Molecular Study of Human Astrovirus in Egyptian Children with Acute Gastroenteritis

Abstract

Introduction: Human astrovirus (HAstV) has been increasingly identified as an important cause of acute gastroenteritis in young children. Limited information is available about the prevalence and genotype distribution of classic HAstV causing acute gastroenteritis in Egyptian children. Methods: Stool samples were collected from 100 infants and children attending the gastroenterology outpatient clinic in Mansoura University Children Hospital and suffering from acute gastroenteritis during the period extending from January 2018 to January 2019. Samples were tested for HAstV using reverse transcription PCR. Genotyping was performed using type-specific reverse transcription nested PCR. Results: Among 100 children included in this study, the detection rate of HAstV was 11% (11 patients). There was a significant difference regarding age between cases positive and negative for HAstV (p=0.005). There was a higher prevalence of HAstV in children aged one year or younger. Significant association was detected between HAstV positive cases and rural residence (p=0.002), summer season (p=0.025) and fever (p=0.017). The HAstV genotypes detected were HAstV-8 (8/11, 72.7%), HAstV-3 (2/11, 18.2%) and HAstV-2 (1/11, 9.1%). Conclusions: This study suggests that HAstV is a common pathogen causing gastroenteritis in Egyptian children especially in rural areas. The most frequent HAstV genotype in our study was HAstV-8.

Introduction

Acute gastroenteritis or infectious diarrhea is a common cause of morbidity and mortality in children below five years. The reported mortality rates in 2016 were 446,000 for children below 5 years. Most of these deaths are in developing countries [1]. Diarrhea is the second leading cause of death among under-5 children in Egypt. Diarrhea leads to loss of large amounts of fluids and essential electrolytes causing varying degrees of dehydration that may end in death. It causes death in 3,500-4,000 under-five children per year in Egypt [2]. In addition, repeated infections in children surviving after diarrheal attacks may lead to lifelong consequences such as growth retardation, impaired cognitive development, and impaired immune response to infection and vaccinations [3].

About 70% of acute gastroenteritis cases in children are caused by enteric viruses [4]. Human astroviruses (HAstVs) are one of the important causes of viral gastroenteritis. HAstVs cause about 10% of acute viral gastroenteritis in children [5].

HAstV is a 28-30 nm in diameter non-enveloped RNA virus. The genome of HAstV has three open reading frames, two of them, ORF1a and ORF1b, encode for the nonstructural proteins including protease (Pro) and RNA-dependent polymerase (RdRp) and the third, ORF2, encodes structural proteins of the capsid. The genotypes of classic HAstV are classified according to the capsid protein into eight distinct genotypes (HAstV-1–HAstV-8) [6]. The distribution of HAstV genotypes differs worldwide. HAstV genotype 1 was detected as the most prevalent genotype with occasional occurrence of HAstV-6, and HAstV-7 in acute gastroenteritis [7]. There are infections attributed to HAstV-8 in some geographical regions [8].

The transmission of HAstV is through fecal-oral route. The replication of the virus in infected children takes place in the intestinal epithelium [7].

However, there are insufficient studies about astrovirus genotypes in Egyptian children with diarrhea. Therefore, the aim of the present study was to evaluate the prevalence of astrovirus and its genotypes in children with acute diarrhea.

Methods

This cross-sectional study involved 100 children with diarrhea from a total of 2300 children attending the gastroenterology outpatient clinic of Mansoura University Children Hospital, Egypt from January 2018 until January 2019. Attending patients live in rural and urban areas of Dakahlia Governorate in the Nile Delta of Egypt. The included patients were complaining of acute gastroenteritis with diarrhea in the last two days. Diarrhea was defined as sudden onset of three or more watery or near watery stools in the previous 48 hours. Children with systemic infections complicated with diarrhea or associated with drug intake were excluded from the study. Samples were tested for bacterial and parasitic causes of diarrhea by direct microscopic examination and bacterial stool culture. Positive cases were excluded.

The study was approved by Mansoura Ethical Committee and informed consent was obtained from the parent of each child.

From each child a single stool sample was obtained in a clean container and transported to the laboratory within 30 minutes. Demographic and clinical data including child name, age, residence, fever, abdominal pain and dehydration status were recorded. The degree of dehydration was estimated according to clinical parameters [9].

Molecular study of astrovirus

Extraction of RNA from stool samples

RNA extraction was done by Trizol™ (Thermo Fisher Scientific, Carlsbad, CA, USA). The steps for RNA isolation including lysis, precipitation, washing and resuspension were performed according to the manufacturer’s instructions. Briefly, for each stool sample, RNA was extracted from 100 mg by using 1 mL Trizol™. Extracted RNA was precipitated using 0.5 mL isopropanol. The RNA pellet was resuspended and washed in 1 mL of 75% ethanol. The pellet was resuspended in 20-50 µL of RNase-free water and deep frozen at -70°C until the reverse transcription procedure.

Detection of astrovirus by reverse transcription PCR (RT-PCR)

Extracted RNA was used for reverse transcription PCR (RT-PCR) for detection of astrovirus ORF2 by the use of the primers listed in

Table 1. Primers used for RT-PCR and genotyping[11]^.

Gene	Primer sequence	Amplicon Size
ORF2	5'CAACTCAGGAAACAGGGTGT3' 5'TCAGATGCATTGTCATTGGT3'	449 bp
AST-S1	5'AACCAAGGAATGACAATGAC3'	212 bp
AST-S2	5'ACCTGCGCTGAGAAACTG3'	158 bp
AST-S3	5'CTGCTTGCATCTGGTCTTTCA3'	119 bp
AST-S4	5'TGATGATGAAGACTCTAA TAC3'	258 bp
AST-S5	5'TAGTAACTTATGATAGCC3'	388 bp
AST-S6	5'TGGCCACCCTTGTTCCTCAGA3'	427 bp
AST-S7	5'CTAGACAACAACACCCCG3'	548 bp
AST-S8	5'GGTAAGTGGTACCTGCTAACTAG3'	599 bp
3' end of the capsid	5'TCCTACTCGGCGTGGCCGC3'

. The method used was described previously by Noel et al. (1995) [10]. At first, denaturation was carried out for viral nucleic acid by incubation at 70°C for 5 minutes and then putting in ice for 2 minutes. Reverse transcription was performed with incubation for 1 hour at 42°C with reverse transcriptase enzyme by the use of SuperScript^TM One-Step RT-PCR with Platinum^® Taq kit (Invitrogen, Carlsbad, CA, USA). Then amplification process was done by 40 cycles of amplification (94°C/30 seconds, 50°C/30 seconds, 72°C/1 minute) and a final extension at 72°C/10 min. RNA from human astrovirus positive patient was used as a positive control and RNase-free water as negative control. The final PCR product (10 µL) was submitted to electrophoresis in 1.5% agar gel and detected by ethidium bromide staining.

Genotyping of astrovirus by multiplex nested/RT-PCR

Genotyping was done for all astrovirus positive samples by multiplex nested/RT-PCR according to Sakamoto et al. (2000) [11]. The used primers are listed in Table 1.

The nested PCR reactions were performed using 2.5 µL of the complementary DNA (cDNA) amplified product of the first PCR reaction positive for astrovirus and one microliter of specific primers pool added to the amplification mixture (Thermo Fisher Scientific, Carlsbad, CA, USA). Nested RT-PCR assay including 40 amplification cycles (94°C/1 min, 45°C/2 min and 72°C/3 min) was carried out, followed by a final extension of 72°C/7 min. Ten microliters of the final product were submitted to electrophoresis in 1.5% agar gel and detected by ethidium bromide staining.

For confirmation of genotypes each positive sample was tested using single primer of the detected genotype using the previous cycling conditions.

Statistical analysis

Data were analyzed by the use of SPPS 22 (IBM Corp., Armonk, NY, USA). Qualitative data was interoperated as number and percentages. Comparison was performed using Chi-square test or Fisher exact test when applicable. The difference in age between patients with positive and negative results for astrovirus was compared using Mann-Whitney test U test. Variable differences were considered significant if p value was <0.05.

Results

This study included 100 children complaining of diarrhea with age range from 3 month up to 7 years old (median 1 year), 58 males and 42 females. They were mainly from rural residence in Dakahlia Governorate (58%) with main complaint of fever (58%) and mild dehydration (49%). Fifteen patients presented with severe dehydration (15%), all these cases were hospitalized for treatment. Higher prevalence of gastroenteritis cases was in summer season (45%). Stool samples were positive for astrovirus by RT-PCR in 11 cases (11%),

Table 2. Demographic and clinical criteria of children presenting with acute gastroenteritis with positive and negative results for astrovirus during the period of the study.

		Total patients	Astrovirus positive	Astrovirus negative	P value
		(N=100), n (%)	(N=11), n (%)	(N=89), n (%)
Sex					0.519a
Male		58 (58)	5 (45.5)	53 (59.6)
Female		42 (42)	6 (54.5)	36 (40.4)
Age					0.005*
Median		1 year	6 months	1 year
(min-max)  25th percentiles  75th percentiles		3 months - 7 years 1 year 3 years	3 months - 3 years 3 months 7 months	3 months - 7 years 1 year 3 years
Season distribution
Warm seasons	Summer	45 (45)	10 (90.9)	35 (39.3)	0.025**
	Spring	29 (29)	1 (0.9)	28 (31.5)
Cold seasons	Winter	12 (12)	-	12 (7.9)
	Autumn	14 (14)	-	14 (15.7)
Residence
Rural		58 (58)	11 (100)	47 (52.8)	0.002**
Urban		42 (42)	-	42 (100)
Symptoms
Abdominal pain		38 (38)	7 (63.6)	31 (34.8)	0.753b
Fever		58 (58)	10 (90.9)	48 (35.9)	0.017**
Dehydration degree   Mild		49 (49)	5 (45.4)	44 (49.4)	0.463b
Moderate		36 (36)	3 (27.3)	33 (37.1)
Severe		15 (15)	3 (27.3)	12 (13.5)

* Significant P value by Mann Whitney U test. Value of the test: 148.5. Effect size (r): 0.15. ** Significant P value by Fisher’s exact test. ^a Non significant p value by Chi-square test. ^b Non significant p value by Fisher’s exact test.

. The genotypes of astrovirus identified by nested RT-PCR were genotype HAstV-8 (8/11. 72.7%), HAstV-3 (2/11, 18.2%) and HAstV-2 (1/11, 9.1%), Figure 1.

There was no statistically significant difference between children with negative and positive RT-PCR for astrovirus regarding sex, abdominal pain and the degree of dehydration. Out of 15 hospitalized children presenting with severe dehydration, three cases were positive for astrovirus representing (3/11; 27.3%). The positive cases for astrovirus were statistically associated with rural residence (p=0.002), summer season (p=0.025) and fever (p=0.017) - Table 2. There was a significant difference regarding age between patients with positive and negative RT-PCR for astrovirus (p=0.005). There was an increased prevalence of astrovirus in children aged one year or younger, although this increase was not significant between age groups,

Table 3. RT-PCR positive samples for astrovirus according to age groups.

		Age group (years)						Total
		≤1.00	1-2	2-3	3-4	4-5	5-7
PCR	Positive	9	1	1	0	0	0	11
	Negative	50	4	20	6	8	1	89
Total		59	5	21	6	8	1	100

p=0.516 by Fisher’s exact test.

.

Discussion

Acute gastroenteritis associated with astrovirus is a common infection in children below 5 years. The astrovirus capsid protein acts like an enterotoxin leading to dysfunction of the intestinal epithelium barrier [12]. There is a lack of international and national surveillance of the epidemiological and molecular characterization of astrovirus associated gastroenteritis in Egypt.

In the current study, astrovirus was detected in 11% of children with acute gastroenteritis. This result matches recent reports in the Republic of Congo (10.3%)[13] and India (12.5%) [14]. However, our finding is higher thanprevious results in Kenya (6.3%) [15], Northwest Ethiopia (3.6%) [8], lower than others obtained in Egypt by El Taweel et al. (2020): 28% [16], and Nigeria: 40.4% [17]. The higher prevalence obtained in Egypt may be due to the inclusion of novel human astrovirus types. In addition, variable prevalence in other countries can be explained by the difference in geographical regions and socioeconomic factors.

Astrovirus diarrhea is a common infection both in developed and developing countries, which suggests that the improvement of the hygiene and living conditions alone cannot prevent such infection [8,13,14,15,16,17,18]. Development of an astrovirus vaccine and its administration in early life may have a role in reducing this infection in early life and even decrease the incidence of astrovirus diarrhea with increasing age [19].

Regarding the clinical features associated with diarrhea in HAstV positive patients, the degrees of dehydration (mild, moderate, and severe) were detected in astrovirus positive cases with no significant association of classic HAstV with any of them (p=0.463). This matches the previous result of Naficy et al. (2000) [20]. Astrovirus was significantly associated with fever (p=0.017). Similarly, Bon et al. (1999)[21] found fever was significantly associated with astrovirus infections.

In the present study, the detection of astrovirus was in children aged from 3 months to 3 years. The infection was prevalent below or at 1 year of age (9/11) with similar results reported from Egypt by El Taweel et al. (2020) [16]. In developed countries, classic HAstV are more common in older children, e.g., in Spain the highest prevalence was in children with age range 2 to 4 years [22]. The decreased prevalence of astrovirus later in childhood may suggest the presence of homotypic immunity against subsequent astrovirus diarrhea [23]. Also, this difference in age distribution may be due to difference in social behavior in different geographic locations.

In the current study, infection with astrovirus was significantly higher in rural areas (11/11) (p=0.002) and in warm months; summer season (10/11) and spring season (1/11) (p=0.025). The distribution of classic HAstV infections in urban and rural areas differs worldwide. The mean incidence of HAstV infections worldwide is higher in rural areas (23% in rural and 7% in urban) [24]. However, higher urban distribution was recorded in some countries like Kenya [15], while in other countries like Ethiopia no difference in classic HAstV infections between urban and rural areas were seen [8]. The heterogeneity in infection and diarrhea between locations suggests that estimates can vary considerably among different geographic settings and in urban and rural areas [23]. Our result differs from previous reports that described higher incidence of classic HAstV infections in cold weather [24], spring [25], and autumn [26]. Other studies found no specific seasonal predominance [27]. HAstV infections also were recorded in summer season [22]. In agreement with our result, HAstV infections in Egypt were more frequently detected during warmer months [20].

This diversity of results could be attributed to the geographic locations and to the complex interaction between different factors including host transmission, social factors and environmental factors. The increased transmission of gastroenteritis in the summer season may be associated with contact with water by drinking and swimming under unsanitary conditions especially in rural areas.

In the current study, the most common genotype was HAstV-8 (72.7%) and HAstV-3 represents the second most prevalent one (18.2%). Similar findings were obtained in previous studies from countries such as Ethiopia [8], while different genotypes were prevalent in other countries like Malawi (HAstV-1 and HAstV-3) [28]. An earlier study in Egypt (2000) reported HAstV-1 as the most prevalent genotype, and HAstV-8 and HAstV-3 as the 3^rd most frequent genotypes [20].

The predominant genotypes vary according to the seasons and the geographical regions, which necessitates the implementation of nationwide epidemiological studies for declaration of this issue. Moreover, the difference in the diagnostic and typing methods leads to heterogeneous data with difficulty in their interpretation.

In the present study we used RT-PCR which detects only classical HAstVs. Hence, the prevalence of the novel human astrovirus types was not detected, which represents a limitation of our study. We did not investigate co-occurrence of human astrovirus with other enteric viruses causing gastroenteritis, like rotavirus and norovirus, which is another study limitation. Most of the studies that found astrovirus co-infection with other pathogen use serological detection of enteric pathogens.[14,23] So, further studies employing more genetic techniques for detection of novel astrovirus and the presence of astrovirus with other enteric pathogens among Egyptian children are recommended.

Conclusions

The present study demonstrated that astrovirus is a common pathogen associated with diarrhea in children, particularly in rural areas. The frequent genotype in the present study was HAstV-8. Further national surveys and studies are required to confirm these findings.