Association of Putative Virulence Genes with HEp-2 Cell Adherence and Biofilm Production in Enteroaggregative E. coli Strains Isolated from Acute Diarrheal and Healthy Children, India

Abstract

Enteroaggregative Escherichia coli (EAEC) is an emerging enteric pathogen that causes acute and chronic diarrhea in developed and industrialized countries in children. EAEC colonizes the human intestine and this ability to form colonies and biofilm is an important step in pathogenesis. Here, we investigated the relationship between known or putative 22 EAEC virulence genes and biofilm formation in isolates derived from acute diarrhea and healthy children and their aggregative adherence (AA) pattern with Hep-2 cell lines. A total of 138 EAEC isolates were recovered from 1210 stool samples from children (age < 10 years) suffering from acute diarrhea and 33 EAEC strains isolated from 550 healthy children (control group) of different Anganwadi centers in Chandigarh region were included. Polymerase chain reaction using the primer pair pCVD432 identified E. coli isolates as EAEC. A total of 22 virulence-related genes have been identified using M-PCR chain reactions. The crystal violet method was used for the quantitative biofilm assay. Aggregative adherence assay was also studied using HEp-2 cell lines. Of 138 EAEC isolates from the acute diarrheal group, 121 (87.6%) EAEC isolates produced biofilm. In our findings, typical EAEC (62%) isolates were strong biofilm producers (37.5%) in the diarrheal group. Among adhesive variants, agg4A (39.6%) and aggA (21.6%) were the most common and were statistically significant (p = 0.01 and p = 0.03 respectively). We reported that the aggR gene along with the typical AA pattern was present in 71.4% of the EAEC strains in the diarrheal group, whereas it was present in 44% of the control group. Other aggR non-dependent genes like ORF3 and eilA may also lead to biofilm formation. In conclusion, there is significant heterogeneity in putative virulence genes of EAEC isolates from children and biofilm formation is associated with the combination of many genes.

Introduction

One of the most frequent causes of morbidity and mortality among infants and children in developing nations is bacterial diarrhea [1]. Children in low-income countries have been affected by enteroaggregative E. coli (EAEC), which has been identified as a frequent cause of acute and chronic diarrhea as well as a cause of their growth retardation [2]. Among six pathotypes of Diarrheagenic E. coli (DEC), EAEC is an emerging food-borne pathotype that can affect international travelers, people with HIV infection, and children who have encountered acute or persistent diarrhea. EAEC has been connected to diarrhea in both developing and developed nations, including the United States, and occasionally leads to significant outbreaks of gastrointestinal illness [3–5]. Notably, a major outbreak in Germany in 2011 due to a hybrid O104:H4 EAEC strain that contained the Shiga toxin gene from EHEC resulted in over 4,300 cases of diarrhea, 900 hospitalizations, and 50 fatalities [6]. The proportion of EAEC attributable to moderate and severe diarrhea in Bangladesh among 12- to 23-month-olds was 9.9% as part of the Global Enteric Multicenter Study (GEMS), a prospective case–control study of 0- to 59-month-old children living in Africa and Asia who experienced moderate to severe diarrhea [7].

EAEC is an exceptional colonizer of the human intestine and this ability to colonies and form a biofilm is an important step in pathogenesis (Fig. 1) [8]. EAEC strains adhere to epithelial cells in a characteristic “stacked-brick” or aggregate formation.EAEC possesses a large and diverse repertoire of adhesins [9]. The aggregative adherence fimbriae (AAF), which are specific to the EAEC, come in at least five different varieties [10]. The aggR gene is a master regulator controlling the transcription of other virulence factors and bacterial adhesin which promotes EAEC colonization [11, 12]. EAEC isolates are categorized into typical (aggR positive) and atypical (aggR negative) isolates based on the presence or absence of aggR gene. The production of the distinctive EAEC stacked-brick adherence pattern and optimal colonization depends on all adhesive factors. Moreover, Serine Protease Autotransporters of Enterobacteriaceae (SPATEs) and an anti-adherence protein (Aap; also known as dispersin) is essential for the formation of biofilm [5, 13]. There are two classes of SPATEs. Among them, pet, sigA, and sat are all cytotoxic to epithelial cells and are classified as Class I SPATEs [14]. Class II SPATEs (sepA, pic) have a wider range of phenotypic traits, though some are known to cleave mucin [13, 15]. Pic, a mucinase encoded on the bacterial chromosome, is present in numerous EAEC and Shigella strains [16]. The Aat secretion system exports the dispersin protein Aap from the cell, where it forms a loosely networked coat on the cell surface [17]. After adhesion, proliferation, and colonization on a surface, the bacteria enclose themselves in exopolymeric substances and bring in more cells to establish microcolonies that are studded with channels filled with fluid [18–21]. Colonies in the biofilm are challenging for bactericidal antibiotics to eradicate because of the restricted antimicrobial penetration, slowed growth rate, and expression of potential resistance genes. Additionally, they may be protected from attack by the intestinal immune system, resulting in long-term infections [19].

Fig. 1.

The current model of EAEC pathogenesis. (Created by using Biorender.com)

In India, EAEC remains the most prevalent pathotype isolated from diarrheal stools in children below the age of 5 years. EAEC was the commonest DEC causing acute community-acquired diarrhea, isolated from 11.4% of children aged 0–10 years and was isolated in 6% of children without diarrhea in a large surveillance study across North India [22]. Here, we have investigated the biofilm-forming potential and aggregative adherence (AA) pattern of EAEC strains isolated from children with acute diarrhea and healthy children and compare their relationship with 22 putative virulence genes.

Material and Methods

Ethics Statement

The Postgraduate Institute of Medical Education and Research (PGIMER) Ethics Committee approved the study (INT/IEC/2017/173). The patient's parents or guardian gave written informed consent.

Selection of EAEC Strains

A total of 138 EAEC isolates were recovered from 1210 stool samples from children (age < 10 years) suffering from acute diarrhea sourced from an earlier study conducted at PGIMER and its referral labs in various regions across North India during the period from 2015 to 2017. Diarrhea was defined as the passage of three or more liquid or semi-liquid stools. Thirteen labs took part in the study from Chandigarh, Haryana, Punjab, Uttarakhand, and Himachal Pradesh. A total of 33 EAEC strains (control group) isolated from 550 healthy children (age < 5 years) of different Anganwadi centers in the Chandigarh region were included in this study from our previous published study [22]. Stool samples from these children were collected in sterile containers, transferred to Cary Blair media, and transported to the laboratory in the cold chain. Briefly, samples were inoculated onto MacConkey agar, ampicillin blood agar, xylose lysine deoxycholate agar (XLD agar), thiosulfate-citrate-bile salts-sucrose agar (TCBS agar), alkaline peptone water(APW) and selenite F broth and incubated at 37°C for 18–24 h.

Organisms were identified by standard biochemical, and Matrix-Assisted Laser Desorption/Ionization-Time Of Flight (MALDI-TOF), which was performed on a MALDI Microflex LT mass spectrometer(BrukerDaltonik GmbH, Bremen, Germany), and confirmed by serotyping using antisera from Denka-Seiken (Japan). Upto three confirmed E. coli, colonies, were selected from each MacConkey agar plate and streaked onto fresh, sterilized nutrient agarand were stored in trypticase soy broth containing 15% glycerol at −80 °C. Further, these E. coli isolates were subjected to M-PCR. We used multiplex PCR (M-PCR) [Multiplex PCR for enterotoxins (heat-labile [LT]and heat-stable [ST]), EPEC (Eae) protein bundle forming protein (Bfp), Shiga toxins (Stx1, Stx2), VTcom for EHEC and (CVD432) for EAEC] which detected different pathotypes of DEC (Table 1). The pCVD432 primer utilized in this study for the detection of EAEC amplified the 630 bp region from the start position 65 to the end position 694 of the CVD432 gene.

Table 1.

List of the target gene, primers sequences, function and amplified PCR product size for detection of DEC pathotypes

Target gene	Primers sequence	Function	PCR product size (bp)	Primer designation	Reference
Bfp	GGAAGTCAAATTCATGGGGGTAT GGAATCAGACGCAGACTGGTAGT	Bundle forming protein	300	Bfp	[23]
Eae	TCAATGCAGTTCCGTTATCAGTT GTAAAGTCCGTTACCCCAACCTG	Inducing the effacement of microvilli and forming of actin pedestals	482	Eae	[23]
Elt	ACGGCGTTACTATCCTCTC TGGTCTCGGTCAGATATGTG	Heat labile toxin	273	LT	[24]
CVD432	CTGGCGAAAGACTGTATCAT AATGTATAGAAATCCGCTGTT	Plasmid gene	630	pCVD432	[25]
estA1	TCTTTCCCCTCTTTTAGTCAG ACAGGCAGGATTACAACAAAG	Heat stable toxin	166	STp	[24]
estA2-4	TTCACCTTTCCCTCAGGATG CTATTCATGCTTTCAGGACCA	Heat stable toxin	120	STh	[24]
stx1	CAGTTAATGTGGTGGCGAAGG CACCAGACAATGTAACCGCTG	Shiga toxin-1	348	Stx1	[23]
stx2	ATCCTATTCCCGGGAGTTTACG GCGTCATCGTATACACAGGAGC	Shiga toxin-2	584	Stx2	[23]
stx1 + stx2	GAGCGAAATAATTTATATGTG TGATGATGGCAATTCAGTAT	Shiga toxin	518	VTcom	[26]

Quantitative Biofilm Assay

A quantitative biofilm assay was conducted by the procedure of the crystal violet method., as described by Naoko et al. [27]. To assess biofilm formation, we inoculated 200 μl of 0.45% glucose-rich Muller-Hinton broth (HiMedia) in 96 well-bottom microtiter polystyrene plates with 5 μl of an overnight Luria broth culture grown at 37 °C with shaking. The sample was incubated at 37°C overnight (18 h). After incubation, the supernatant was discarded, and each well was washed with phosphate buffer saline (PBS) and stained with 200 µl of 0.5% crystal violet. After staining, the plate was washed gently to remove the loosely adhered cells and extra stain, and crystal violet was dissolved by adding ethanol, and OD was taken at 570 nm in an ELISA reader. EAEC Isolates having OD value > 1 were interpreted as strong biofilm producers. Bacteria with OD value > 0.5 < 1 were considered weak biofilm producers, and isolates with OD value < 0.4 were noted as non-biofilm producers.

Detection of Virulence Factors by PCR

Further investigation into the 22 virulence genes (virulence genes, including aap, pet, sigA, pic, sepA, sat, aaiC, agg4A, aafA, astA, aggR, sat, ORF3, aggA, agg3A, aafC, aar, eilA, capU, air, espY2, and rmoA) of EAEC was carried out via M-PCR [28]. Table 2 describes the virulence genes, target primer sequences, concentrations, annealing temperatures, and PCR product sizes for various virulence genes.

Table 2.

Primers used for the 4 multiplex polymerase chain reactions (M-PCRs) and 3 monoplex PCRs, target gene description, base-pair size, annealing temperature, and primers concentration

Multiplex PCR	Gene/Target	Description of Target	Primer Sequence (5’–3’)	PCR Product, bp	Annealing Temperature Primer Concentration (_C), pmol/lL	GenBank Accession No	Refrences
M-PCR-1	astA	EAST-1 heat-stable toxin	ATGCCATCAACACAGTAT GCGAGTGACGGCTTTGTAGT	110	58/20	L11241	[29]
	pet	Plasmid-encoded toxin	GGCACAGAATAAAGGGGTGTTT CCTCTTGTTTCCACGACATAC	302	58/25	AF056581	[30]
	sigA	IgA protease-like homolog	CCGACTTCTCACTTTCTCCCG CCATCCAGCTGCATAGTGTTTG	430	58/30	NC_004337	[31]
	pic	Serine protease precursor	ACTGGATCTTAAGGCTCAGGAT GACTTAATGTCACTGTTCAGCG	572	58/25	AF097644	[30]
	sepA	Shigella extracellular protease	GCAGTGGAAATATGATGCGGC TTGTTCAGATCGGAGAAGAACG	794	58/25	Z48219	[30]
	sat	Secreted autotransporter toxin	TCAGAAGCTCAGCGAATCATTG CCATTATCACCAGTAAAACGCACC	932	58/25	AE014075	[31]
M-PCR-2	ORF3	Cryptic protein	CAGCAACCATCGCATTTCTA CGCATCTTTCAATACCTCCA	121	57/35	AB261016.2	[31]
	aap	Dispersin, protein	GGACCCGTCCCAATGTATAA CCATTCGGTTAGAGCACGAT	250	57/25	Z32523	[31]
	aaiC	AaiC, secreted protein	TGGTGACTACTTTGATGGACATTGT GACACTCTCTTCTGGGGTAAACGA	313	57/25	AB255435.1	[31]
	aggR	Transcriptional activator	GCAATCAGATTAARCAGCGATACA CATTCTTGATTGCATAAGGATCTGG	426	57/25	Z18751	[31]
M-PCR-3	agg4A	AAF/IV fimbrial subunit	TGAGTTGTGGGGCTAYCTGGA CACCATAAGCCGCCAAATAAGC	169	57/25	EU637023	[31]
	aggA	AAF/I fimbrial subunit	TCTATCTRGGGGGGCTAACGCT ACCTGTTCCCCATAACCAGACC	220	57/25	Y18149	[31]
	aafA	AAF/II fimbrial subunit	CTACTTTATTATCAAGTGGAGCCGCTA GGAGAGGCCAGAGTGAATCCTG	289	57/25	AY344586	[31]
	agg3A	AAF/III fimbrial subunit	CCAGTTATTACAGGGTAACAAGGGAA TTGGTCTGGAATAACAACTTGAACG	370	57/25	AF411067	[31]
	aafC	Usher, AAF/II assembly unit	ACAGCCTGCGGTCAAAAGC GCTTACGGGTACGAGTTTTACGG	491	57/25	AF114828	[31]
M-PCR-4	ORF61	Plasmid-encoded hemolysin	AGCTCTGGAAACTGGCCTCT AACCGTCCTGATTTCTGCTT	108	57/25	J02459.1	[31]
	eilA	Salmonella HilA homolog	AGGTCTGGAGCGCGAGTGTT GTAAAACGGTATCCACGACC	248	57/25	CP009685.1	[31]
	capU	Hexosyltransferase homolog	CAGGCTGTTGCTCAAATGAA GTTCGACATCCTTCCTGCTC	395	57/25	AF134403	[31]
	air	Enteroaggregativeimmunoglobulin repeat protein	TTATCCTGGTCTGTCTCAAT GGTTAAATCGCTGGTTTCTT	600	57/25	CP009685.1	[31]
Monoplex PCR	espY2	Non-LEE-encoded type III secreted effector	CGCAAAAGATCCGGAAAATA TCAGCATTGCTCAGGTCAAC	216	59/25	ECSP_0073	[31]
Monoplex PCR	rmoA	Putative hemolysin expression- modulating protein	TTACCTTACATATTTCCATATC CGAAAACAAAACAGGAATGG	210	60/25	ECUMN_0072	[32]
MonoplexPCR	shiA	shiA-like inflammation suppressor	CAGAATGCCCCGCGTAAGGC CACTGAAGGCTCGCTCATGATCGCCG	292	57/25	ECB_03517	[32]

HEp-2 Cell Adherence Assay

All EAEC isolates were subjected to HEp-2 adherence tests [23]. HEp-2 cell monolayers (50–70% confluence) were grown on circular 13-mm glass coverslips. Twenty microliters of bacterial cultures (2X106 bacteria) were injected into each well, and the plates were incubated for 3 h in a humid, 5% CO₂ atmosphere. From the monolayers, the culture medium was aspirated. The monolayers were then fixed in 70% aqueous methanol for five minutes, stained with 10% Giemsa stain, and seen under a light microscope.

Statistical Analysis

A two-tailed chi-square test was used to compare groups. If low predicted values constrained the study, Fisher's exact test was used when the comparison between two groups was performed. Odds ratio (OR) and 95% confidence intervals (CIs) were calculated using the GraphPad PRISM program.

Results

Biofilm Formation by Clinical and Non-Clinical Isolates EAEC Isolates

Based on quantitative biofilm assay, all EAEC strains were categorized into three major categories, such as strong biofilm producers, weak biofilm producers, and non-biofilm producers. Most of the EAEC strains isolated from healthy and acute diarrheal children were able to form a biofilm; however, there was no statistically significant difference in biofilm formation among EAEC in both groups as shown in Table 3.

Table 3.

Biofilm forming potential of EAEC isolates from clinical and non-clinical sources

	N (%)
	Strong biofilm formation	Weak biofilm formation	Non-biofilm formation	Biofilm formation	P value
Acute diarrhea (n = 138)	40 (28.9)	81 (58.6)	17 (12.3)	121 (87.6)	0.8
Healthy children (n = 33)	14 (42)	15 (45)	4 (12)	29 (87.8)	1.0
Total(n = 171)	54	96	21	150

Statistically significant (P < 0.05). When biofilm-forming EAEC was compared with non-biofilm forming, EAEC from diarrheal and healthy group. Data was analyzed using Fischer’s exact test

Association Between the Presence of Putative EAEC Virulence Genes and Biofilm Formation in Isolates from Acute Diarrhea

We observed an association between the presence of aggR and biofilm formation. Of the 121 EAEC isolates from the acute diarrheal group, 77 (57.8%) of 121 biofilm producers contained aggR versus only 5 (29.4%) of 17 of the non-biofilm producers as shown in Table 4. Among these isolates, aggR was detected in 23 (62%) of 40 of the strong biofilm producers and 47 (61%) of the weak biofilm producers. Toxin gene astA was the most prevalent and was present in 88.4% of biofilm-forming EAEC isolates whereas dispersion gene aap was detected in 70% of EAEC isolates and was statistically significant (P = 0.02). Among adhesive variants, agg4A (39.6%) and aggA (21.6%) were the most common and were statistically significant (p = 0.01 and p = 0.03 respectively). However, we found that the presence of additional putative virulence factors, such as astA, pet, pic, sigA, and sat did not significantly impact the biofilm phenotype as shown in Table 4.

Table 4.

Biofilm forming potential in EAEC isolated from children with acute diarrhea

Virulence genes	Strong biofilm formation n = 40 (%)	Weak biofilm formation n = 81 (%)	Non biofilm formation n = 17 (%)	All BF + n = 121 (%)	P = value
astA	34 (85%)	73 (90.1%)	14 (82.3%)	107 (88.4%)	0.4
sigA	1 (2.5%)	3 (3.7%)	0 (0)	4 (3.3%)	1.0
pic	5 (12.5%)	11 (13.5%)	2 (11.7%)	16 (13.2%)	1.0
sepA	3 (7.5%)	10 (12.3%)	1 (5.8%)	13 (10.7%)	1.0
sat	4 (10%)	14 (17.2%)	2 (11.7%)	18 (14.8%)	1.0
pet	5 (12.5%)	4 (4.9%)	1 (5.8%)	9 (7.4%)	1.0
ORF3	29 (72.5%)	60 (74%)	10 (58.8%)	89 (73.5%)	0.2
aap	33 (82.5)	52 (64%)	7 (41%)	85 (70%)	0.02*
aaiC	8 (20%)	12 (14.8%)	2 (11.7%)	20 (16.5%)	1.0
aggR	25 (62%)	50 (61%)	5 (29.4%)	75 (61.9%)	0.01*
agg4A	18 (45%)	30 (37%)	12 (70.5%)	48 (39.6%)	0.01*
aggA	12 (30%)	14 (17.2%)	8 (47%)	26 (21.4%)	0.03*
aafA	3 (7.5%)	2 (2.4%)	0 (0)	5 (4.1%)	1.0
agg3A	3 (7.5%)	4 (4.9%)	0 (0)	7 (5.7%)	0.5
aafC	4 (10%)	5 (6.1%)	2 (11.7%)	9 (7.4%)	0.6
aar	18 (45%)	33 (40%)	13 (76.4%)	51 (42%)	0.5
eilA	18 (45%)	35 (43.2%)	4 (23.5%)	53 (43.8%)	0.1
capU	25 (62.5%)	38 (46.9%)	10 (58.8%)	63 (52%)	0.7
espY	15 (37.5%)	28 (34.5%)	7 (41.1%)	43 (35.5%)	0.7
rmoA	20 (50%)	38 (46.9%)	4 (23.5%)	58 (47.9%)	0.07
shiA	11 (27.5%)	15 (18.5%)	4 (23.5%)	26 (21.4%)	0.7
air	11 (27.5%)	14 (17.2%)	3 (17.6%)	25 (20.6%)	1.0

*Statistically significant (P < 0.05). When biofilm-forming EAEC from diarrheal group were compared with non-biofilm forming EAEC isolates. Data was analyzed using Fischer’s exact test

Association Between the Presence of Putative EAEC Virulence Genes and Biofilm Formation in EAEC Isolates from Healthy Children

In EAEC isolates from healthy children, toxin gene astA (93%) was the most prevalent among biofilm-forming EAEC isolates and was statistically significant (p = 0.0004). The presence of other putative virulence factors, such as aap, astA, pet, pic, sigA, and Sat was not significantly correlated with the biofilm phenotype, and the number of virulence genes was lower in the healthy group than in the acute diarrheal group (Table 5).

Table 5.

Biofilm forming potential in EAEC isolates from healthy children

Virulence genes	Strong biofilm formation n = 14 (%)	Weak biofilm formation n = 15 (%)	Non-biofilm formation n = 4 (%)	All BF + n = 29 (%)	P value
astA	13 (92.8)	14 (93.3)	3 (75)	27 (93)	0.0004*
sigA	1 (7.1)	1 (6.6)	0 (0)	2 (6.8)	1.0
pic	2 (14.2)	1 (6.6)	0 (0)	3 (10.3)	1.0
sepA	1 (7.1)	2 (13.3)	0 (0)	3 (10.3)	1.0
sat	1 (7.1)	1 (6.6)	1 (25)	2 (6.8)	0.3
pet	5 (35.7)	3 (20)	2 (50)	8 (27.5)	0.56
ORF3	9 (64.2)	5 (33.3)	4 (100)	14 (48.2)	0.1
aap	7 (50)	7 (46.6)	4 (100)	14 (48.2)	0.1
aaiC	3 (21.4)	4 (26.6)	1 (25)	7 (24.1)	1.0
aggR	6 (42.8)	3 (20)	2 (50)	9 (31)	0.5
agg4A	2 (14.2)	1 (6.6)	0 (0)	3 (10.3)	1.00
aggA	1 (7.1)	2 (13.3)	1 (25)	3 (10.3)	0.4
aafA	1 (7.1)	1 (6.6)	0 (0)	2 (6.8)	1.0
agg3A	0 (0)	1 (6.6)	0 (0)	1 (3.4)	1.0
aafC	0 (0)	1 (6.6)	0 (0)	1 (3.4)	1.0
aar	9 (64.2)	5 (33.3)	4 (100)	13 (44.8)	0.1
eilA	7 (50)	5 (33.3)	3 (75)	12 (41.3)	0.3
capU	6 (42.8)	6(40)	3 (75)	12 (41.3)	0.3
espY	2 (14.2)	1 (6.6)	1 (25)	3 (10.3)	0.4
rmoA	5 (35.7)	6 (40)	4 (100)	11 (37.9)	0.03*
shiA	4 (28.5)	3 (20)	2 (50)	7 (24.1)	0.2
air	2 (14.2)	1 (6.6)	1 (25)	3 (10.3)	0.4

*Statistically significant (P < 0.05). When biofilm-forming EAEC were compared with non-biofilm forming, EAEC from healthy sources. Data was analyzed using Fischer’s exact test

Based on virulence gene analysis in isolates from acute diarrheal and control samples, multiple unique combinations of virulence markers were found among the EAEC isolates in our study(Table 6). Gene combination astA, aap, aggR, agg4A, and aggA was most prominent among cases (23.1%) whereas, in controls gene combination astA, aar was most common (24%)(Table 6). However, the aggR gene was present in combination with other virulence genes in strong biofilm-producing EAEC strains in the acute diarrheal group.

Table 6.

Combination of virulence markers associated with forming potential among diarrheal and control EAEC strains

EAEC (No. of strains)	Gene combination	No. (%) of strains
Biofilm forming EAEC isolates from diarrheal group (121)	astA, aap, aggR, agg4A, aggA	28 (23.1)
	aap, aggR	4 (3.3)
	astA, aggR, agg4A, aggA	5 (4.1)
	astA, aap, aggR	6 (4.9)
	astA, aap, aggR, agg4A	12 (9.9)
	astA, aap, aggR, aggA	1 (0.8)
	astA,,aggR, aggA	6 (4.9)
	astA, aap, agg4A,aggR	11 (9.0)
	aap, aggR	2 (1.6)
	astA	15 (12.3)
	astA, aap	4 (3.3)
	none	4 (3.3)
Biofilm forming EAEC isolates from Control group (29)	astA, aap, aggR, agg4A, aggA, aar	2 (6.8)
	astA, aap, agg4A, aar	1 (3.4)
	astA, aggR, aar	1 (3.4)
	astA, aap, aar, aggA	2 (6.8)
	astA, aap, aggR, agg4A, aar	3 (10.3)
	astA, aap, aggR, aggA	1 (3.03)
	astA, aar, aap	2 (6.8)
	astA, aap, aggR, aggA	2 (6.8)
	astA, aar	7 (24)
	astA, aap	5 (17)
	aar	2 (6.8)
	astA	2 (6.8)

Hep-2 Adherence Assay

Three different AA types—type A, the typical AA pattern with a honeycomb formation; type B, an AA pattern without the typical honeycomb formation; and type C, an AA pattern with lined up cells—were seen in the present study, based to the AA patterns of EAEC identified by the HEp-2 adhesion test. In EAEC strains, the occurrence of each AA pattern is depicted in which depicts the typical AA pattern (type A) demonstrated by the majority of EAEC strains (66%) from the acute diarrheal group (Figs. 2 and 3). Type A adherence pattern was present in 71.4% of the EAEC strains in the acute diarrheal group carrying the aggR gene, whereas it was present in 44% of EAEC isolates among the control group (Table 7).

Fig. 2.

Incidence of each AA pattern in EAEC from acute diarrheal and healthy children

Fig. 3.

Aggregative adherence (AA) patterns of EAEC were detected by the HEp-2 cell adherence test. (A) A type: typical AA pattern with honeycomb formation. (B) B type: AA pattern without typical honeycomb formation. (C) C type: AA pattern with lined-up cells

Table 7.

Presence of aggregative adherence patterns in typical EAEC (tEAEC) and atypical (aEAEC) isolates among diarrheal and control group

Virulence genes	Diarrheal group				Control group				P value
	All BF + n = 121 (%)	Type A	Type B	Type C	All BF + n = 33 (%)	Type A	Type B	Type C
aggR present (tEAEC)	75(61.9%)	55 (71.4%)	15 (19.4%)	7 (9.0%)	9 (31%)	4 (44%)	2 (22%)	3 (33%)	0.0001*
aggR absent(aEAEC)	47 (38.8%)	13 (27.6%)	21 (44.6%)	13 (27.6%)	24 (72.7%)	9 (37%)	8 (33%)	7 (29%)	0.0007*

*Statistically significant (P < 0.05). When biofilm-forming tEAEC and aEAEC were compared among diarrheal and control group. Data was analyzed using Fischer’s exact test

Discussion

EAEC is an emerging enteric pathogen whose modes of pathogenicity are not truly understood largely due to its genetic heterogeneity [24]. However, the expression of biofilm has been regarded as an uncontested virulence factor among EAEC isolates. Biofilm formation is a complicated event that may encompass numerous virulence factors. EAEC strains found in the small intestine and colon mucosa produce significant amounts of biofilm and cause a change in the endothelium, consequences in prolonged term of diarrheal episodes of acute and persistent diarrhea in infants [25]. EAEC is also linked with growth impairment caused by malabsorption that, possibly, would occur as a result of thick biofilm production [25, 26]. Here, we investigated the occurrence of numerous EAEC virulence determinants and their relationship with biofilm formation among EAEC strains isolated from acute diarrheal and healthy children from India. Also, the biofilm formation ability of EAEC isolates carrying or lacking aggR was examined. Of 138 EAEC isolates from the acute diarrheal group, 121 (87.6%) EAEC isolates produced biofilm, which is higher than the rates reported elsewhere {Japan (77%) and Mexico (58%)}[29, 30]. We found that 28 and 58% of EAEC isolates from the acute diarrheal group were strong and weak biofilm producers respectively. In a previous study from Baltimore, it was found that 21% of EAEC isolates of human origin were strong biofilm producers, whereas 32% were weak biofilm producers[29]. The heterogeneity of EAEC strains may account for this geographic variation in biofilm formation [30]. In our observations, the portion of EAEC isolates that were biofilm producers was equivalent in acute diarrheal and control groups and may describe in part the potential of EAEC to cause colonization of the gut.

In EAEC, the master regulator aggR controls both plasmid- and chromosomally-encoding genes and helps in the activation of AAF-related genes [12, 31]. Furthermore, it appears that other genes, including aatA (the dispersion transporter) and aap (dispersion), are also under the control of aggR [32]. However, atypical EAEC, such as AggR-negative EAEC, is also recognized as an enteric pathogen in outbreaks of food-borne illness [33]. In our findings, tEAEC (62%) isolates were strong biofilm producers as compared to EAEC (37.5%) in the acute diarrheal group. These findings support observations from studies conducted in Mexico and Mongolia that also noted the involvement of the aggR gene with biofilm formation and found that the presence of aggR was statistically associated with biofilm production (P = 0.01) [29, 30]. In our findings, among healthy biofilm-forming EAEC isolates, the gene combination of astAandaar (24%) was most prevalent. As aar gene is a negative regulator of aggR, it may be responsible for lower virulence of EAEC from the control group despite biofilm formation. We reported that EAEC isolates possessing the aap, astA, agg4A, and aggA in an aggR context formed more biofilm than those with an aggR-lacking background. This implies that aggR governs other genes required for biofilm development in EAEC. We also looked into the role of dispersin and its transporter in biofilm production. Dispersin is a 10-kDa secretory protein defined by the aap gene that coats the microbial surface which promotes the distribution of EAEC on the intestinal epithelium. Dispersion protein was reported in 62.8% of acute diarrheal EAEC isolates (p-0.02) and 48% of control strains. Our findings show that the dispersin gene may play an important role in biofilm formation. Similarly, a previously published study from the United States shows that the importance of the dispersion-like genes is not definite, but they may lead to adherence and/or biofilm development in these strains. Also, biofilm formation was linked to the presence of the aatA gene that codes for a transporter protein that is a homolog of E. coli the outer membrane protein, TolC. In EAEC, this transporter protein enables the export of dispersion across the outer surface and provides the target for the extensively used molecular probe pCVD432.

Our study demonstrated the importance of adhesive variants agg4A and aggA in biofilm formation which were found in a majority (61%) of isolates. Only a minority (9.9%) expressed either aafA or agg3A. We observed that 12% of the EAEC strains were unable to develop a biofilm, which may be due to the expression of adhesive factors other than AAFs or inadequate AAF expression. We assume that EAEC adhesins are allelic and that biofilm development might be a conserved phenotype that all AAF family members share. In an earlier study, EAEC adhesins were found to be allelic(one of two or more genes that may occur alternatively at a given site on a chromosome)in nature, where biofilm formation was shared by all members of the AAF family [34]. Further investigation into adhesive factors may result in other crucial findings regarding EAEC colonization and adhesion. Two of aggR independent genes encoding ORF3 and eilA may also have a role in adherence and colonization. The eilA gene contributes to EAEC virulence by activating the type 3 secretion system (T3SS) and its effectors [35]. We reported that eilA gene was present in 47.9% of biofilm-forming EAEC isolates. A study from Baltimore shows that strains carrying this gene show more adherence to epithelial cells and form more biofilm [35].In our study, the gene encoding ORF3 was detected in 73.5% of the biofilm-forming EAEC isolates. In Mali, the ORF3 gene, which encodes a cryptic protein, was the most frequently identified virulence gene among children [36]. A study from the USA reported that the ORF3 gene was involved in biofilm formation and mutation in the ORF3 gene results in loss of adherence properties of the bacteria [37].

In this study, three different AA types—type A, the typical AA pattern with a honeycomb formation; type B, an AA pattern without the typical honeycomb formation; and type C, an AA pattern with lined-up cells—were seen in the present study, based to the AA patterns of EAEC identified by the HEp-2 adhesion test. It depicts the heterogeneous nature of EAEC isolates. The variation in adhesive fimbriae may be the reason for the variation in adhesion pattern. The majority of the AA pattern was present in the aggR-positive strains, along with a honeycomb formation and noticeably stronger biofilm formation, which may be related to transcriptional activation of the adhesins genes by the aggR gene. We reported that the aggR gene along with the typical AA pattern was present in 71.4% of the EAEC strains in the acute diarrheal group, whereas it was present in 44% of EAEC isolates among the control group. On the other hand, additional study is required on the pathogenesis and adhesion mechanisms of the aggR negative strains with an atypical AA pattern.

In summary, the present research shows that biofilm formation is a common phenomenon among EAEC isolates obtained from acute diarrheal and healthy children. However, a smaller number of EAEC isolates sourced from healthy children is the limitation of this study. The tEAEC strains are more frequent biofilm producers than atypical ones. In vitro biofilm production from EAEC isolates is attributed to aggR, and its regulated genes aap, aatA, and adhesive EAEC genes located on the plasmid, as well as on its chromosome, are linked with biofilm formation. Though aggR presence leads to higher and stronger biofilm production, it is not necessary. Other aggR non-dependent genes like ORF3 and eilA may also lead to biofilm formation. In conclusion, there is significant heterogeneity in putative virulence genes of EAEC isolates from children and biofilm formation is associated with the combination of many genes.

Author Contributions

VM: conceptualization, methodology, formal analysis, investigation, visualization, and writing—original draft. HK: formal analysis, methodology. BM: investigation, resources. NT: investigation and writing—review and editing, visualization, resources, supervision.

Funding

This research was supported by the University Grant Commission (UGC) (Sr. No. 20614305077, Ref No: 22/06/2014(i) EU-V).

Declarations

Conflict of interest

The authors have declared that no conflict of interest exists.