfer et al.Nature Microbiology

Nov 2017

Can microbiota transfer phenotypes across generations? A study of IBD

Microbiota communities until pup sacrifice at week 21

Unweighted UniFrac

PC1 (16%)

PC2 (9%)

PC3 (8%)

Transmission of taxa across generationsWildtype IL10-/-

Serina Robinson Dan KnightsTonya Ward

Colonic pathology in IL-10-/- pups at week 21, according tothe microbiota to which their mothers were exposed

Control Antibiotic-perturbed (STAT)

SummaryNeither pups nor mother received antibioticsEnhanced disease signal is entirely microbial

Antibiotic effect crosses generationsInheritance also based on microbial genes

HistologyActivity

Index (HAI)

Antibiotic impact on long-term physiology through microbiota changes

PLoS Pathogens 2015

Solutions

Antimicrobial use in USA farm animals, 2013

Source food Antimicrobial agents in surveys

Concentration Country

Shrimp Fluoroquinolones 0.1-1.0 ng/g USA

Salmon, trout, shrimp tissues Fluoroquinolones 0.28-16 ng/g Canada

Swine, chicken, shrimp tissues Fluoroquinolones 1-100 ng/g China

Bob veal, heavy calves, sows, heifers, market hogs, non-formula-fed veal, roaster pig

Sulfonamides 0.1-1 ppm USA

Bull meat Moxidectin (milbemycin) 89.13 ppb USA

Goat meat Oxytetracycline 4.66 ppm USA

Market hog, roaster pig meat Carbadox 47-110 ppb USA

Catfish, basa Fluoroquinolones 1.9-6.5 ppb ChinaHoney Erythromycin 50-1776 ng/g Turkey

Corn, green onion, cabbage Chlortetracycline 2-17 ng/g USA

Antibiotic residues in food

All 18 detected58.3% of children

were positive

H Wang et al. Environ Sci Technol 2015; 49: 5070-9

Age group

Sex

Detection of antibiotics in urine by use category

Category N

A. Human 4B. Veterinary 3C. Both 11

A B C

A B C

Number of courses taken

Age (years)

USADuring period Cumulative

2 2.73 2.733 - -

10 8.17 10.9020 6.78 17.6840 13.38 31.0665 19.93 50.98

Adapted from L. Hicks NEJM 2013; 368:1461 (USA)

Cumulative antibiotic use in the USA, by age

Cumulative antibiotic use in the USA and Sweden

Number of courses taken

Age (years)

USA SwedenDuring period Cumulative

During period Cumulative

2 2.73 2.73 - -3 - - 1.39 1.39

10 8.17 10.90 2.90 4.2820 6.78 17.68 2.52 6.8040 13.38 31.06 5.92 12.7265 19.93 50.98 8.48 21.20

Adapted from L. Hicks NEJM 2013; 368: 1461 (USA)A. Ternhag NEJM 2013; 369: 1175 (Sweden)

Diversity loss in the microbiome in 3 model locales

Science 2016

Next steps for the microbiome?

Science 2016

Medicine of the future: new analyses of child health

Analysis

Globalmicrobes

Personalmicrobes

Hostmarkers

• Genotype• Biomarkers

New approach to optimize child health?

Analysis

Globalmicrobes

Personalmicrobes

Administer

Hostmarkers

• Genotype• Biomarkers

New algorithm for preventing illnesses?

Analysis

Globalmicrobes

Personalmicrobes

Administer

Assess forpositivity,phenotypes

Hostmarkers

• Genotype• Biomarkers

Drug the microbiome

The dark side of antibioticsSlide Number 2 Obesity trends in US adults: changing physiology �Slide Number 4Slide Number 5Slide Number 6Slide Number 7Slide Number 8Slide Number 9Slide Number 10Slide Number 11Slide Number 12Slide Number 13Slide Number 14Slide Number 15Slide Number 16Slide Number 17Slide Number 18Slide Number 19Slide Number 20Slide Number 21Slide Number 22Slide Number 23Slide Number 24Slide Number 25Slide Number 26Slide Number 27Slide Number 28Slide Number 29Slide Number 30Slide Number 31Slide Number 32Slide Number 33Slide Number 34Slide Number 35Slide Number 36Slide Number 37Slide Number 38Slide Number 39Slide Number 40Slide Number 41Antibiotics used in farm animals to promote their growthUsing mice to examine the effects of antibioticsSlide Number 44Effects of combining high fat diet and antibioticsHFD and antibiotic both contribute to body fatDuraSTAT: Are changes durable with limited antibiotic exposure?Morphometric changes with limited antibiotic exposureEffects of STAT on intestinal Th17 populations Fecal community structure at 3 weeks Fecal community structure at 8 weeksIs microbe-induced obesity transferable?Expression of genes involved in intestinal defenses �in the microbiota of donor and recipient miceSlide Number 54Slide Number 55Ileal gene expression in PAT or control pups and damsSlide Number 57Microbiota communities until pup sacrifice at week 21Slide Number 59Slide Number 60Slide Number 61Slide Number 62Slide Number 63Slide Number 64Slide Number 65Slide Number 66Slide Number 67Slide Number 68Slide Number 69Slide Number 70Slide Number 71Slide Number 72Slide Number 73Blank Page