Intestinal microbiota and juvenile idiopathic arthritis:current understanding and future prospective

Le Xin·Feng He·Sen Li·Zhi-Xuan Zhou·Xiao-Lin Ma

Abstract Background Juvenile idiopathic arthritis (JIA) characterized by arthritis of unknown origin is the most common childhood chronic rheumatic disease,caused by both host genetic factors and environmental triggers.Recent evidence has mounted to focus on the intestinal microbiota,a potentially recognized set of environmental triggers affecting JIA development.Here we offer an overview of recently published animal and human studies that support the impact of intestinal microbiota in JIA.Data sources We searched PubMed for animal and human studies publications with the search terms“intestinal microbiota or gut microbiota”and“juvenile idiopathic arthritis or juvenile chronic arthritis or juvenile rheumatoid arthritis or childhood rheumatoid arthritis or pediatric rheumatoid arthritis”.Results Several comparative studies have demonstrated that intestinal microbial alterations might be triggers in disease pathogenesis.Alternatively,a slice of studies has suggested environmental triggers in early life might disrupt intestinal microbial colonization,including cesarean section,formula feeding,and antibiotic exposure.Aberrant intestinal microbiota may influence the development of JIA by mediating host immune programming and by altering mucosal permeability.Conclusions Specific microbial factors may contribute to the pathogenesis of JIA.Intensive studies,however,are warranted to investigate the causality between intestinal dysbiosis and JIA and the mechanisms behind these epidemiologic relationships.Studies are also needed to design the best interventional administrations to restore balanced intestinal microbial communities.

Keywords Antibiotic exposure·Cesarean section·Immune system reprogramming·Intestinal microbiota·Juvenile idiopathic arthritis

Introduction

Juvenile idiopathic arthritis (JIA) is the most common chronic rheumatic condition in childhood.It is not a single disease entity,but rather a group of genetically heterogeneous and phenotypically distinct disorders.JIA is thought to be multi-factorial with a strong interaction between genetic susceptibility and environmental triggers [1].The innate immune system appears to play a central role in the pathogenesis of systemic onset JIA (SoJIA) [2].Conversely,other forms of JIA are generally accepted to be driven by T cells and have classically been assumed to be linked to the increase in pro-inflammatory cytokines,such as tumor necrosis factor,interleukin-1 (IL-1),and IL-6.However,T helper 17 (Th17) cells,which secrete the pro-inflammatory cytokine IL-17 and IL-22,have recently been implicated in the pathogenesis of JIA [3].In JIA patients,these proinflammatory responses can be counterbalanced by specialized T cells known as IL-10-producing regulatory T cells(Tregs) [4].However,recent publications indicate that innate immunity is also involved in the pathogenesis of other JIA forms [5,6],not only in SoJIA [2].

The mammalian intestine is colonized by trillions of microorganisms.The colony of microbial populations(including bacteria,viruses and fungi),which inhabit on and in the host,is commonly referred to as microbiota.Intestinal microbiota initially develop immediately following birth and follow a consecutive process.Thanks to highthroughput nucleotide sequencing technologies,the composition of microbiota and host—microbe interactions have been well elucidated.Complex microbial communities play an essential role in the development of the immune system,protecting the host against colonization,and over-growth and invasion of pathogens.Recently,accumulating evidence supports microbiota as a potential contributor to the development of a wide variety of microbiota-related immune diseases including inflammatory bowel disease (IBD) and rheumatoid arthritis (RA) [7,8].The well-known collaboration between our common bacteria and the mucosa is challenged by these findings.Disruption of this balance can most likely lead to illness.

Recently,published studies have demonstrated that the composition of fecal microbiota in new-onset JIA patients was altered compared to healthy controls (HC) and the mode of either delivery or feeding,and antibiotic exposure in early life was correlated to subsequent JIA development.Here we review recent publications to address the hypothesis that alteration of intestinal microbiota predisposes the development of JIA.

Microbiota and animal models of arthritis

Existing evidence has indicated that intestinal microbiota have a crucial effect on arthritis development.Interleukin-1 receptor antagonist-knockout (IL1rn−/−) mice,K/BxN mice and SKG mice can spontaneously develop arthritis in conventional facility [9— 11].

In 2008,a study in Brazil reported IL1rn−/−mice did not develop arthritis when they were housed under germfree (GF) environmental conditions [12].In contrast,colonization of these GF IL1rn−/−mice with one indigenous microbiota,Lactobacillus bifidus,led to the rapid onset of T cell-mediated arthritis through activation of Toll-like receptor 2 (TLR 2) and TLR 4.K/BxN mice are KRN T cell receptor transgenic mice on the C57BI/6xNOD genetic background [13].In 2010,an American team demonstrated that GF-housed K/BxN mice developed an attenuated arthritis,whereas mono-colonization with segmented filamentous bacteria (SFB),one particular species ofClostridia-related bacteria,was sufficient to trigger the development of Th17 cell-mediated inflammatory arthritis in K/BxN mice [14].Furthermore,arthritis was strongly inhibited in K/BxN mice,which were treated with vancomycin or ampicillin from birth.The human leucocyte antigen B27/β 2 -microglobulintransgenic (HLA-B27/β2m-Tg) rat is a leading translational model of spondyloarthritis,which is also dependent upon an intact microbiota [15].In 2014,Lin et al.[16]reported significant differences in cecal lumen/mucosa microbiota between HLA-B27/β2m-Tg and wild-type rats and identified an increase inPrevotellaspp.as well as a decrease inRikenellaeaein HLA-B27/β2m-Tg rats compared to the wild-type rats.The alteration of intestinal microbiota in HLA-B27/β2m-Tg rat was associated with HLA-B27 expression,and its metabolite,such as proprionate,attenuated the development of HLA-B27-associated inflammatory disease[16,17].SKG mice are the ones that bear mutation of ZAP-70W 163C gene on the BALB/c background.Recent studies have revealed GF-housed SKG mice do not develop arthritis,colonization of GF-housed SKG mice with microbiota,such asPrevotella coprioraltered Shaedler flora,initiated the onset of Th17 cell-dependent arthritis by a zymosan injection,which is a fungal β-glucan and can elicit the development of severe arthritis in GF-housed SKG mice [18].

Intestinal microbial dysbiosis in children with JIA

Although the role of the intestinal microbiota in JIA is poorly understood,seven studies have revealed that intestinal microbial composition is altered in patients with JIA(Table 1).Two additional studies are not discussed because of limited data.One study included one severe patient with polyarticular JIA (poly-JIA) [26].The other study included JIA patients as the control group for IBD patients.However,minimal comparisons of microbiota data between JIA group and healthy control group were extracted [27].As shown in Table 1,the abundance of Bacteroidetes/Bacteroides,Firmicutes/Faecalibaterium prausnitzii,andBifidobacteriumare most commonly altered in patients with JIA compared to pediatric HC.In addition to heterogeneities in JIA classification,differences in populations,geography approaches and the use of immunosuppressive medications among study individuals may explain some of the different results observed.

Bacteroidetes/Bacteroidesare consistently reported with an increased abundance in patients with JIA [19,20,22— 24,28].At the level of phylum,Hissink Muller et al.[22]reported that phylum Bacteriodetes was significantly decreased in eight treatment-naïve Dutch patients with rheumatoid factor-negative poly-JIA relative to that in 22 agematched HC.In this study,IS-pro,a 16S—23S interspacer region-based profiling method,was used [22].By sequencing the bacterial hypervariable V4—V5 regions of 16S ribosomal RNA (rRNA) gene,scientists in Finland observed that treatment-naïve patients (n=30) with oligo-or poly-JIA or enthesitis-related arthritis (ERA) had higher abundance of phylum Bacteroidetes in the intestine compared to 27 HC[20].At the level of genus,in the Finnish study [20],JIA patients had an increased abundance ofBacteroides.In 2014,Stoll et al.[19]published the first study of intestinal microbiota in American children with 25 ERA and 13HC.The variable region,V4 region of 16S rRNA gene,was sequenced and two distinct clusters were identified by principal coordinates analysis.Although ERA patients had a nonstatistically significant increase inBacteroides,cluster analysis revealed one cluster was enriched inBacteroidesand the other inAkkermansia muciniphila[19].In an Indian study,Aggarwal et al.[23]investigated the intestinal microbiota in 33 treatment-naïve ERA patients compared to 14 agematched HC using sequencing of V3 region of 16S rRNA gene.They found that ERA patients had more abundantBacteroides,accompanied byPrevotellanceae.At the level of species,Hissink Muller et al.[22]found JIA patients had significant increased abundance ofBacteroides fragilis(B.fragilis),along with significantly diminished abundance ofPrevotella multisacharivoraxandAlistipes finegoldii.In 2018,Stoll et al.[24]published a multicenter study and evaluated the intestinal microbiota by sequencing V4 region of 16S rRNA gene among 30 treatment-naïve ERA patients and 19 HC.They reported that ERA patients had increased abundance ofB.fragilis.Aggarwal et al.[23]revealed that ERA patients had higher abundance of some species belonging to genusBacteroides,namelyB.fragilis,B.plebeiusandB.eggerthiicompared to HC.

Table 1 Altered composition of intestinal microbiota involved in patients with juvenile idiopathic arthritis

A significant reduction in phylum Firmicutes was revealed in a study by Tejesvi et al.[20],but not in a study by Hissink Muller et al.[22].Di Paola et al.[21]published a study in 29 Italian children with JIA and 29 HC through amplifying hypervariable V5—V6 regions of 16S rRNA gene.Poly-JIA patients had a decreased trend of abundance ofFaecalibateriumgenus and significantly enriched in abundance ofVeillonellaceae.In addition,patients with ERA had a decreased trend of abundance of thePeptostreptococcaceaefamily and a significant increase inClostridium cluster XIVb.Both poly-JIA and ERA patients had a higher abundance ofRuminococcaceaethan HC.At the level of species,the abundance ofFaecalibaterium prausnitzii(F.prausnitzii) was statistically significantly decreased in ERA patients in the study by Stoll et al.in 2014 [19].In contrast to the previous finding,the same team [24]reported a small,albeit non-statistically significant,increase in the abundance ofF.prausnitziias a whole in 2018.However,at the level of strain,Stoll et al.[24]in 2018 suggested that ERA patients had a decreased abundance of the antiinflammatoryF.prausnitziiA2-165 strain and an increased abundance of non-regulatoryF.prausnitziiL2/6 compared to pediatric HC.In 2019,van Dijkhuizen et al.[25]conducted the largest multicenter study to date to examine intestinal microbiota using 16S rRNA-based metagenomics in treatment-naïve patients with 78 Italian and 21 Dutch JIA and 107 geographically matched HC.The authors showed that JIA patients had increased richness ofF.prausnitzii,Erysipelotrichaceae,Enterococcus,ParabactteroidesandRuminococccaceae,and had decreased reduction ofAllobaculum,Gemellaceae,Propionibacterium acnesandTuricibacter.Random forest models successfully distinguished JIA patients from HC in both Italian and Dutch cohorts.In the Dutch cohort,however,no single microbiota demonstrated statistically significant differences after controlling for multiple comparisons.Of note,statistically significant differences in age were reported between JIA patients and HC in the Italian cohort.

In addition,inconsistent reports exist regarding theBifidobacterium.Stoll et al.[19]in 2014 suggested the growing richness ofBifidobacteriumin the fecal samples of ERA patients;surprisingly,the same team identified a significantly decreased abundance of the same genus in 2018 [24].

Contributions of risk factors in early life to subsequent development of JIA

The composition of intestinal microbiota and microbial colonization can be disturbed by mode of delivery,mode of feeding,and antibiotic exposure in early life.

Mode of delivery

Mode of delivery plays a major role in the composition of intestinal microbiota in infancy and early childhood.In vaginal-delivered infants,the composition of neonate’s intestinal microbiota is defined primarily by the mother [29].Neonates of caesarean section (C-section) are deprived of contact with maternal or vaginal microbes [28].Compared with vaginal deliveries,C-section neonates are prone to a lower magnitude of intestinal strict anaerobes,such asBifidobacteriaandBacteroidesand facultative anaerobes,such asLactobacillus,are more often colonized withClostridum difficile(C.difficile) [30].The diminished microbial diversity and colonization may eventually give rise to an insufficient stimulation of the immune system,which in turn reduces Th1 immune responses (lower levels of chemokines CXCL10 and CXL11) [31].C-section infants initially harbored higher diversity of intestinal microbiota than did in vaginal-delivered infants [32].Interestingly,previous studies have reported that C-section infants are more likely to have higher risk of JIA development compared to those normally delivered [33— 35](Table 2).

Although considerable evidence has indicated a longterm effect of C-section on microbial composition of infants[32],the study by Chu et al.[43]suggested the opposite:no discernable differences in intestinal community structure or functions were observed between infants vaginally delivered or C-section regardless of age of time (including initially at birth and age of 6 weeks).Further research is needed to explore the underlying mechanism behind this phenomenon.

Mode of feeding

Breast-feeding can modify intestinal microbiota and exerts a paramount effect on intestinal microbial colonization [44].A body of convincing evidence has indicated that the intestinal microbiota of breast-fed infants was likely to be dominated byBifidobacteria,whereas the microbiota of formula-fed infants included a more diverse flora [45,46].This difference seems to remain after weaning [46];however,the number ofBifidobacteriawas comparable between both groupsin two Dutch studies [30,47].In addition,Bacteroides,Lactobacilli,C.difficileandEscherichia coliwere observed more often in exclusively formula-fed infants (aged 1 month)than in exclusively breast-fed infants [30].In another study,C.difficileandAkermansiawere more abundant in formulafed infants even at 4 months of age [48].

Table 2 Evidence for potential risk factors in early life to subsequent development of juvenile idiopathic arthritis

The potential association between breast-feeding and susceptibility to JIA has been stated previously [36— 39].One small study stated that breast-feeding imposed a protective effect on subsequent JIA development,especially for oligo-JIA [36].Another large study identified that a longer duration of breast-feeding protects against subsequent JIA development [39].However,two other studies did not reveal a strong association between preceding breast-feeding and subsequent development of JIA [37,38].

Antibiotic exposure

Antibiotic exposure is well known to produce considerable but often transient perturbations in the indigenous intestinal microbiota [49].Long-term exposure can alter the biodiversity of intestinal microbiota [50].Antibiotic exposure early in life dramatically reduced the diversity and the richness of microbiota initially following birth to several months,though these were restored within the first year [32].Exposure to antibiotics early in life resulted in not only a delayed maturation of microbiota but also reduced abundance ofBifidobacteriaandBacteroidesin infants [30,32].Even in preschool children,antibiotic exposure,especially macrolides,leads to long-lasting alterations of microbial composition with a reduction inBifidobacteriumand an increase inBacteroidetes[51].

Two recently published case—control studies have investigated the association between antibiotic exposure and subsequent JIA development [40,42].Both studies indicate that antibiotic exposure in early life was significantly associated with increased risk of subsequent JIA development in a dose-dependent manner (Table 2).However,the impact of antibiotic category or spectrum of activity coverage on JIA development varied.The United Kingdom (UK) study did not have sufficient statistical power to address the differences between anti-anaerobic and anti-aerobic antibiotics with respect to later JIA development [42].Although not initially analyzed [40],Arvonen et al.[41]further grouped the antibiotics into anti-anaerobic and anti-aerobic types in a recently published review paper in light of the same method used in the UK study.In this adjusted model,children exposed to a combination of diverse antibiotic categories were at higher risk compared to those exposed to a category of either antianaerobic or anti-aerobic antibiotics (Table 2).It is speculated that the greater the overall perturbation of microbiota,the higher the risk of later developing JIA.

We were greeted by a burly proprietor3, whose smile faded when he discovered we were foreigners with a limited command of the French language. He turned his back and retreated to the kitchen, muttering under his breath and slowly shaking his head from side to side. Not a good start.

Hypothesis of mechanism

Intestinal microbiota are sufficient to induce joint inflammation [52,53].In animal models of arthritis [52],GF conditions had no arthritis,but colonization ofBacteroidesspp.,includingB.fragilis,successfully resulted in articular inflammation [52,53].In addition,the intestinal microbiota can play not only a critical role in inducing alteration of mucosal permeability [54],but also in the adaptive immune system in autoimmune arthritis [14].Bacteroidesspp.andAkkermansia mucinipphiacan facilitate inflammatory processes by increasing the access of the bacteria to the intestinal immune system [55].Intestinal microbiota can increase mucosal permeability and internalization of microbiota in the mucosa through enterotoxin secretion and mucin degradation,and finally induce the damage of mucosal barrier and increase penetration of commensal microbiota [15,56].Consistent with these findings,increased intestinal permeability has been reported in patients with JIA [57,58].Products produced by intestinal microbiota could impact the immune system at distal sites,such as joints,without leaving the intestine [14].Moreover,intestinal microbiota can induce the differentiation of Th17 cells in the small intestine lamina propria (SILP) [59— 61]and can activate inflammation in the joints by migration of T cells into peripheral lymphoid tissue [62].

In summary,intestinal microbial dysbiosis causes dysregulation of the immune system possibly through regulating the development of T cell subsets,particularly Th17 cells and Tregs,and by increasing mucosal permeability.Combined with host genetic susceptibility and environmental triggers,intestinal microbial dysbiosis may cause autoimmunity and local inflammation in extra-intestinal sites,such as joints (Fig.1).

Microbiota-mediated host immune system programming

Intestinal microbes can exert a strong impact on the appearance and/or maintenance of special CD4+T cell subsets[59— 61].Intestinal microbiota interact with the host immune system to direct against and to shape the differentiation of both effector and regulatory T cell subsets in the intestine(Fig.1).In all conditions,dendritic cells (DC) are deemed to be the initial target of mediators generated either by culprit intestinal microbes or in response to them,such as serum amyloid A (SAA),adenosine-5´-triphosphate (ATP),polysaccharide A (PSA) and short-chain fatty acids (SCFAs).However,the precise mechanism by which specific intestinal microbiota induce the development of Th17 cells and Tregs and maintain mucosal homeostasis remains to be elucidated.

T helper 17 cells

Th17 cells,a lineage of CD4+T cells,play a significant role in the development of autoimmune disease by producing cytokines,such as IL-17 and IL-22 [63].The presence of IL-22 is required for the direct induction of regenerating islet-derived protein 3 gamma (RegIIIγ),a secreted antimicrobial lectin produced by colonic epithelial cells[64].RegIIIγ favors host—bacterial mutualism by regulating the spatial relationships between intestinal microbiota and host [65].Th17 cells are considerably abundant in the SILP,where their accumulation must require specific luminal commensal microbiota [59,60].The differentiation of Th17 cells in the SILP is dramatically inhibited in selective antibiotic-treated or GF mice [12,14,60],whereas colonization of SFB facilitates the appearance of Th17 cells in the SILP [59].Therefore,the development of Th17 cells is hypothesized to be regulated by intestineintrinsic mechanisms.The tight adherence of SFB to the host epithelium in the terminal ileum of mice upregulates the levels of SAA,which in turn acts on lamina propria dendritic cells (LPDCs) to facilitate Th17 cell differentiation,possibly through the local cytokine milieu,including IL-6 and IL-23 [59,66].However,the exact mechanism by which SFB facilitates Th17 cell development is as yet not fully known.Thus far,no study has investigated how SFB is involved in the pathogenesis of JIA.

Fig.1 Model of the intestinal microbiota-mediated development of T cell subsets and the potential role of intestinal microbiota involved in development of JIA.Intestinal microbial dysbiosis causes dysregulation of immune system possibly through regulating the development of Th17 cells in a SAA-or ATP-dependent manner and development of Tregs in PSA-,SCFAs or folate-dependent manner,and increasing mucosal permeability.Combined with host genetic susceptibility and environmental triggers,intestinal microbial dysbiosis may cause autoimmunity and local inflammation in extra-intestinal sites,such as joints.LPDCs lamina propria dendritic cells,SAA serum amyloid A protein,ATP adenosine-5´-triphosphate,IL interleukin,Th17 T helper 17,RegIIIγ regenerating islet-derived protein 3 gamma,FOXP3 +Forkhead box P3,Tregs regulatory T cells,PSA polysaccharide,SCFAs short-chain fatty acids,TGFβ transforming growth factor-beta,PTPN22 protein tyrosine phosphatase,non-receptor type,22,MIF macrophage migration inhibitory factor,SLC solute carrier,WISP3 WNT1-inducible-signaling pathway protein 3

Luminal ATP,which can be derived from commensal bacteria,promotes Th17 cell differentiation in mice by activating a unique subset of LPDCs,CD70 high CD11c low,which expresses Th17-prone cytokines,such as IL-23p19[67].In line with this observation,GF-mice exhibit considerable levels of ATP and a fewer Th17 cells in the intestinal LP [67].Of note,the microbiota that regulate the development of Th17 cells in the intestine might be in a species-specific pattern.It is supported by the fact that colonization of GF-housed mice with microbiota in healthy mice,but not in rats or humans,restores the development of Th17 cells in the intestine [68].

Regulatory T cells

Tregs were originally identified as CD4+CD25+Forkhead box P3 (FOXP3+).Tregs exert inhibitory control of immune responses by producing IL-10 and transforming growth factor-beta [66,69].It is speculated that the development of Tregs might be dependent upon the intestinal microbiota,supported by the evidence that the number of Tregs is dramatically reduced in GF mice compared to conventionally housed mice [70,71].Accumulating studies have suggested that increased numbers and the immunoregulatory function of Tregs are induced by the presence of several components of indigenous intestinal microbiota(such asClostriumspecies clusters IV and XIVa orB.fragilis)[70,72,73].Tregs are mainly induced from naïve CD4+T cells through PSA [73,74]or SCFA [72,75,76]produced by intestinal microbiota.

PSA is produced and exported byB.fragilis,a Gramnegative symbiont predominately colonized in the outer mucus layer of the colon.Its special zwitterionic structure is crucial to the growth and efficient colonization ofB.fragilisas well as mediating its interactions with host [77].TLRs are innate immune receptors recognizing conserved structure of bacteria.B.fragilis-derived PSA has been implicated not only to activate TLR2 on DCs,but also to be processed and presented to T cells by DCs [78,79].PSA can inhibit inflammation either by inducing IL-10 production or by facilitating the number and function of IL-10-producing FOXP3+Tregs [73,74].

SCFAs are intestinal microbiota-derived bacterial fermentation products and have anti-inflammatory properties by binding to free fatty acid receptors or by promoting epigenetic alternations in the host [80].SCFAs are inhibitors of histone deacetylases (HDAC) and ligands for G protein-coupled receptor (GPR) [77].SCFAs promote anti-inflammatory effect in colonic DCs and macrophages and enable them to induce differentiation of Tregs and IL-10-producing T cells in a GPR43-dependent manner or in a GPR109A (encoded byNiacr1)-dependent manner [76,81].SCFAs also regulate the number and function of peripheral Tregs through HDAC inhibition.Numerous studies have previously stated that SCFAs regulate the size and function of colonic FOXP3+Tregs pool in an HDAC-dependent manner to maintain colonic homeostasis [72,82].Inhibition of HDAC9 elevates the number of Tregs,fortifies the immunosuppressive function of FOXP3+Tregs under homeostatic conditions,and alleviates the development of Tregs-mediated colitis[83].Alternatively,the increased numbers of Tregs and enhanced immunoregulatory functions of these cells were observed through manipulation of SCFAs treatment [72,76].

Critical time window early in life

Future perspective

Targeting intestinal microbiota by means of fecal microbial transplantation (FMT) and prebiotics/probiotics has recently started to emerge as a novel therapeutic option in the context of rheumatic diseases.Probiotics are defined as follows:“Live microorganisms which when administered in adequate amounts confer a health benefit on the host”[86].FMT,which presumably restores a normal intestinal microbiota and regulates the immune homeostasis of a recipient,has been used successfully in adult patients with IBD and has been proven to work better in younger individuals [87,88].Although the protocol for a clinical trial of FMT treatment in adult psoriatic arthritis has been published [89],FMT has not been evaluated in patients with arthritis.

Significant concerns regarding the safety of FMT remain.Impaired intestinal functions (e.g.,faster gastrointestinal transit,intestinal barrier dysfunction,innate immune activation and anxiety-like behavior) [90]and serious infections caused by multi-drug-resistant microbiota regarding FMT have been reported [89].In addition,currently the standardization of fecal FMT protocols for various diseases regarding donor selection (including age,gender,and health status),preparation (fresh or frozen),and dosage regimen(single or multiple doses) has not reached an agreement [91],which makes it difficult to predict the efficacy of FMT.This evidence implies that the manipulation of FMT as a therapeutic strategy requires cautions,and intensive research is needed to evaluate its safety.

A growing number of clinical trials for testing prebiotics/probiotics have shown favorable tolerability and potential benefits for restoring healthy microbiota in infants [92,93].However,a recent randomized controlled trial of probiotics suggested that probiotic treatment in ERA patients was well tolerated but failed to indicate any significant immunological or clinical effects compared to treatment by nonsteroidal anti-inflammatory drugs [94].Again,a recent systematic review has revealed that probiotic supplementation does not alter the fecal microbial composition [95],with the exception of one study in which increased beta diversity was noticed in the probiotic group compared to the placebo[96].Thus,we need more evidence to evaluate the safety and effectiveness of prebiotics and probiotics for patients with JIA before we can make any kind of recommendations to clinical practice.

Conclusions

In summary,specific microbial factors may contribute to the pathogenesis of JIA.This effect may have been initiated at the time of postnatal intestinal colonization and may reflect the influence of antibiotic regimens.Challenges that still remain to be overcome include the causality identification between intestinal microbial dysbiosis/antibiotic regimens and JIA development.Better translational studies are needed to investigate the mechanisms behind these epidemiologic relationships and to design the best interventional administrations to restore balanced intestinal microbial communities.

Author contributionsXL wrote the initial draft of the manuscript.HF,LS and ZZX revised critically and contributed to the writing of the manuscript.MXL designed the review,revised critically and contributed to the writing of the manuscript.All authors have seen and approved the final version of the manuscript.

FundingThis work was financially supported by National Natural Science Foundation of China (No.81701591) and Research Foundation of Capital Institute of Pediatrics,China (No.FX-2019-01).

Compliance with ethical standards

Ethical approvalNot needed.

Conflict of interestThe authors have no conflict of interest to declare with regards to the work presented.