Importance of the Gut Microbiome in Preterm Infants

Abstract

Birth represents the start of an incredible journey for the individual and the microbes which reside within and upon them. This interaction between human and microbe is essential for healthy development. Term infants are colonized by bacteria at birth, and thereafter the diet is the most important factor shaping the gut microbiome, in particular receipt of human milk. Human milk contains viable bacteria and numerous components that modulate the bacterial community, including human milk oligosaccharides (HMOs) which promote the growth of Bifidobacterium species. Notably, Bifidobacterium spp. are the primary bacterium used in probiotic supplements, owing to their association with positive outcomes in cohort studies and range of beneficial properties in mechanistic experiments. Preterm infants born <32 weeks' gestation encounter an unnatural beginning to life, with housing in “sterile” incubators, higher rates of caesarean delivery and antibiotic use, and complex nutritional provision. This reduces Bifidobacterium abundance and overall microbial diversity. However, this also presents an opportunity to use probiotics and prebiotics (e.g., HMOs) to restore “normal” development. Much work has focused in this area over the past two decades and, while more work is needed, there is promise in symbiotic intervention to modulate the microbiome and reduce disease in preterm infants.

Development of the Gut Microbiome in Term and Preterm Infants

Following birth, neonates are rapidly colonized by microbes that play fundamental roles in health and disease. To a bacterial cell, the human body is like a universe, with different planets (i.e., body sites) containing distinct atmospheres (i.e., temperature, oxygen levels, acidity, humidity, etc.) and resources (i.e., food sources and metals). This predisposes the body to colonization by specific microorganisms that are able to survive in that environment and use the available resources to replicate. The gut contains the largest density of microorganisms, termed the gut microbiome, which have important roles in protection from pathogens, immune system training, the breakdown of dietary compounds, and many other functions [1]. Over the first year of life, the infant gut microbiome is highly dynamic, providing a window of opportunity in which to seed a potentially beneficial microbiome and reduce the risk of early- and later-life diseases. In term infants, birth mode and breastfeeding are the most important variables for shaping the early life microbiome [2]. Infants born vaginally have a higher prevalence and relative abundance of Bacteroides spp., which an enteric bacterium and is most likely transferred via the fecal-oral route during delivery [3]. Receipt of human milk is the primary factor shaping the infant gut microbiome over the first year of life, increasing the relative abundance of Bifidobacterium spp. and reducing the relative abundance of pathobionts [2, 4]. Other factors reported to influence infant gut microbiome development include geographical location and having furry pets in the household [2]. However, such influences are only observed up to ~1.5 years of life, after which the gut microbiome becomes increasingly individualized and stable. This early life host-microbiome cross talk and immune development is hypothesized to have important roles in long-term health and is directly correlated to an increased risk of obesity, allergy, asthma, and other disorders later in life [5].

Preterm infants born <32 weeks of gestation encounter an unnatural beginning to life, with housing in “sterile” incubators, higher rates of caesarean delivery and antibiotic use, inability to feed at the breast, complex nutritional supplementation, and parenteral nutrition that bypasses the gastrointestinal tract. Such factors are necessary to maximize survival in this vulnerable population, but as a consequence, the gut microbiome and microbial-host cross talk is abnormal when compared to term infants [6]. In comparison to term infants, preterm infants have lower microbiome diversity and enrichment of pathobionts including Staphylococcus, Klebsiella, Escherichia, Enterobacter, and Enterococcus, and a reduction in potentially beneficial bacteria including Lactobacillus and Bifidobacterium [7].

Microbially Mediated Disease in Preterm Infants

Owing to an altered microbiome, coupled with an immature intestinal architecture (e.g., leaky gut) and an underdeveloped immune system, preterm infants are predisposed to intestinal diseases such as necrotizing enterocolitis (NEC). NEC is the leading cause of death in preterm infants surviving the initial days of life, where exaggerated inflammation cascades to epithelial damage, ischemia, and ultimately necrosis of the bowel. Around 50% of cases will require surgery and of those going to theatre, the survival rate is around 50%. It is likely there are numerous pathways underlying NEC pathogenesis that ultimately manifest in a common endpoint of intestinal necrosis. To this end, work is currently ongoing to characterize and better understand different NEC subtypes. Currently, the most well-characterized is the toll-like receptor (TLR)-4 pathway. Work by Hackam and Sodhi [8], as well as others, have shown that TLR4 expressed specifically on the surface of intestinal epithelial cells (and not immune cells), is activated when the gut is exposed to lipopolysaccharides (LPS) found on the outer membrane of Gram-negative bacteria. TLR4 is crucial for normal fetal development and, as such, is expressed in high levels during this period. Since LPS would not be expected in the gut during this period, activation of TLR4 receptors is not a problem; however, TLR4 expression remains high following premature birth. Once activated, it results in a three-pronged destructive process whereby (1) the nuclear factor kappa B (NF-KB) pathway is activated, leading to increased inflammation, (2) enterocyte destruction, and (3) failure to restore epithelial damage owning to reduced replication and migration of crypt-derived stem cells. This leads to increased translation of enteric bacteria and reduced blood flow to the site of damage, reflecting the common inflamed and necrotic pathology.

Late-onset sepsis (LOS) is another common disease of preterm infants, with greater prevalence but lower overall mortality when compared to NEC. While LOS is not solely a disease of intestinal origin, there is mounting evidence to support the gut as a common route for translocating bacteria. For example, common enteric bacteria of preterm infants, such as Klebsiella and Escherichia, are regularly identified as the causative agent in LOS [9]. There are also reports in preterm infants receiving dietary probiotics where the probiotic strain is detected in diagnostic blood culture [10]. Ward et al. [11] further showed the uro- pathogenic Escherichia coli strains detected in the infant gut could also be isolated in diagnostic blood culture.

Human Milk and Probiotics to Increase Bifidobacterium in Preterm Infants

Once a baby is born prematurely, maternal breast milk is the single most protective factor against NEC [12]. However, infants receiving mother's own milk still develop the disease, suggesting numerous other factors to be involved, including the variable composition of nutrients and other components of human milk. In addition, extremely preterm infants cannot be directly breastfed, necessitating expression, freezing, and nasogastric feeding. Human milk contains viable maternal bacteria that colonize the infant gut [13, 14], but expressed human milk delivered by nasogastric tube is enriched with potential pathogens and reduced levels of beneficial Bifidobacterium [15]. Human milk also contains an abundance of human milk oligosaccharides (HMOs), which are a family of structurally diverse sugars, absent from standard preterm formula milk [16]. HMOs are unbound and cannot be digested by the infant, so they reach the lower gastrointestinal tract intact where they act as growth substrates for specific bacteria, most notably Bifidobacterium [4]. While their primary role is likely to serve as a pre- biotic, there is mounting evidence that they play important roles in epithelial function and immune development. Seminal work by Bode and colleagues [17] has shown a specific HMO, disialyllacto-N-tetraose (DSLNT), was associated with protection against NEC in humans, which was validated in a rat model [18]. This was recently supported by a large observational study of 33 NEC cases, where DSLNT could predict the risk of NEC with a sensitivity and specificity of 0.9 [19]. Separate work has shown that the HMO 2'-fucosyllactose (2'FL) re-duced stimulation of TLR4, suppressing activation of the NF-KB inflammatory pathway and ameliorating LPS-stimulated inflammation [20]. While 2'-FL is the most abundant HMO in human milk, nonsecretor mothers lacking a functional FUT2 allele, who represent ~20% of the population, will lack or produce only trace levels of a1-2 fucosylated oligosaccharides. Notably, no association between 2'FL and NEC risk has been observed in cohort studies, and there is no report of maternal secretor status (i.e., 2'FL presence/absence in milk) as a factor that impacts the likelihood of developing NEC.

Given the importance of the gut microbiome in NEC and the improvements in sequencing technologies that allow bacteria to be surveyed, a large number of studies over the past decade have sequenced stool in preterm infants with NEC and matched controls. Despite improved understanding of preterm infant gut microbiome development, there has been no consistent bacterium or combination of bacteria that reproducibly associate with NEC onset. The most consistent associations are evident at the phylum level, with a higher abundance of proteo- bacteria found in infants who are diagnosed with NEC. On the other hand, there has been evidence that a higher bacterial diversity and higher Bifidobacteriumspp. is associated with protection against NEC [9, 19, 21-25]. Recent work utilizing HMO profiling of mothers' milk and metagenomic sequencing of preterm infant gut microbiome showed the HMO composition in human milk was associated with altered microbiome development. Specifically, infants receiving low concentrations of DLSNT had reduced transition into mature microbiome community types which were dominant in Bifidobacterium spp. [20]. Such studies highlight the importance of diet-microbe-host interaction, but in observational studies utilizing clinical samples it is difficult to determine cause or effect. Possible models for further investigating this interaction at the preterm epithelial surface are proposed in a subsequent section. Animal models have also been used to investigate potential mechanisms, showing bacteria such as Bifidobacterium spp. may increase gut and immune maturation, which could reduce the risk of NEC in preterm infants [22].

The findings of enriched Bifidobacterium in healthy infants have led, in part, to an increased attention and use of probiotics within neonatal intensive care units. Despite this, human preterm probiotic trials have yielded inconsistent results, potentially reflecting the lack of knowledge to underpin probiotic choice, the optimal dosage, the timing and frequency to supplement, what other prebi- otics are needed for probiotic colonization/replication and beneficial functioning, variable colonization rates, and cross-contamination between randomized groups [26]. Nonetheless, recent meta-analyses show that, overall, probiotics significantly reduce NEC, and their use is supported by the European Society for Paediatric Gastroenterology Hepatology and Nutrition Working Group for Probiotics and Prebiotics [27]. However, a recent clinical report from the American Academy of Pediatrics concluded that based on the current evidence and the lack of FDA regulation, they do not support the routine use of probiotics in preterm infants [28]. Such conflicting recommendations highlight the need for further work to optimize the use of probiotics in preterm infants.

Next-Generation Organoid Models for Studying Preterm Intestinal Health and Disease

Better understanding of the interaction between bacteria and infant gut epithelial cells holds incredibly exciting possibilities to better predict, diagnose, and manipulate the microbiome of preterm infants at risk of disease. Animal models have provided important advances; however, the microbiome varies significantly between different animal species, and thus the direct relevance and translation of findings into humans is challenging [29]. The anatomical differences between animal models and neonatal gut, as well as the inability to model extreme gutprematurity (i.e., <32 weeks of gestation), represent further limitations of animal models. To overcome these longstanding hurdles and advance upon associations by investigating underlying mechanisms, several groups have turned to human intestinal-derived organoids (HIOs). HIOs are generated from resected intestinal tissue by stimulating Lgr5+ stem cells isolated from intestinal crypts, and a recent method described a method for establishing lines from preterm infants [30]. HIOs remain morphologically representative of the segments of intestine they were taken from (e.g., ileum will be distinct from colon); are composed of all the major cell types of the intestinal epithelium (i.e., enterocyte, Paneth, goblet, neuroendocrine, and stem); and are physiologically active (i.e., secrete mucus and swell in response to enterotoxins) [29]. Furthermore, they retain the genetic and epigenetic susceptibility and immune programming of the host [31].

Owing to the different oxygen demands of host and microbe, until recently it was not possible to simultaneously co-culture viable HIOs with viable anaerobic enteric bacteria. However, several systems have recently been engineered to recapitulate the steep oxygen gradient across the epithelium. In such systems, the lumen (i.e., apical surface) is anaerobic to support the growth of bacteria, whereas the mucosa (i.e., basolateral surface) is oxygenated to maintain HIO viability. A microfluidic anaerobic intestine-on-a-chip has been developed and shown to sustain cell viability of anaerobic bacteria [32]. A different model called the organoid-anaerobe co-culture system can be setup using nonspecialist, com-mercially available equipment, and additionally allows easy adjustment of the basolateral oxygen concentration [33]. HIO co-culture systems may therefore be a robust and relevant model to systematically explore host responses to bacteria and other stimuli (e.g., milk bioactive components) in the human preterm intestine, allowing comprehensive mechanistic investigation of diet-host-microbiome interaction.

Concluding Remarks

Over the past decade, our understanding of the preterm infant gut microbiome has advanced immensely, but there is still much to learn. One challenge relating to clinical cohorts is the need for multi-site, multi-geographical cohorts, improving the power of observations and the generalizability of findings. Another challenge facing researchers relates to moving beyond the huge number of associations that have resulted from omic profiling of maternal and infant clinical samples, to understand cause or effect, and unravelling the potential mechanisms underpinning these associations. As demonstrated by the consistent observation of a single HMO, DLSNT, being higher in control infants across different studies (and thus cohorts), there is enormous potential for the development of novel biomarkers and therapeutic/preventative strategies. For instance, prioritizing donor milk high in DLSNT for preterm infants and/or supplementing the diet with synthesized DSLNT where required. Such possibilities, as well as other emerging therapies, require deeper understanding and optimization for use in preterm infants before being rolled out. The risk in considering only a single component of a complex system of diet-microbe-host interaction is epitomized with probiotics, where simply providing so-called beneficial bacteria without other components (e.g., prebiotics) to support their growth and beneficial functioning has failed to yield consistent and widespread improvements to outcomes for preterm infants. Nevertheless, personalized or stratified approaches to modulate the gut microbiome though symbiotics is one emerging area that holds genuine potential to reduce the risk of devastating disease in preterm infants.

Conflict of Interest Statement

C.J.S. declares performing consultancy for Astarte Medical and receiving lecture honoraria from Danone Early Life Nutrition and Nestlé Nutrition Institute but has no share options or other conflicts.

References

1    Young VB. The role of the microbiome in human health and disease: an introduction for clinicians. BMJ. 2017;356:j831.
2    Stewart CJ, Ajami NJ, O'Brien JL, et al. Temporal development of the gut microbiome in early childhood from the TEDDY study. Nature. 2018;562:583-8.
3    Korpela K, Helve O, Kolho KL, et al. Maternal fecal microbiota transplantation in cesarean-born infants rapidly restores normal gut microbial development: a proof-of-concept study. Cell. 2020;183:324-34.e5.
4    Vatanen T, Franzosa EA, Schwager R, et al. The human gut microbiome in early-onset type 1 dia-betes from the TEDDY study. Nature. 2018;562:589-94.
5    Davis MK. Breastfeeding and Chronic Disease in Childhood and Adolescence. Pediatr Clin North Am. 2001;48:125-41.
6    Stewart C, Skeath T, Nelson A, et al. Preterm gut microbiota and metabolome following discharge from intensive care. Sci Rep. 2015;5:17141.
7    Masi AC, Stewart CJ. The role of the preterm in-testinal microbiome in sepsis and necrotising enterocolitis. Early Hum Dev. 2019 Nov;138:104854.
8    Hackam DJ, Sodhi CP. Toll-like receptor - mediated intestinal inflammatory imbalance in the pathogenesis of necrotizing enterocolitis. Cell Mol Gastroenterol Hepatol. 2018;6:229-38.e1.
9    Stewart CJ, Embleton ND, Marrs ECL, et al. Lon-gitudinal development of the gut microbiome and metabolome in preterm neonates with late onset sepsis and healthy controls. Microbiome. 2017;5:75.
10    Esaiassen E, Cavanagh P, Hjerde E, et al. Bifido-bacterium longum subspecies infantis bacteremia in 3 extremely preterm infants receiving probiotics. Emerg Infect Dis. 2016;22:1664-6.
11    Ward DV, Scholz M, Zolfo M, et al. Metagenomic sequencing with strain-level resolution implicates uropathogenic E. coli in necrotizing enterocolitis and mortality in preterm infants. Cell Rep. 2016;14:2912-24.
12    Patel AL, Kim JH. Human milk and necrotizing enterocolitis. Semin Pediatr Surg. 2018;27:34-8.
13    Lackey KA, Williams JE, Meehan CL, et al.
What's normal? Microbiomes in human milk and infant feces are related to each other but vary geographically: The INSPIRE Study. Front Nutr. 2019;6:45.
14    Stewart CJ, Marrs ECL, Nelson A, et al. Develop-ment of the preterm gut microbiome in twins at risk of necrotising enterocolitis and sepsis. PLoS One. 2013;8:e73465.
15    Moossavi S, Sepehri S, Robertson B, et al. Com-position and variation of the human milk micro-biota are influenced by maternal and early-life factors. Cell Host Microbe. 2019;25:324-35.e4.
16    Bode L. Human milk oligosaccharides: every baby needs a sugar mama. Glycobiology. 2012;22:1147-62.
17    Autran CA, Kellman BP, Kim JH, et al. Human milk oligosaccharide composition predicts risk of necrotising enterocolitis in preterm infants. Gut. 2018;67:1064-70.
18    Jantscher-Krenn E, Zherebtsov M, Nissan C, et al. The human milk oligosaccharide disialyllacto-N- tetraose prevents necrotising enterocolitis in neo-natal rats. Gut. 2012;61:1417-25.
19    Masi A, Embleton N, Lamb C, et al. Human milk oligosaccharide DSLNT and gut microbiome in preterm infants predicts necrotising enterocolitis. Gut. 2020 Dec 16;gutjnl-2020-322771. doi: 10.1136/gutjnl-2020-322771.
20    He Y, Liu S, Kling DE, et al. The human milk oli-gosaccharide 2'-fucosyllactose modulates CD14 expression in human enterocytes, thereby attenu-ating LPS-induced inflammation. Gut. 2016;65:33-46.
21    Stewart CJ, Embleton ND, Marrs ECL, et al. Tem-poral bacterial and metabolic development of the preterm gut reveals specific signatures in health and disease. Microbiome. 2016;4:67.
22    Fanning S, Hall LJ, Cronin M, et al. Bifidobacte- rial surface-exopolysaccharide facilitates commensal-host interaction through immune modulation and pathogen protection. Proc Natl Acad Sci USA. 2012;109:2108-13.
23    Olm MR, Bhattacharya N, Crits-Christoph A, et al. Necrotizing enterocolitis is preceded by increased gut bacterial replication, Klebsiella, and fimbriae-encoding bacteria. Sci Adv. 2019;5:5727-38.
24    Pammi M, Cope J, Tarr PI, et al. Intestinal dysbiosis in preterm infants preceding necrotizing enterocolitis: a systematic review and meta-anal- ysis. Microbiome. 2017;5:31.
25    Stewart CJ, Fatemizadeh R, Parsons P, et al. Using formalin fixed paraffin embedded tissue to characterize the preterm gut microbiota in necro- tising enterocolitis and spontaneous isolated per-foration using marginal and diseased tissue. BMC Microbiol. 2019;19:52.
26    Patel RM, Underwood MA. Probiotics and necro-tizing enterocolitis. Semin Pediatr Surg. 2018;27:39-46.
27    van den Akker CHP, van Goudoever JB, Shamir R, et al. Probiotics and preterm infants: a position paper by the European Society for Paediatric Gas-troenterology Hepatology and Nutrition Committee on Nutrition and the European Society for Paediatric Gastroenterology Hepatology and Nu-trition Working Group for Probiotics and Prebi- otics. J Pediatr Gastroenterol Nutr. 2020;70:664- 80.
28    Poindexter B. Use of probiotics in preterm infants. Pediatrics. 202U47:e202w51485.
29    Blutt SE, Crawford SE, Ramani S, et al. Engineered Human Gastrointestinal Cultures to Study the Microbiome and Infectious Diseases. Cell Mol Gastroenterol Hepatol. 2017;5:241-51.
30    Stewart CJ, Estes MK, Ramani S. Establishing human intestinal enteroid/organoid lines from preterm infant and adult tissue. Methods Mol Biol. 2121;185-198.
31    Kraiczy J, Nayak KM, Howell KJ, et al. DNA methylation defines regional identity of human intestinal epithelial organoids and undergoes dynamic changes during development. Gut. 2019;68:49-61.
32    Jalili-Firoozinezhad S, Gazzaniga FS, Calamari EL, et al. Complex human gut microbiome cultured in anaerobic human intestine chips. bioRxiv. 2018;421404. doi:10.1101/42L104
33    Fofanova TY, Stewart C, Auchtung JM, et al. A novel human enteroid-anaerobe co-culture system to study microbial-host interaction under physiological hypoxia. bioRxiv. 2019;555755. doi: 10.1101/555755.