FIELD

The present application relates to a composition comprising at least three bacterial genera, a method for determining a risk of developing asthma or other wheezing disorders or allergy in a child, a method for prophylactic or therapeutic treatment of a child in risk of developing said disorders, and a composition comprising at least three bacterial genera for use as a medicament.

BACKGROUND

Asthma, the most common chronic childhood disorder and a major cause of emergency room visits, hospitalizations, as well as daycare and school absence, has increased in incidence dramatically in the recent half century. The beginning of our lives is an important period for susceptibility to environmental exposures from which lasting effects may be imprinted on the developing immune system through complex gene-environment interactions. The human microbiome contains as many as 10 ¹⁴ bacterial cells, roughly on par with our human cells. Most bacteria are in the gastrointestinal tract, where they provide essential stimulation of the child's developing immune system . From birth, humans are continuously exposed to multiple exposures that influence microbiome ecology. The composition of the gut microbiome matures within the first years of life, and the microbiome has the ability to affect host immune maturation; perturbing homeostasis during this important period of development could lead to asthma, allergy, and other immunologic disorders. Thus, the microbiome may be an intermediary player in the interaction between the host and its environment in the extrinsic mechanisms that determine the transition from health to chronic disease.

Laboratory models have provided proof of the microbiome's ability to modulate immune functions of the host (Cho, I. & Blaser, M . J . The human microbiome: at the interface of health and disease. Nat. Rev. Genet 13, 260-270 (2012)) . Delivery mode has been studied intensively and the associations of caesarean section-birth with asthma (Sevelsted, A., Stokholm, J . & Bisgaard, H . Risk of Asthma from Caesarean Section Depends on Membrane Rupture. J. Peadiatrics (2016)) and other chronic childhood disorders have been largely consistent (Sevelsted, A., Stokholm, J ., Bpnnelykke, K. & Bisgaard, H. Cesarean section and chronic immune disorders. Pediatrics 135, e92-98 (2015)) . Furthermore, it is clear that delivery by Caesarean section greatly influences the earliest microbiome and its development. Antibiotics in the first year of life also have been confirmed both to perturb the intestinal ecosystem, as well as being associated with asthma risk.

A recent study (Arrieta, M .-C. et a!. Early infancy microbial and metabolic alterations affect risk of childhood asthma . Sci. Transl. Med. 7, 307ra l52 (2015)) identified associations between lower Faecalibacterium, Lachnospira, Veillonella, and Rothia levels and an early allergic wheezy phenotype at age 1 year.

Approximately 20% of children develop asthma-like symptoms in early life, and there is an urgent need for a better understanding of disease mechanisms to establish prevention strategies. For disease prevention, in some embodiments, a medical professional may intervene before the onset of disease in order to affect disease risk. There is a consequently a need for a targeted, and in some embodiments prophylactic, treatment of children at risk for developing asthma or other wheezing disorders. The present application fulfils such a need.

OBJECT

It is an object of embodiments described herein to provide a method for determining a risk of developing asthma or other wheezing disorders in a child and to provide a prophylactic and/or therapeutic method for the treatment thereof.

SUMMARY

It has now been found that lower abundances of one or more of the specific bacterial genera Roseburia, Bifidobacterium, Faecalibacterium, Alistipes, Ruminococcus, Ruminococcaecea , Bacteroides, Blautia, Lachnospiraceae incertae sedis, and Dialister during the first year of life are associated with increased risk of asthma, allergy and other wheezing disorders in children.

So, a first aspect relates to a composition comprising at least three bacterial genera chosen from Roseburia, Bifidobacterium, Faecalibacterium, Alistipes, Ruminococcus,

Ruminococcaecea, Bacteroides, Blautia, Lachnospiraceae incertae sedis, and Dialister.

A second aspect relates to a method for determining a risk of developing asthma or other wheezing disorders or allergy in a child comprising :

(a) identifying in a fecal sample from said child the abundance of at least one bacterial genus chosen from Roseburia, Bifidobacterium , Faecalibacterium, Alistipes, Ruminococcus, Ruminococcaecea , Bacteroides, Blautia, Lachnospiraceae incertae sedis, and Dialister;

(b) comparing the abundance to the abundance in a control group;

wherein a lower abundance in the child compared to a control group indicates an increased risk of developing asthma or other wheezing disorders or allergy.

A third aspect relates to a method for prophylactic or therapeutic treatment against asthma or other wheezing disorders or allergy of a child, comprising determining the child's risk of developing asthma according to the second aspect, and treating the child for asthma or other wheezing disorders or allergy if said risk is increased.

A fourth aspect relates to a method for prophylactic or therapeutic treatment of a child in risk of developing asthma or other wheezing disorders or allergy, comprising the steps:

a) identifying in a fecal sample from said child the abundance of at least one bacterial genus chosen from Roseburia, Bifidobacterium, Faecalibacterium , Aiistipes,

Ruminococcus, Ruminococcaecea , Bacteroides, Blautia, Lachnospiraceae incertae sedis, and Dialister;

b) comparing the abundance to the abundance in a control group;

c) if a lower abundance of at least one bacterial genus, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine or ten bacterial genera chosen from Roseburia, Bifidobacterium, Faecalibacterium, Aiistipes, Ruminococcus, Ruminococcaecea, Bacteroides, Blautia, Lachnospiraceae incertae sedis, and Dialister compared to a fecal sample from a control group is identified, administration of a medicament for inhibiting asthma or other wheezing disorders or allergy.

A fifth aspect relates to a method for prophylactic or therapeutic treatment of a child in risk of developing asthma or other wheezing disorders or allergy, comprising the steps:

a) identifying in a fecal sample from said child the abundance of at least one bacterial genus chosen from Roseburia, Bifidobacterium, Faecalibacterium , Aiistipes, Ruminococcus, Ruminococcaecea, Bacteroides, Blautia,

Lachnospiraceae incertae sedis, and Dialister;

b) comparing the abundance to the abundance in a control group; and if a lower abundance of at least one bacterial genus, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine or ten bacterial genera chosen from Roseburia, Bifidobacterium, Faecalibacterium, Aiistipes, Ruminococcus,

Ruminococcaecea, Bacteroides, Blautia, Lachnospiraceae incertae sedis, and Dialister compared to a fecal sample from a control group is identified, administration of a composition according to the first aspect. A sixth aspect relates to a composition according to the first aspect for use in the prevention or treatment of asthma or other wheezing disorders or allergy in a child.

A seventh aspect relates to a method for improving the microbiome in a child at risk of developing asthma or other wheezing disorders or allergy comprising administering the composition according to the first aspect to the child.

LEGENDS TO THE FIGURES

Figure 1 shows relative abundances in 1-year fecal samples among the 20 most abundant bacterial genera. Comparison of each genus is shown with respect to asthma at age 5 years in all children, and stratified by maternal asthma.

Figure 2 shows differences between asthmatic and non-asthmatic children in bacterial genera abundance at 1-year among children born by asthmatic mothers. Bacteroides,

Parabacteroides, Alistipes, Prevotella, Barnesiella, Odoribacter, Bacteroidales,

Porphyromonadaceae, and Butyricimonas belong to the phylum Bacteroidetes;

Faecaiibacterium, Blautia, Clostridium XlVa, Veillonella, Roseburia, Anaerostipes, Clostridium XI, Clostridium sensu stricto, Lachnospiracea incertae sedis, Ruminococcaceae, Dialister, Flavonifractor, Ruminococcus, Clostridium XVIII, Clostridiales, Clostridiaceae 1,

Erysipelotrichaceae incertae sedis, Clostridium IV, Dorea, Megasphaera, Butyrici coccus, Oscillibacter, Enterococcus, Turicibacter, Megamonas, Sporobacter, and Acidaminococcaceae belong to the phylum Firmicutes; Enterobacteriaceae, Pasteurellaceae, Sutterella, Escherichia Shigella, Parasutterella, Neisseria, Morganella, and Aiphaproteobacteria belong to the phylum Proteobacteria; Bifidobacterium, Eggerthella, Collinsella and Gardnerella belong to the phylum Actinobacteria ; and Fusobacterium belong to the phylum Fusobacteria.

DETAILED DISCLOSURE

Definitions

In the present context the term "lower abundance" in relation to bacterial genera refers to a content of the bacterial genus in question with a relative abundance of less than the median, such as less than 49, 48, 47, 46, 45, 44, 43, 42, 41, 40, 39, 38, 37, 36, 35, 34, 33, 32, 31, 30, 29, 28, 27, 26 %, such as in the lowest quartile (lowest 25%) compared to the relative abundances of the genus present in a control group. In the present context the term "higher abundance" in relation to bacterial genera refers to a content of the bacterial genus in question with a relative abundance of more than the median, such as more than 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66,

67, 68, 69, 70, 71, 72, 73, 74 %, such as in the highest quartile (highest 25%) compared to the relative abundances of the genus present in a control group.

In the present context the term "control group" refers to a group of healthy children, i.e. children who did not develop asthma or other wheezing disorders or allergy by the age of 5 years. Samples from a control group may include samples that were obtained before a patient was classified as part of the control group.

In the present context the term "asthma" refers to long-term inflammatory disease of the airways of the lungs characterized by recurrent wheeze (airway obstruction because of inflammation, causing difficulty in breathing, coughing and/or wheezing sounds when breathing against the obstruction).

In the present context the term "other wheezing disorders" refers to recurrent episodes of wheeze (airway obstruction because of inflammation, causing difficulty breathing, coughing and/or wheezing sounds when breathing against the obstruction), but does not fulfil the diagnosis of asthma.

In the present context the term "allergy" refers to an abnormal reaction of the body to a previously encountered allergen introduced by inhalation, ingestion, injection, or skin contact, often manifested by itchy eyes, runny nose, wheezing, skin rash, or diarrhea.

In the present context the term "probiotics" refers to living microorganisms which are intended for ingestion for humans and animals.

In the present context the term "synbiotics" refers to a combination of probiotics and prebiotics, wherein the term "prebiotics" refers to food ingredients that induce the growth or activity of beneficial microorganisms. Non-limiting examples of prebiotics include non- digestible fibre compounds.

Specific embodiments

The composition of the gut microbiome at the age of about 1 year has been found to be skewed in children who later develop asthma and other wheezing disorders or allergy. The effect is particularly pronounced in children born to asthmatic mothers. Children, who later become asthmatic, thus experience a delayed maturation of their gut microbiome compared to their healthy peers.

A first aspect thus relates to a composition comprising at least three bacterial genera chosen from Roseburia, Bifidobacterium, Faecaiibacterium, Alistipes, Ruminococcus,

Ruminococcaecea , Bacteroides, Biautia, Lachnospiraceae incertae sedis, and Diaiister. It has thus been found that bacteria of the above genera are present in the gut microbiome of children, who later become asthmatic, at lower levels than in the gut of a control group of children. Administration of the above composition is thus believed to be able to correct the gut microbiome.

In some embodiments the composition comprises at least four, five, six, seven, eight, nine or ten bacterial genera chosen from Roseburia, Bifidobacterium , Faecaiibacterium, Alistipes, Ruminococcus, Ruminococcaecea, Bacteroides, Biautia, Lachnospiraceae incertae sedis, and Diaiister. In some embodiments the composition comprises at least 10 ⁴ , 10 ⁵ , 10 ⁶ , 10 ⁷ , 10 ⁸ , or 10 ⁹ CFU of at least three, four, five, six, seven, eight, nine, or ten bacterial genera chosen from Roseburia, Bifidobacterium, Faecaiibacterium, Alistipes, Ruminococcus,

Ruminococcaecea , Bacteroides, Biautia, Lachnospiraceae incertae sedis, and Diaiister.

In some embodiments, the composition comprises at least four, five, six, seven, eight, nine or ten bacterial genera chosen from Roseburia, Bifidobacterium , Faecaiibacterium, Alistipes, Ruminococcus, Ruminococcaecea, Bacteroides, Biautia, Lachnospiraceae incertae sedis, and Diaiister, wherein pure strains of bacteria are combined. In some embodiments, purity comprises a pool of sequences where at least 95%, 96%, 97%, 98%, 99% members share 16S sequence identity or other sequence identity to their respective master cell bank reference strain. For instance, if the composition comprises at least four genera in equal amounts, each of the genera would be expected to represent at least 23.75% (25% x 95%) of the total sequence identity, as evaluated by 16S sequencing or other sequencing.

In some embodiments, the composition consists essentially of at least four, five, six, seven, eight, nine or ten bacterial genera chosen from Roseburia, Bifidobacterium, Faecaiibacterium , Alistipes, Ruminococcus, Ruminococcaecea , Bacteroides, Biautia, Lachnospiraceae incertae sedis, and Diaiister.

In some embodiments, the composition may contain no more than 1,000 CFU per dose of bacteria of the genera Pasteurellaceae, Escherichia/Shigella, Megasphera, Fusobacterium, Neisseria, Megamonas. It has thus been found that bacteria of the above genera are present in the gut microbiome of children, who later become asthmatic, at higher levels than in the gut of a control group of children. In some embodiments, the composition may contain no more than 750, 500, 250, 100, 50, 25, or 10 CFU per dose of bacteria of the genera

Pasteurellaceae, Escherichia/Shigella, Megasphera, Fusobacterium, Neisseria, Megamonas. In some embodiments, the composition contains no bacteria or no detectable amounts of bacteria from the genera Pasteurellaceae, Escherichia/Shigella, Megasphera, Fusobacterium, Neisseria, Megamonas. In some embodiments, no more than 1%, 2%, 3%, 4%, or 5% of a pool of 16S sequences or other sequence of the composition are attributable to the presence of Pasteurellaceae, Escherichia/Shigella, Megasphera, Fusobacterium, Neisseria, Megamonas.

A simple method for determining a risk of developing asthma or other wheezing disorders in a child is provided by method for determining a risk of developing asthma or other wheezing disorders or allergy in a child comprising :

(a) identifying in a fecal sample from said child the abundance of at least one bacterial genus chosen from Roseburia, Bifidobacterium , Faecaiibacterium, Alistipes, Ruminococcus,

Ruminococcaecea , Bacteroides, Blautia, Lachnospiraceae incertae sedis, and Dialister;

(b) comparing the abundance to the abundance in a control group;

wherein a lower abundance in the child compared to a control group indicates an increased risk of developing asthma or other wheezing disorders or allergy.

In some embodiments a lower abundance in relation to bacterial genera refers to a content of the bacterial genus in question with a relative abundance of less than the median, such as less than 49, 48, 47, 46, 45, 44, 43, 42, 41, 40, 39, 38, 37, 36, 35, 34, 33, 32, 31, 30, 29, 28, 27, 26 %, such as in the lowest quartile (lowest 25%) compared to the relative abundances of the genus present in a control group. In some embodiments a lower abundance refers to a content of less than 35, 34, 33, 32, 31, 30, 29, 28, 27, 26 %, such as in the lowest quartile (lowest 25%) . In some embodiments a lower abundance refers to a content of less than 30, 29, 28, 27, 26 %, such as in the lowest quartile (lowest 25%).

A method for prophylactic or therapeutic treatment against asthma or other wheezing disorders or allergy of a child is also provided, comprising determining the child's risk of developing asthma according to the above method, and treating the child for asthma or other wheezing disorders or allergy if said risk is increased.

In some embodiments the abundance of at least two, three, four, five, six, seven, eight, nine or ten bacterial genera chosen from Roseburia, Bifidobacterium , Faecaiibacterium, Alistipes, Ruminococcus, Ruminococcaecea, Bacteroides, Blautia, Lachnospiraceae incertae sedis, and Dialister are compared between the child and the control group and have a lower abundance in the child than the control group. It has also been found that a higher abundance of one or more bacterial genera chosen from Pasteurellaceae, Escherichia/Shigella, Megasphera, Fusobacterium, Neisseria, Megamonas during the first year of life are associated with an increased risk of developing asthma or other wheezing disorders or allergy in children.

In some embodiments the method for determining a risk of developing asthma or other wheezing disorders or allergy in a child further comprises (i) identifying in a fecal sample from said child the abundance of at least one bacterial genus chosen from Pasteurellaceae, Escherichia/Shigella, Megasphera, Fusobacterium, Neisseria, Megamonas;

(ii) comparing the abundance to the abundance in a control group, wherein a higher abundance in the child compared to a control group indicates an increased risk of developing asthma or other wheezing disorders or allergy.

In some embodiments a higher abundance in relation to bacterial genera refers to a content of the bacterial genus in question with a relative abundance of more than the median, such as more than 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74 %, such as in the highest quartile (highest 25%) compared to the relative abundances of the genus present in a control group. In some embodiments a higher abundance refers to a content of more than 65, 66, 67, 68, 69, 70, 71, 72, 73, 74 %, such as in the highest quartile (highest 25%) . In some embodiments a higher abundance refers to a content in the highest quartile (highest 25%).

In some embodiments the abundance of at least two, three, four, five, or six bacterial genera chosen from Pasteurellaceae, Escherichia/Shigella, Megasphera, Fusobacterium, Neisseria, Megamonas are compared between the child and the control group and have a higher abundance in the child than the control group.

In some embodiments treating the child for asthma or other wheezing disorders or allergy comprises anti-asthmatic treatments with e.g. beta2-agonists, corticosteroids etc.

In some embodiments of the method the child is an infant of the age from 1 day to 2 years, such as 1 day to 18 months, such as 1 week to 18 months, such as 2 weeks to 18 months, such as from 2 weeks to 16 months, such as from 2 weeks to 1 year. In this period of life the microbial population of the gut matures and an immature composition at this period associates with an increased risk of asthma, other wheezing disorders or allergy. Thus supplementation in this period may prevent disease and/or improve a child's overall microbiome profile. The composition may be used as a medicament. The composition in the form of a

medicament may be administered in the form of a pharmaceutical formulation, e.g., in admixture with a suitable pharmaceutical excipient, diluent or carrier selected with regard to the intended route of administration and standard pharmaceutical practice. The excipient, diluent and/or carrier must be "acceptable" in the sense of being compatible with the other ingredients of the formulation and not deleterious to the recipient thereof. Acceptable excipients, diluents, and carriers for therapeutic use are well known in the pharmaceutical art, and are described, for example, in Remington: The Science and Practice of Pharmacy. Lippincott Williams and Wilkins (A.R. Gennaro edit. 2005). The choice of pharmaceutical excipient, diluent, and carrier can be selected with regard to the intended route of administration and standard pharmaceutical practice. Although there are no physical limitations to delivery of the compositions, in some embodiments oral delivery is used for delivery to the digestive tract because of its ease and convenience, and because oral compositions readily accommodate additional mixtures, such as water, breastmilk, cow's milk, yogurt, and infant formula. Bacteria can be also administered via naso/orogastric gavage, recta I ly, or via fecal route (by enema). Other delivery forms, such as food, nutraceutical, supplement, probiotic or synbiotic, may be prepared in analogy with the above.

Furthermore, in children born to an asthmatic mother, by caesarean section or by mothers treated with antibiotics during pregnancy or delivery, or in children treated with antibiotics during delivery or during the first year of life, the bacterial composition is skewed in early life because of this. Delivery by caesarean section has in multiple studies been associated with later development of asthma. The composition can be used to establish a healthy bacterial composition in early life in children treated with antibiotics during delivery or during the first year of life, in children born by caesarean section or in children born by mothers treated with antibiotics during pregnancy or delivery, and may thereby protect the child from asthma and other wheezing disorders. In some embodiments, the composition can be used to establish a healthy bacterial composition in early life in children who had antibiotic treatment during delivery (within 1 hour of delivery), during the first week of life, during the first three weeks of life, during the first month of life, during the first three months of life, during the first six months of life, during the first nine months of life, during the first year of life, or during the first 2 years of life.

As childhood asthma and allergy are related diseases, the compositions may also be used for the prevention of allergy. EXAMPLE 1

METHODS

A study was conducted at CO PS AC (Copenhagen Prospective Studies on Asthma in

Childhood) in accordance with the guiding principles of the Declaration of Helsinki and was approved by the Local Ethics Committee (H-8-2008-093), and the Danish Data Protection Agency (2015-41-3696) . All parents gave written informed consent before enrolment.

Study Population

The COPSAC2010 cohort is a population-based birth cohort of 700 children recruited at one- week of age and has been followed prospectively at the COPSAC research unit with deep clinical phenotypeing at 11 scheduled visits during the first 5 years of life. The study pediatricians collected all information during these clinical visits scheduled at 1 week, 1, 3, 6, 12, 18, 24, 30, and 36 months, and yearly thereafter. Additional acute visits were arranged whenever the children experienced lung or skin symptoms. The symptom burden between visits was captured with daily diary cards monitoring : (1) significant troublesome lung symptoms including components of cough, wheeze, and dyspnea; (2) skin symptoms; and (3) respiratory infections. The study pediatricians acted as general practitioners for the cohort and were the only physicians responsible for diagnosis and treatment of asthma, allergy, and eczema adhering to predefined algorithms.

Primary End-point

Asthma was diagnosed based on a previously detailed quantitative symptom algorithm requiring all of the following criteria : (1) verified diary recordings of 5 episodes of

troublesome lung symptoms within 6 months, each lasting at least 3 consecutive days; (2) symptoms typical of asthma including exercise-induced symptoms, prolonged nocturnal cough, and persistent cough outside of common colds; (3) need for intermittent rescue use of inhaled 2-agonist; and (4) response to a 3-month course of inhaled corticosteroids and relapse upon ending treatment. Remission was defined by 12 months without relapse upon cessation of inhaled corticosteroid treatment. For analyses, the cross-sectional ongoing asthma diagnosis at age 5 years and the time to disease onset of either transient or persistent asthma (ongoing diagnosis after age 5 years) was used .

Allergic sensitization was determined at 6 and 18 months of age as any skin prick test (SPT) >2 mm (ALK-Abello, Horsholm, Denmark) and by specific IgE (s!gE) ³ 0.35 kUa/L against milk, egg, dog or cat (ImmunoCAP; Thermo Fischer Scientific, Allerod, Denmark). Children classified as "not sensitized" were both SPT and specific IgE negative for all tested allergens.

Samples collection and preparation

Fecal samples were collected 1 week, 1 month and 1 year after birth. Each sample was mixed with 1 ml_ of 10% vol/vol glycerol broth (SSI, Copenhagen, Denmark) and frozen at -80°C. DNA was extracted using MoBio PowerSoil kits on an EpMotion 5075, amplified using a two- step PCR reaction with 515F and 806R primers flanking the V4 region of 16S rRNA gene, and sequenced using V2 kit (PE250bp) and the MiSeq platform (lllumina Inc., San Diego CA). Fastq-files demultiplexed by the MiSeq Controller Software were trimmed for amplification primers, diversity spacers, and sequencing adapters, mate-paired and quality filtered (USEARCH v7.0.1090). UPARSE was used for Operational Taxonomic Unit (OTU) clustering as recommended, in particular removing singletons after dereplication. Chimera checking was performed with USEARCH. Representative sequences were classified at 0.8 confidence threshold (Mothur v.1.25.0 wangQ function). FastTree and Mothurs align. seq() function were used to construct a phylogenetic tree. Alignments were built against the 2013 version of Greengenes.

Statistics and data analysis

All data analysis was performed in the statistical software package R version 3.3.0, with the package phyloseq to handle the microbiome data. Differences in relative abundances at the genus level were analyzed using Wilcoxon rank-sum. We used sparse Partial Least Squares (PLS) modeling of asthma at age 5-years after filtering genera (prevalence >5% of children, >0.01% mean relative abundance) and log-transforming relative abundances, using half the lowest nonzero value as a pseudocount. We selected the optimum number of input variables using repeated 10-fold cross-validation of the Area Under the Curve (AUC) statistic to avoid overfitting. Clustering analysis was performed using partitioning around medoids (PAM) clustering from the package cluster. The Silhouette index calculated from weighted UniFrac distances was used for determining the optimal number of clusters in the data. Microbiota- by-age z-scores (MAZ) were calculated with non-asthmatics as training set: Microbial maturity (MM)= (predicted microbiota age-median microbiota age). MAZ = (MM/SD) of predicted microbiota age. Associations between PAM clusters and MAZ and asthma were quantified by Cox proportional hazards regression. A significance level of 0.05 was used in all analyses. RESULTS

At least one fecal sample was analyzed for 690 (99%) of these children. Of these children,

356 (52%) were boys, 390 (57%) had at least one older child in the home at birth, 148 (21 %) were delivered by Caesarean section, and 311 (46%) were treated with antibiotics during the first year of life. The mean maternal age at delivery was 32.3 years and 179 (26%) of the mothers had physician-diagnosed asthma .

Genus abundance and risk of asthma

Among the 648 children with follow-up to age 5, the prevalence of ongoing asthma at age 5 year 5 was 9% (N =60) . For the 20 most abundant genera present in the 1-year samples, an increased risk of asthma at age 5-years was found to be associated with a somewhat higher abundance of Veillonella (asthma vs. non-asthma ; median relative abundance, 0.94% vs. 0.29% ; P=0.035) and with lower abundance of Roseburia (0.27% vs. 0.66% ; P=0.042), Alistipes (0.04% vs. 0.35% ; P=0.002), and Flavonifractor (0.05% vs. 0.07%; P=0.002) . If the child was born to an asthmatic mother, 8 of these 20 genera were significantly associated with the development of asthma in the child, whereas there were no significant associations if the mother did not have asthma . In children born to asthmatic mothers, asthma at age 5 years was negatively associated with relative abundance in the age 1-year sample of the genera Faecalibacterium (0.59% vs. 3.27% ; P=0.010), Bifidobacterium (0.47% vs. 2.27%; P=0.006), Roseburia (0.01% vs. 0.76%; P O.001), Alistipes (0.01% vs. 0.39%; P=0.003), Lachnospiraceae incertae sedis (0.05% vs. 0.19% ; P=0.018), Ruminococcus (< 0.01% vs. 0.13%; P=0.004) and Dialister (<0.01% vs. 0.15%; P=0.007 and potentially positively correlated with Veillonella (1.41% vs. 0.23%; P=0.039) . See Figure 1.

To further examine the associations between the microbial composition at age 1-year and asthma at age 5 in children born to asthmatic mothers, a cross-validated sparse PLS model was constructed to identify jointly contributing taxa at age 1-year that would predict asthma in these children. This resulted in a one-component model based on the relative abundances of 60 genera . The model demonstrated a high predictive capacity for asthma (cross-validated AUC 0.76) and the many contributing taxa suggest a global delayed microbial maturation in these children at age 1-year. This analysis revealed the taxa most associated with asthma; negative loadings (protective) : Roseburia, Bifidobacterium, Faecalibacterium, Alistipes, Ruminococcus, Ruminococcaecea, Bacteroides, Dialister, Blautia, and Lachnospiraceae incertae sedis; and positive loadings (accelerators) : Pasteurellaceae, Escherichia/Shigella, Megasphera, Fusobacterium, Neisseria, Megamonas, see Figure 2. Veillonella also has a positive loading, but the study of Arrieta, M.-C. et a/. (2015) suggests potentially protective effects of Veillonella.

Community types and risk of asthma

The PAM clustering algorithm was used to find the optimal compositional separation in the microbial populations of all the faecal samples by weighted UniFrac distances, which divided the samples into two distinct PAM clusters (silhouette width, 0.30), largely representing the child's age at sampling, with PAM cluster 1 (N = 1019) mainly composed of the 1-week and 1- month samples and PAM cluster 2 (N=677) chiefly composed of the 1-year samples. As such, use of these PAM clusters may be representative of the age-related maturation of the intestinal microbial populations. The risk of developing persistent asthma if the child's microbiome remained in the immature PAM cluster 1 at 1-year of age (N =34) was compared to children with transition to PAM cluster 2 (N = 589). During the first 5 years of life, the risk of developing persistent asthma was increased (adjusted hazard ratio (aHR) 2.87 [1.25- 6.55], P=0.013) if the microbiome remained in PAM cluster 1 at age 1-year. This effect was driven solely by the 147 children born to asthmatic mothers (aHR 12.99 [4.17- 40.51], P O.001), whereas there was no microbial effect on asthma development for the 442 children of non-asthmatic mothers (aHR 0.56 [0.07- 4. 16], P=0.57), indicating significant interaction, P=0.011.

Microbial maturity and risk of asthma

To further explore the microbial maturation process, we calculated microbiota-by-age z- scores (MAZ) for all time points as a marker of maturity. During the first 5 years of life, the risk of developing persistent asthma was increased with low maturity (below median) in 1- year MAZ (HR 1.77 [1.02-3.07], P=0.043). Low microbial maturity was only associated with asthma in children born to asthmatic mothers (HR 6.53 [1.93-22.06), P=0.003), and not in children of non-asthmatic mothers (HR 0.92 [0.46-1.84), P=0.81), indicating significant interaction, P=0.006.

Asthma with or without allergic sensitization

To explore whether the results were characterizing a specific asthmatic phenotype, the children were further subdivided by allergic sensitization in childhood. The microbial populations by means b-diversity at age 1-year were significantly different whether the child had asthma with or without sensitization (4 categories) (F=1.7, R ² =0.9%, P=0.020).

Especially the phenotype of asthma with allergic sensitization was associated with skewed b- diversity. Again, this was mainly apparent among children born to asthmatic mothers (F=2.9, R ² =5.9%, P<0.001) with no significant effects among children born to non-asthmatic mothers F=0.7, R ² =0.5%, (P=0.887).

DISCUSSION

In a cohort of 700 children with the primary objective of making accurate asthma diagnoses and comprehensive phenotyping, it was found that the gut microbiome at age 1-year was associated with asthma at age 5-years. Parallel observations were apparent for genus abundance, and maturation by means of PAM clustering and MAZ. The effect was especially pronounced in children born to asthmatic mothers, suggesting a mechanism of susceptibility to microbial impacts among these children. Furthermore, a skewed microbial composition at age 1-year was especially associated with an asthmatic phenotype also characterized by allergic sensitization.

The microbial populations of the children were not associated with maternal asthma status, thereby diminishing the possibility of maternal asthma being the causative factor for both microbial composition and asthma risk. Maternal asthma was a key effect modifier between the microbiome and asthma risk in our study. The lack of significant effects in children born to non-asthmatic mothers, and of paternal contribution point to susceptibility to host- microbial interactions specifically for these children. This could arise from an inborn immune deviation determined by maternal asthma status, as we have previously reported in this cohort. Stronger heritability of maternal over paternal asthma has been described, consistent with our findings, which points towards mechanisms beyond genetic effects.