Probiotics for the prevention or treatment of allergic diseases

The role of microbiota in emerging models of allergy pathogenesis 

It is now proposed that allergic disease results from a fundamental failure of underlying immune regulation. Microbial exposure arguably provides the strongest environmental signal for normal postnatal maturation of the immune system and also induces the maturation of antigen presenting cells and T-regulatory cells, which are essential for programming and regulating the T-cell response. It appears likely that microbial activation of regulatory pathways through microbial pattern recognition molecules (Toll-like receptors [TLRs]) plays a central role in reducing the risk of immunologically mediated disease, including TH2-mediated allergic responses, and possibly also TH1-mediated autoimmune disease, such as type I diabetes. Thus, reduced microbial exposure may be contributing to the rising rates of this wide spectrum of immune diseases. This is likely to be of greatest relevance in early life when immune programming is initiated and less significant in relation to a mature immune system in older children and adults.

The gut microbiota is the major source of microbial exposure, composed of 1014 microorganisms, or 10 times the number of cells in the entire body with a 30 times larger total genome than the human genome. Microbial colonization of the gastrointestinal tract, linked with lifestyle and/or geographic factors, may be important an determinant of the heterogeneity in disease prevalence throughout the world,5 and ongoing cohort studies are focusing in detail on this complex question. These suggestions are supported by observations that germ-free mice do not develop tolerance in the absence of a gut microbiota in addition to the observed differences in the composition of the gut microbiota between infants living in countries with a high and a low prevalence of allergy and between healthy infants and infants with allergy (see summary5).

Probiotics: mechanisms of action 

Probiotics are defined as living microorganisms which, on ingestion in certain numbers, exert health benefits beyond inherent general nutrition. There is now good evidence that certain strains of lactobacilli and bifidobacteria can influence immune function through a number of different pathways (summarized in Table II) including effects on enterocytes, antigen presenting cells (including both circulating monocytes and local dendritic cells [DCs]), regulatory T cells, and effector T and B cells. Importantly, however, the relationship between the various reported effects (Table II) and clinical consequences of treatment are unknown. Because there are very few studies in which several allegedly probiotic strains have been compared, it is not known to what extent a finding using a certain bacterial strain is relevant for other strains, even of the same species. To date, there are only a few strains, limited to lactobacilli, that have been reasonably well documented in clinical studies, mostly against infectious gastroenteritis and lactose intolerance.

Table II. Immunologic pathways affected by probiotics

	Pathway	Documented effects in human beings and/or animals	Proposed mechanism of immunomodulation
	Local effects
	Effects on mucosal barrier	Repair and maintenance of intestinal barrier and tight junctions	Reduced permeability and reduced systemic penetration of allergens/antigens7
		Increased mucous production
	Enterocytes	Reduced cell signaling via nuclear factor-κB local inflammation6	Reduced local inflammation/promotion of tolerogenic conditions
		Increased production of TGF-β and prostaglandin E2, which promote tolerogenic DCs9
	Innate mucosal recognition (TLR)	Anti-inflammatory effects of probiotics mediated by TLR9 8	Inhibition of TH2 allergic responses: mechanisms unclear
		Possible changes in TLR2 in vitro37	TLR2/4 agonists shown to reduce inflammation in murine lung
	DCs	Increased activity of DCs in human gut10	Promotes tolerogenic DC (IL-10 production)
	Effector T cells	TH1 skewed responses observed12	Inhibition of TH2 differentiation?
			Effects of T-cell trafficking?
	Treg	CD4+CD28+ T cells associated with oral tolerance	TGF-β produced locally (including by enterocytes) promotes tolerogenic DC, local IgA, and Treg activity
		T-cell–producing IL-10 and TGF-β associated with oral tolerance
		Probiotics: Increased TGF-β (TH3) Treg13
	B cells and antibodies	Colonization: increased lymphoid tissue	Promotes tolerogenic microenvironment (as above); IgA may reduce systemic antigen load
		Probiotics: increased local IgA production
	Systemic effects
	Monocytes	Probiotics: increased circulating monocytes19, 20	Mechanisms not known
	T cells	Probiotics: increased TH1 differentiation	Secondary to effects on T cells trafficking through gut?
	B cells/IgA	Probiotics: increased IgA production in other tissues (respiratory tract)18	Secondary to effects on B cells trafficking through gut?
	Stem cells	Probiotics: increased circulating bone marrow derived CD34+ stem cells21	Mechanisms not known

Treg, Regulatory T cells.

Locally in the gut, there is evidence that commensal gut bacteria help reduce local inflammation,6 and at least 1 probiotic strain has the capacity to maintain the integrity of the intestinal barrier,7 potentially reducing systemic antigen load. At least some of the anti-inflammatory effects appear to be mediated through TLR, including TLR98 and possibly TLR2 and TLR4 expressed on enterocytes. Intestinal microbiota also promote enterocyte production of TNF-β and prostaglandin E2, which promote the development of tolerogenic DCs.9 Other studies have also shown that probiotics directly enhance the activity of human DC populations10, 11 to promote TH1 differentiation.12 Consistent with notions that bacteria promote regulatory function, there is also preliminary evidence that probiotics promote immunoregulatory activity in the gut. In animal studies, probiotic supplementation can induce regulatory T-cell populations (bearing TGF-β),13 and human studies also suggest an increase in the in vitro production of regulatory cytokines (IL-10) after probiotic ingestion.14 The effects may be limited to certain species, as indicated by a recent study in which Lactobacillus reuteri and Lactobacillus casei, but not Lactobacillus plantarum, primed monocyte-derived DCs to drive the development of regulatory T cells.15 These regulatory T cells produced increased levels of IL-10 and were capable of inhibiting the proliferation of bystander T cells in an IL-10–dependent fashion. Interestingly, the 2 former species, but not L plantarum, bound the C-type lectin DC-specific intercellular adhesion molecule 3–grabbing nonintegrin, and blocking antibodies to DC-specific intercellular adhesion molecule 3–grabbing nonintegrin inhibited the induction of the regulatory T cells by these probiotic bacteria.

Intestinal microbiota also influences IgA production in distal sites (respiratory tract). The gastrointestinal tract makes up a critical part of the integrated common mucosal immune system, which includes mucosal surfaces across anatomically remote locations (namely the gastrointestinal tract and respiratory tract). It is well recognized that mucosa-homing IgA-producing B cells and effector T cells mature in the gut mucosa before seeding to distal mucosal sites in the respiratory tract. This provides a possible explanation for how gut microbiota appear to enhance systemic TH1 responses16, 17 and IgA production in remote tissues.18

It is less clear how probiotics influence other bone marrow–derived populations of cells that do not traffic directly through the gut. Circulating monocytes in infant animals mature at significantly different rates depending on enteric microbiota exposure,19 with 2-fold lower function in germ-free animals.20 Potential marrow effects are supported by studies showing that changes in gut microbiota have effects on bone marrow CD34+ progenitor populations entering the circulation.21 Clearly a better understanding of the systemic effects of gut microbiota is necessary to explore causal pathways and the potential of probiotics as preventive and therapeutic agents.

Evidence of clinical benefits of probiotics in treatment of allergic disease 

Most studies to explore the role of probiotics in the treatment of allergic disease have focused on early manifestations of allergy, namely food allergy and atopic dermatitis. Studies in older individuals with established respiratory disease have failed to show any improvement in asthma22 or allergic rhinitis,23 although one larger study reported improved quality of life in patients with allergic rhinitis.24 This is consistent with findings in animal studies in which microbial products have more significant effects when immune responses are still developing than once sensitization is established.

There are now a number of studies that have addressed use of probiotics in young children with atopic dermatitis with encouraging (but not strong) effects.4, 25, 26, 27, 28 The studies are limited to 3 species of lactobacilli. The first study to address this demonstrated a clinical improvement (> 50% reduction in Scoring Atopic Dermatitis [SCORAD] index) in infants with atopic dermatitis and cow's milk allergy when fed probiotic-fortified hydrolyzed whey formula (n = 13) during the 1-month treatment period compared with a placebo group (which also subsequently improved).25 The second study included children with mild disease (median SCORAD of 16 at inclusion) and documented a complete resolution in all participants after 6 months, although this occurred more rapidly in the group receiving probiotics (Lactobacillus GG [n = 9] or Bifidobacterium lactis [n = 9]).4 These were both very small studies and only included infants with mild disease who are less likely to be atopic and less likely to develop persistent cutaneous disease or new respiratory allergy. More recently, a further study in a larger cohort26 reported a slightly greater reduction in SCORAD (−26.1 vs −19.8; P = .036) after 4 weeks of treatment in IgE sensitized infants with atopic dermatitis (age around 6 months) receiving the same probiotic strain (Lactobacillus GG). It is of note that clinical effects of this probiotic were only seen in children with evidence of allergic sensitization and not in children with atopic dermatitis but no sensitization who received the same probiotic. This suggests that atopic dermatitis is a heterogeneous condition and that the effect of immune modifying agents such as probiotics will depend on the pattern of disease. A number of subsequent studies with other strains of lactobacilli have also suggested some favorable effects on atopic dermatitis extent and severity. In one of the studies,27 in which Lactobacillus rhamnosus and L reuteri were given in combination, 56% of the patients experienced improvement of the eczema compared with 15% in the placebo group (P = .001). The second study demonstrated improved SCORAD in 92% of the children receiving a L fermentum strain compared with 63% in the placebo group (P = .01).28 In general, however, the effects are not strong or not evident at all.29

In 2 studies, probiotic administration (Lactobacillus rhamnosus GG17 and Lactobacillus fermentum PCC16) was associated with increased polyclonal TH1 IFN-γ responses in the infants. Of note, the improvement in atopic dermatitis was directly proportional to the increase in IFN-γ responses to Staphylococcus enterotoxin B (r = 0.445; P = .026).16

In summary, although there are studies suggesting favorable effects of probiotics on atopic dermatitis, it is generally accepted that larger, controlled studies with well defined probiotic bacteria and perhaps mixtures of several such strains are needed to determine the role of these products in therapy. The lack of effect of these products in older individuals (with asthma and allergic rhinitis)22, 23, 24 suggests that any beneficial effects could be limited to early life before allergic disease is established.

Role of probiotics in prevention of allergic disease 

It is logical from an immunologic standpoint to explore the benefits of probiotics very early in life when immune responses are still developing, and there are now a number of studies addressing the role of probiotics in primary allergy prevention. The first study to assess the role of probiotics in this context administered L rhamnosus to mothers (starting 2-4 weeks before delivery) and to infants in the first 6 months of life. This was reported to reduce the incidence of eczema at 2 years by around 50%.3 Although the cumulative effect on eczema was still evident at 4 years, there was no reduction in respiratory allergy, IgE levels, or allergic sensitization.30 Effects on underlying immune response were not reported, and a number of methodologic concerns have been raised about the study.31 A major concern was that many of the children (28 out of 64) included in the probiotic supplement group did not receive probiotics directly, because the supplement was given to the mother if babies were breast-fed. These issues have made the results difficult to interpret.

There are now 3 other published studies and at least 7 other studies still in progress (summarized in Table III) to examine the effects of various probiotic strains for allergy prevention, most using direct infant supplementation. The first of these (using Lactobacillus acidophilus) failed to show any reduction in allergic disease despite changes in colonization.32 Rather, there was a concerning increase in sensitization and in IgE-associated atopic eczema. The second showed a reduction in atopic eczema (odds ratio, 0.66; 95% CI, 0.46-0.95; P = .025) but no effects on sensitization or other allergic disease.33 Of note, this study used a combination of strains and prebiotic galacto-oligosaccharides (as detailed in Table III). The third study showed no effect of L reuteri on the prevention of allergic disease or sensitization.34 However, subgroup analyses showed that probiotics were linked with less IgE-associated atopic eczema (8% vs 19% in placebo-treated group; P < .05) and less sensitization in a subgroup with atopic mothers (17% vs 31%; P < .05). Prospective analysis of these populations is necessary to assess long-term outcomes, particularly any possible effects on respiratory allergy. The results of the other studies (Table III) are awaited with interest. At this stage it is not appropriate to recommend probiotics for allergy prevention. Despite all of the immunomodulatory effects described in experimental models, so far none of these studies has shown any clear effect on preventive sensitization or any allergic disease other than eczema.

Table III. Summary of probiotic primary prevention studies (completed and in progress)

		Study protocol				Outcomes
	Investigators and location of study	Population characteristics	Organisms and dosage	Prenatal duration	Postnatal duration	Reduction in eczema	Reduction in sensitization	Reduction in other AD	Effect on colonization
	Kalliomaki, Isolauri, and others, Turku, Finland3, 33	Any first-degree relative with allergic disease	L rhamnosus GG, 1 × 1010 CFU daily	Yes	6 mo	Yes	No	No	Yes
		n = 132 completed (of initial 159)	Only to mother if breast-feeding postnatally	2-4 wk before delivery	∗only directly to baby if not breast-feeding	(at 2 and 4 y)
	Taylor, Prescott, and others, Perth, Australia32	Mother with SPT+ allergic disease	L acidophilus (3 × 108 CFU daily)	No	6 mo	No	No†	No	Yes
		n = 189 completed (of initial 230)			(direct to infant)	(at 1 y)
	Kukkonen, Kuitunen, and others, Helsinki/Tampere, Finland33	One or both parents with allergic disease	L rhamnosus GG and LC705 (both 5 × 109 CFU twice daily); and Bifidobacterium breve and Proprionibacterium freudenreichii (both 2 × 109 CFU twice daily)	Yes	6 mo	Yes	No	No	Yes
		n = 925 completed (of initial 1223)		2-4 wk before delivery	(direct to infant)∗	(at 2 y)
	Abrahamsson, Oldaeus, and others, Linköping, Sweden34	Any first-degree relative with allergic disease	L reuteri (1 × 108 CFU daily)	Yes	12 mo	No‡	No‡	No	Not known
		n = 188 completed (of initial 232)		2-4 wk before delivery	(direct to infant)	(at 2 y)
	Kopp and others, Germany	Any first-degree relative with allergic disease	L rhamnosus GG, 1 × 1010 CFU daily	Yes	6 mo	Study complete
		n = 94 completed (of initial 102)	To mother if breast-feeding postnatally for 3 mo and than to the neonates for additional 3 mo	4-6 wk before delivery		Clinical analysis complete (awaiting publication)
						No difference in proliferative response or T-cell reactivity in cord blood and in maternal blood between serum and placebo groups
	Thornton, Morgan, and others, Swansea, United Kingdom	Any first-degree relative with allergic disease	Lactobacillus salivaris (6.25 × 108 CFU daily) and Lactobacillus paracasei (1.25 × 108 CFU daily) and Bifidobacterium infantis (1.25 × 108 CFU daily) and Bifidobacterium bifidum (1.25 × 108 CFU daily)	Yes	6 mo	Study not complete
		n = 600 intended		2-4 wk before delivery	(direct to infant regardless of feeding method)	Outcomes will be assessed at 6 mo, 2 and 5 y
	Wickens, Crane, and others, Wellington and Auckland, New Zealand	One or both parents with allergic disease	L rhamnosus (1 × 1010 CFU daily) or B lactis (1 × 1010 CFU daily)	Yes	2 y	Study not complete
		n = 450 intended		2-5 wk before delivery	(to infant regardless of feeding method)	Outcomes will be assessed at 3,6, 12, 18, and 24 mo
	Tang and others, Melbourne, Australia	Any first-degree relative with allergic disease	L rhamnosus (2 × 1010 CFU daily)	Yes	No	Study not complete
		n = 200 intended		2-4 wk before delivery	(only to mother)	Outcomes will be assessed at 3,6, 12, and 24 mo
	Skek, Aw, and others, Singapore	Any first-degree relative with SPT+ allergic disease	L rhamnosus (1 × 109 CFU daily) and Bifidobacterium longum (6 × 108 CFU daily)	No	6 mo	Study not complete
		n = 300 intended			(in infant formulae)	Outcomes will be assessed at 1, 3, 6, 12, and 24 mo and 5 y
	Niers, Rijkers, Hoekstra, and others, Utrecht, The Netherlands	Past or present atopic disease in either mother or father plus at least 1 sibling	Lactococcus lactis Bifidobacterium infantis Bifidobacterium bifidum (all 1 × 109 CFU daily)	Yes	12 mo	Initial assessment at 3 mo now completed (awaiting publication)
		n = 120 intended		4 wk before delivery	(direct to infant)	Study not complete
	Lau, Wahn, and Hamelmann, Berlin, Germany	Infants at age 4 wk with at least 1 atopic parent	Streptococcus faecalis DSM 16440 and Escherichia coli DSM 17252 (combined at 1.5-4.5 × 107 daily)	No	6 mo (wk 5 to end of 7th mo)	Primary outcome: effect on eczema at age 7 and 12 mo; follow-up period after 6-mo treatment period up to age 3 y
		n = 650			(direct to child)	Secondary outcome: sensitization and other allergic symptoms; gut flora will be studied§

AD, Allergic disease; CFU, colony-forming units.

∗ Probiotic group also received prebiotic supplement (galacto-oligosaccharides). † Sensitization more common in probiotic group. ‡ Probiotics: less IgE-associated eczema in the second year, and less sensitization in subgroup with atopic mothers. § This study is using heat-killed (rather than live) bacteria.

Factors that may account for varied effects of probiotics in different studies 

Possible explanations for the varied results include differences in the bacterial strains used, host factors that could influence microbial responsiveness and allergic propensity, and other environmental factors that could influence colonization or immune development (Fig 1). First, there are significant variations in the strains claimed to be probiotic. It is also of note that the studies that suggested preventative effects started supplementation in pregnancy,3, 33, 34 whereas the study that showed increased sensitization did not.32 This may indicate that the supplementation to the mothers in late pregnancy is of a crucial importance. It is of interest to note that in 1 study, the levels of IL-10 were higher and TGF-β lower in colostrum of mothers receiving a probiotic compared with placebo-treated mothers.35 Second, there are differences in host susceptibility to microbial influence and to colonization with a particular strain of bacteria. Functional genetic polymorphisms in microbial recognition pathways are well described (including TLR), and it is likely that this could result in individual variation in the effects of probiotics. Similarly, there is some heterogeneity in the level of allergic risk in these studies. Third, there are likely to be many environmental factors that influence both colonization (such as maternal microbiota and other sources of microbial exposure, delivery method, antibiotics, prebiotics in the diet, and general microbial burden) and immune development. All of these factors are likely to make robust meta-analyses problematic to perform as more studies are completed.

View Large Image Download to PowerPoint

Fig 1. Factors that could explain the varied effects of probiotics.

Future directions: prebiotics? 

Although some studies have reported benefit in the treatment and prevention of atopic eczema, none has had any clear effects on the development TH2 mediated allergic responses. It appears unlikely that supplementation with a single probiotic strain would be sufficient to have a major influence on the very diverse intestinal microbiota and the complex interaction between the gut bacteria and the host. This has led to a shift in interest to dietary substrates that could have a more global effect on gut microbiota—namely, prebiotics (nondigestible, fermentable oligosaccharides that stimulate the growth of Bifidobacterium and Lactobacillus species). Altering the intake of foods containing these products can directly influence the composition and activity of intestinal microbiota. This could explain some of the protective effects of grains and cereals that have been seen in epidemiologic studies. At this stage, there are still very little data to confirm directly the immunologic or therapeutic effects of prebiotic supplements, but a number of studies are underway.

Conclusion 

Although there is a sound theoretical basis for anticipating benefits of probiotic supplementation in allergic disease, there is currently insufficient data to recommend this as a part of standard therapy in any allergic conditions or for prevention. Although there has been early promise in atopic dermatitis, it is generally accepted that more studies are needed to confirm this, and that any benefits are not likely to be great. However, faced with the stress and severe discomfort that can be associated with atopic dermatitis, many families are still choosing to try probiotics in conjunction with their prescribed treatment. Although the microbiotica in probiotic preparations are generally safe, it is possible that some products could contain milk products and may cause anaphylaxis in children with milk allergy.36 Furthermore, although only 1 prevention study has reported adverse outcomes in relation to sensitization (with increased risk32), the significance of this needs to be examined in further studies. These observations provide a cautionary note amid the continuing public enthusiasm for probiotics.