The microbiome’s role in pediatric intestinal failure and associated liver disease

Gray's anatomy schematic of the liver.

On Wednesday we talked about children who suffer from short bowel syndrome (SBS), specifically highlighting the role of the gut microbiome alongside its relationship to parenteral nutrition (PN).  In addition to demonstrating microbial dysbiosis in children with SBS, intestinal dysbiosis was noticed in patients receiving PN, while those who had been weaned off showed bacterial overgrowth. 

SBS can also lead to intestinal failure (IF) and further complications.  Among the many, PN and disrupted bowel function have both been shown to lead to IF-associated liver disease (IFALD), which can result in severe illness and even death.  Steatosis, a pathological indication used to describe abnormal lipid retention in cells, has been observed in liver histological samples. 

The cause of IFALD remains unclear, but findings from studying other liver disorders suggest involvement of the gut microbiome.  Intestinal overgrowth has been postulated for quite some time now, but evidence is lacking and the exact biological underpinnings that lead to liver injury remain unclear.  Researchers in Finland sought to address this, estimating that IF-induced disruptions to the gut microbiome of pediatric patients played a direct role in causing this liver damage. 

Twenty-three pediatric patients developing IF were selected for this study.  Researchers collected fecal samples to analyze microbiota populations and took liver biopsies to examine inflammatory damage and fibrotic tissue morphology.  In line with previous findings, the microbiomes of patients with IF had limited bacterial diversity and species richness as compared to those of healthy children and adults. 

A strong correlation was observed between microbiota composition and liver steatosis, and different microbiota strains were shown to be associated with different stages of the disease progression.  Additionally, an overabundance of microbiota in the Proteobacteria phylum was observed in patients undergoing PN.  The Proteobacteria phylum contains many opportunistic pathogenic bacteria, including E. coli.  Interestingly, Proteobacteria species produce lipopolysaccharides, which are known toxins to the liver.  The researchers were also able to model that bacterial composition was a strong predictor of liver steatosis than nutrition history or bowel length post-resection surgery. 

These findings led the researchers to propose that intestinal resection, alongside PN, disrupts the intestines, and consequently native microbiota populations.  The disruption and species decline invites opportunistic bacteria, such as those in the Proteobacteria phylum, to populate the intestine.  These bacteria release of lipopolysaccharides into the blood stream, and upon reaching the liver, induce inflammatory toxic damage leading to steatosis. 

This study complements Wednesday’s discussion and helps us makes better sense of a convoluted disease complication that has drastic consequences.  Understanding the microbiome’s influence on post-SBS liver disease can help clinicians make informed decisions to rescue pediatric patients from these ailments.  

Multiple studies show connection between early life antibiotics and obesity

Childhood obesity is a growing problem not just in the United States, but around the globe. While diet and exercise are basic ways to reduce weight, there can be many complex causes of weight gain. Stress, caloric intake and genetics have all been seen to have a correlation with weight gain. Recent studies, however, have suggested a possible microbiome connection, through a correlation between obesity and the antibiotic exposure. A recent review published in Advances in Obesity, weight management & control described various studies on this subject.

 The researchers discussed five studies that were related to childhood obesity and antibiotic intake. In a Canadian study of 616 children participating in a Study of Asthma, Genes and the Environment survey, a significant relationship was found, particularly for boys, between obesity at ages 9 and 12 and first antibiotic exposure at 3-12 months. In the UK, the Avon Longitudinal Study of Parents and Children found that antibiotic prescription before 6 months of age led to higher Body Mass Index during 10 to 38 months, while later antibiotic intake did not lead to this association. In a study done by Bailey with 65,480 children, 69% of children had infant or preadolescent obesity if they had taken antibiotics during the first 23 months of their life.

 In a fourth study discussed by the article, a large cross sectional study ISSAC was performed in New Zealand on 74,946 male children between the ages of 5 and 8. The study found that boys who had taken antibiotics during their first year of life had higher BMIs later in life. Finally, in a Danish study by Ajslev, 28,364 children were administered antibiotics before 6 months of age and were followed up for the next 7 years. A significant association between antibiotics and obesity was found only for boys.

 Based on these studies, it seems apparent that altering the microbiome through antibiotic use early in life is associated with an increase in BMI.  As many people have recognized in recent years, farmers give antibiotics to their livestock to help them gain weight, and a similar phenomenon may be occurring in humans.  Scientists like the AMI’s advisory board member Martin Blaser have taken up this cause, and he has lectured about this connection many times in recent years.  We know that the microbiome impacts weight gain through diet and nutrition, so it should come as no surprise that decreasing the abundance of many bacterial species through antibiotics may cause obesity.

The microbiome of children with short bowel syndrome

Infants who suffer from necrotizing enterocolitis, a condition we’ve written about on the blog, often must have the dead portion of their intestines surgically removed. Some children who undergo this procedure, as well as others with congenital malformations of the bowel, suffer from a condition called short bowel syndrome (SBS). SBS results when nutrients are unable to properly absorb in the intestines and intravenous feeding is often needed to ensure these children have the nutrients needed to survive. It is thought that the intestinal microbiome plays an important role in the ability to remove children from intravenous feeding, however the microbiome of these children had never previously been mapped.

Scientists in Sweden successfully mapped the microbiome of children with SBS that were diagnosed in the neonatal period. They collected fecal samples from 11 children with SBS and 7 healthy siblings who served as controls for the study. Children that were on parenteral nutrition (PN), or intravenous feeding, had significant intestinal dysbiosis compared to the children who had been weaned off of PN and suffered from a condition called small bowel bacterial overgrowth (SBBO), a condition that is known to prevent weaning off of PN.

In 6 of the 11 patients with SBS, and specifically those still on PN, Enterobacteriacae dominated the guts of these children. While those children who were off PN had more diverse microbiomes than those on PN, only one of those 5 children had diversity levels on the same level of the control individuals without SBS.

This microbial dysbiosis in children with SBS is in line with mappings of individuals with other bowel conditions such as Crohn’s disease and necrotizing enterocolitis. While this study was small in the number of children studied, it was the first to study the microbiome of children with this serious condition. Currently, children with SBS are often given probiotics but the results of this have been conflicting and there are often complications such as septicemia. Future studies will be important to look at how the microbiome can be altered to treat this dysbiosis so that children can be weaned off of parenteral nutrition.

The nasal microbiome of infants may impact risk of developing asthma

Many lower respiratory illnesses have been shown to associate with specific lung, throat and nasal bacteria, but the role of the microbiome is still unclear, and mechanisms for the connection have yet to be proven.  Of particular interest is asthma, which affects around 7% of people in the US, and increases a person’s risk for many other conditions.  While it is normally diagnosed in toddlers, scientists believe that the groundwork for the disease is actually laid during infancy.  With that in mind, researchers in Australia performed the first longitudinal study of infants’ nasopharyngeal (nose and throat) during the first year of their lives, and tracked episodes of respiratory illness during that time.  They discovered a strong connection between the microbiome and respiratory illness, including asthma, and last month published their results in Cell Host and Microbe.

The researchers collected nasopharyngeal microbiome samples from 234 infants at different time points during their first year of life.  Most infants’ microbiomes were dominated by the following species: Moraxella catarrhalis, Streptococcus pneumoniae, Staphylococcus aureus, Haemophilus influenzae, and Alloiococcus otitidis.  Interestingly, this was true for infants regardless of birth delivery mode (i.e. cesarean or vaginal) as well as length of breast feeding.  In contrast, having a furry animal in the house tended to increase the abundance of Streptococcus, but having older siblings tended to decrease it.  In addition, there were strong seasonal effects on the microbiome, with Haemophilus being associated with the summer, and Moraxella the winter.  In children with respiratory illness, Haemophilus, Moraxella, and Streptococcus were most frequently measured, and Staphylococcus, Alloiococcus, and Corynebacterium least frequently measured.

When the scientists compared their results with the asthma outcomes of the children at 5 years old they noticed one significant trend.  Colonization by Streptococcus at around 2 months old, which was asymptomatic at the time and occurred in 14% of infants tested, was strongly connected to chronic wheezing (itself an indication of asthma) at 5 years old.  They suggest that the developing airways in infants may be especially vulnerable to Streptococcus.

This long term study does a really nice job of defining how the microbiome grows and develops in the airways of infants – something which previously hadn’t been performed at such a large scale.  While this study alone does not figure out exactly what the microbiome’s role is in childhood respiratory illnesses, it does provide a baseline for future studies to work off of.   

Commensal bacteria likely prevent wound healing after certain surgeries

Gastrointestinal bobbins used to connect pieces of the colon and create anastomoses

Often times when people undergo colon surgeries the colon must be re-adjoined, creating what is called an anastomosis.  Around 50% of the time this anastomosis will not heal properly, and bacteria can escape from the gut into the body.  Even with improvements to surgical techniques the rates for anastomotic leak have not really improved over the years.  In fact, due to the likelihood of a patient developing this sometimes-deadly ailment, many surgeons will not even attempt to re-adjoin a colon, and instead just create a stoma (i.e. divert the colon outside the body).  It has been known that bacteria somehow cause anastomotic leaks, so intravenous antibiotics are given as standard care, but they have not been highly effective because the mechanism behind the bacterial involvement is still unknown.  However, a big step forward in elucidating the microbiome’s contribution to this disease was made by a group out of the University of Chicago.  They recently discovered that common gut bacteria, Enterococcus faecalis and Proteus mirabilis, actually hinder the bodies natural wound repair process.  They published their results in the Science Translational Medicine last week.

 First, the scientists performed a colon resection on mice and then tracked which ones developed anastomotic leaks and which ones healed normally.  Eventually they sacrificed the animals and made a number of measurements on the resectioned colons, which allowed them to compare the results between the groups.  The first thing they discovered was that enzymes responsible for degrading collagen were much higher in mice that had the leaky anastomoses. (Collagen a molecule produced by the body that is important to reattaching the sections.)  Next, they cultured organisms that were taken from the sites of the anastomoses and measured their ability to degrade collagen.  They found that two members of the communities, Enterococcus faecalis, and Proteus mirabilis, had the highest collagen degradation activity.  Interestingly, only some strains of E. faecalis had high collagen degradation activity, and it was these exact strains that were found in the leaky anastomoses, whereas the healthy anastomoses contained the other, non-collagen degrading, E. faecalis.  Moreover, when the researchers transplanted these virulent strains of E. faecalis into the guts of mice that just underwent colon sectioning these mice developed leaky anastomoses regardless of whether or not they were given antibiotics intravenously.  However, when antibiotics were applied directly to the anastomoses and E. faecalis was not able to grow, leakiness did not develop.  In a final experiment the researchers cultured bacteria from humans that had just undergone colon surgery and discovered that their anastomoses contained E. faecalis and other collagen degrading bacteria (although only one developed anostomatic leak).

 The researchers actually went a step further and pinpointed the actual genes that were associated with degrading the collagen, and I leave it to our most interested viewers to read the paper and find out more.  Thanks to research like this, and better, more targeted antibiotics, doctors are a step closer to curing leaky anastomoses.

Our immune system selectively chooses which gut bacteria to keep, and which to eliminate

We have come a long way in our understanding how exactly pathogenic bacteria can invade and populate the gut.  Yet, there still remains uncertainty as to how exactly our immune system responds to and eliminates these infectious bacteria.  A recent study addressed this by investigating the immune response to pathogenic bacteria in mice guts. 

Some Escherichia coli can be pathogenic and infect the human gastrointestinal tract.  In these instances, these Gram-negative bacteria attach to and populate the gut and cause lesions to the epithelium through a well-characterized attaching-and-effacing behavior.   It is currently understood that IgG antibodies are produced in response to E. coli infection, but the exact cellular underpinnings as to how the bacteria are eliminated are unknown. 

To model this, researchers infected germ free mice with Citrobacter rodentium, a bacterial strain known to carry genes that exhibits effacement pathology in mice.  The specific genes of interest that induce enterocyte effacement (LEE) are referred to as a pathogenicity island, loci responsible for virulent behavior, and they are present in both E. Coli and C. rodentium.   The researchers measured adaptive immunity reaction in response to C. rodentium infection, and specifically looked to see if LEE - the virulent bacterial signature - was down-regulated. 

It was found that the LEE virulent strain was down-regulated concomitant to an increase in release of IgG antibodies.  These IgG antibodies were found to be specific to the LEE virulent expression, as supported by significant IgG binding affinity to the virulent strain.  The IgG antibodies eliminated the specific C. rodentium phenotype that expressed the LEE loci, and upon binding to the bacteria, they were removed by neutrophils.

Interestingly, the C. rodentium avirulent phenotype that lacked the LEE was not eliminated by IgG antibodies.  However, these bacteria were subsequently outcompeted by other microbiota populations.  Together, this information suggests that IgG could selectively eliminate the C. rodentium virulent phenotype, and innate immunity could eventually remove the non-virulent populations. 

This study provides excellent insight into how our immune system can distinguish between good and bad bacteria in addition to describing the underlying cellular mechanism.  Defining the molecular underpinnings of antibody action will allow us to make significant advancements in therapeutic approach.  Understanding the molecular pathways is a critical first step toward pharmacotherapeutic intervention, and this study could potentially lead to the development of some exciting advancements in the future.