Interaction of Giardia intestinalis and Systemic Oxidation in Preschool Children in the Western Highlands of Guatemala

Abstract

Background: 

Guatemala is a country with the highest prevalence of stunting in under-5 children in the Americas, with a national average of 49.8%. Asymptomatic intestinal colonization with Giardia intestinalis is common in Guatemalan preschoolers and has been implicated as a factor in linear growth retardation. The potential mechanisms of any giardiasis-growth interaction have not been exhaustively explored.

Objectives: 

The aim of the present study was to describe urine oxidative stress biomarkers and erythrocyte antioxidant enzyme activity, and to explore any association with prevalence or intensity of G intestinalis infection in preschoolers attending 3 government-subsidized day care centers in the Guatemalan Western Highlands.

Methods: 

Samples of feces, urine, and red blood cell (RBC) hemolysate were collected in a total of 74 preschoolers enrolled in 3 day care centers. Giardia prevalence and a proxy index for intensity were assessed by enzyme-linked immunosorbent assay (ELISA). Urinary biomarkers of oxidative damage to DNA (8-hydroxydeoxyguanosine [8-OHdG]) and to lipid (F2t 15-Isoprostane [F2-Iso]) were measured by ELISA. The erythrocyte activity of catalase (CAT), superoxide dismutase (SOD), glutathione reductase (GSR), and glutathione peroxidase (GPX) were measured by respective spectroscopic substrate-based reaction assays.

Results: 

Median values of RBC CAT activity (P = 0.016) and urine F2-Iso (P = 0.023) differed between children who were positive (n = 39) and negative (n = 35) for Giardia. Similarly, G intestinalis intensity was significantly and positively associated with urinary F2-Iso (r = 0.446, P < 0.001), RBC SOD (r = 283, P = 0.014), and RBC CAT (r = 0.260, P = 0.025).

Conclusion: 

The optical density reading of the fecal ELISA assay for G intestinalis has potential as a proxy for the intensity of infestation. In this respect, there exists an association of this intensity with indicators of the systemic oxidation.

What Is Known

• Giardia intestinalis is a protozoon that causes giardiasis because of poor hygienic environment in low-income countries.
• Asymptomatic intestinal colonization with Giardia intestinalis is common in Guatemalan preschoolers.
• Little is known about the relation of oxidative stress and antioxidation activity with this infection.

What Is New

• Associations between red blood cell antioxidant enzymes, urine oxidative stress biomarkers, and giardiasis were found in this population.
• Oxidation was also related to the intensity of asymptomatic giardiasis.
• Oxidation may represent a potential mediating factor in growth impairment associated with the intensity of Giardia colonization in the intestine.

Growth retardation has been identified as the highest priority issue in the public health agenda for the Republic of Guatemala in Central America ⁽¹⁾. The nation ranks highest in the Western Hemisphere, with the prevalence of 49.8% in children between 6 and 50 months in the most recent national survey ⁽²⁾. Moreover, the stunting rates are the highest in the rural areas, and indigenous Guatemalans are more stunted at all ages than their nonindigenous compatriots ^(3,4). Environmental factors in addition to dietary deficiencies have been imputed as determinants of linear growth retardation ^(5,6), but the nature and mechanisms of their influence on growth are incompletely understood.

The literature is variable in terms of a negative influence of intestinal infection with Giardia intestinalis on growth. The earliest confirmation of an adverse effect actually came from Guatemala, in which children with the greatest time excreting Giardia cysts in early life were the shortest in stature ^(7,8), found relatively higher infections of G intestinalis in Guatemalan children with moderate malnutrition based on weight. In our review of the literature since 2004, the majority of studies confirmed an adverse association of Giardia infection and linear or ponderal growth ^(9–19), whereas a small minority did not detect any association ^(20,21).

With a common database for fecal assays for G intestinalis and battery of biomarkers of oxidation or the antioxidative system in urine, plasma, and erythrocytes in 74 preschool children in an environment, that is, Guatemalan day care centers, with proven predisposition to transmission of the protozoa ⁽²²⁾, we tested the hypothesis of an association of asymptomatic giardiasis and oxidative stress. We present here the findings related to this exploration of associations.

METHODS

Geographical and Ecological Setting

The Western Highlands in Quetzaltenango, located 220 km from Guatemala City at 2657m above sea level, were the setting for this research project. It is a generally agricultural region with a temperate climate. Principle crops include maize, coffee, and fruit. The economy is based on agriculture, livestock, commerce, finances, and tourism.

Three day care centers in different locations were assigned by the Secretaría de Obras Sociales de la Esposa del Presidente (SOSEP, Secretariat of Beneficial Works of the First Lady) for availability to our study. The semiurban center (center A) was located in La Esperanza, 2 miles away from downtown Quetzaltenango, with all of the basic services and ∼80% of indigenous attendees. The marginal urban setting (center B) was located in the outskirts of the city of Quetzaltenango in La Puerta del Llano, with great majority of indigenous attendees, most of whom were children of families that moved from the rural areas to run small business in markets; they were living in overcrowded conditions, some of them without basic sanitary services. The rural setting (center C) was situated 15 miles away from Quetzaltenango proper in La Estancia, San Martín Chile Verde; 100% of the children attending this center were indigenous preserving their original customs. Although the services at the distinct sites were common and equivalent; the varying ethnic mixture and environmental circumstances produced variation in some of the living habits, pastimes, and physical characteristics among attendees of the different centers.

Enrollment of Subjects

Eligible children had to be attending 1 of the selected centers and be 2 to 7 years old; they had to have ≥80% attendance during the 8 observation-collection weeks. Subjects had to be apparently healthy and with no restriction related to the menu offered by the SOSEP system. Children were excluded from particular analyses when they did not adhere to the urine, fecal, or blood collections. Full exclusion resulted when they declined to participate in the project or the caregivers did not sign consent form.

Ethical approval was granted for the study protocol by the human subjects committee of the Center for the Studies of Sensory Impairment, Aging, and Metabolism (CeSSIAM). The SOSEP's director for the Quetzaltenango area gave authorization to realize the study. To provide all of the food items on the menu, the diet offered to the children was complemented when required by funding from the investigators.

Collection of Fecal, Urinary, and Red Cell Samples

Two samples of feces were collected: the first during the seventh and the second during the eighth week of the cycle. Feces were immediately processed and homogenized, and an aliquot of ∼2 g was stored in a −20°C freezer to be analyzed for G intestinalis by an enzyme-linked immunosorbent assay (ELISA) method. We attempted to collect a 24-hour urine sample during the same eighth and final observation-collection week.

Urine collection was started when a child arrived at the day care center (from 7:00 AM to 8:00 AM); SOSEP personnel and investigators started the collection at each day care center using Vacutainer plastic 24-hour collection container (no. 36499, Becton-Dickinson, Franklin Lakes, NJ). Parents continued the collection at home. The sample collection was finished after 24 hours at the center. The collection process was repeated if incomplete.

Red blood samples were collected after an extraction of whole blood by an experienced phlebotomist using Vacutainer plastic 24-hour collection container (no. 36499, Becton-Dickinson, Franklin Lakes, NJ) and Vacutainer 4-mL tubes with ethylenediaminetetraacetic acid (EDTA) (no. 367861, Becton-Dickinson, Franklin Lakes, NJ). Samples were centrifuged to separate red blood cells (RBCs) from plasma; both were stored in criogenic vials (no. 5000-0012, Nalgene Nunc International Corp., Rochester, NY) at −80°C until shipment to the Institute of Nutrition and Food Technology, Center of Biomedical Research, University of Granada, Granada, Spain, to determine antioxidant enzyme activity in RBCs.

Measurement of the Fecal G intestinalis Concentration

G intestinalis prevalence and intensity were derived in Guatemala using the ProSpecT Giardia EZ Microplate Assay (no. 245896 Remel Inc., Lenexa, KS). The absorbance value was used as a proxy for intensity recognizing that values >300 optical density (O.D.) cannot be further resolved ⁽¹⁹⁾.

Measurement of Urinary Biomarkers of Oxidation

The frozen aliquots of urine in cryogenic tubes were shipped on dry ice to Granada. Measure of oxidative damage to DNA was based on the 8-hydroxydeoxyguanosine (8-OHdG) assay (JaICA, Nikken SEIL Co, Ltd, catalog no. IM-KOGHS 040914E, Shizuoka, Japan) and that to lipid by 15-Isoprostane F_2t [F2-Iso] (no. EA84102606, Oxford Biomedical Research Inc., Rochester Hills, MI).

Measurement of Erythrocyte Biomarkers of Oxidation

The red cell hemolysates were shipped in cryogenic tubes on dry ice to Granada. Hemoglobin (Hb) concentration was determined by the use of Drabkin reagent (no. D5941, Sigma, St. Louis, MO). The final concentration of Hb was adjusted to 5 mg/mL for antioxidant enzyme analyses. Catalase (CAT) activity was determined using Aebi method ⁽²³⁾. Superoxide dismutase (SOD) was analyzed by spectrophotometry ⁽²⁴⁾. Glutathione reductase (GSR) was determined using the method by Carlberg and Mannervik ⁽²⁵⁾. Glutathione peroxidase (GPX) was analyzed using the procedure developed by Flohé and Günzler ⁽²⁶⁾.

Data Handling and Statistical Analysis

SPSS version 20 (IBM, Chicago, IL) was used to create a database. Descriptive statistics were expressed as median, 95% confidence interval, and minimum and maximum. Association of values was tested using the Spearman rank-order coefficient. Comparisons between positive and negative for G intestinalis infection subjects were done using Student t test or Mann-Whitney U test as appropriate. A probability level of ≤0.05 was accepted as statistically significant and of ≤0.10 as indicative of a trend.

RESULTS

Characteristics of the Sample

Of the 87 children enrolled in the study, 74 (85%) were sampled for both fecal antigens and the array of urinary and red cell oxidation biomarkers; 36 were girls, and 38 were boys. The ages ranged from 23 to 81 months, with a mean of 58 ± 15 months and a median of 59 months. Figure 1 illustrates the division of the children by sex and age in relation to the specific day care centers and provides data on the median ages by sex.

FIGURE 1: Characteristics of the sample.

Pattern of Intestinal Colonization With G intestinalis

Based on the cut off criterion of positivity and negativity for Giardia antigen in the ELISA assay, 35 subjects were negative (47%) and 39 were positive (53%). With respect to the distinct day care center sites, the Giardia prevalences were as follows: A, 16%; B, 78%; C, 56%, respectively. There was a significant difference in prevalence by site (P < 0.001 by 3 × 2 χ²).

Distribution of Biomarkers of Oxidative Stress

The median, 95th confidence intervals, and the minimum and maximum values are provided in Table 1. There are no cut off criteria established for normative excretion amounts or red cell enzymes for children in this age group. Interestingly, 8-OHdG and F2-Iso median values were different between sexes (P = 0.03 and P = 0.04, respectively), even when calculating the total amount excreted per day and adjusting to weight. The median daily excretion values for the former biomarker were 3.2 μg/day in boys and 2.1 μg/day in girls (P = 0.008), whereas for the latter it was 1.0 and 0.6 μg/day (P = < 0.001), respectively.

TABLE 1: Descriptive statistics for the urinary and enzymatic biomarkers of oxidation

Interaction of Giardia Infection and the Oxidation-Antioxidation System

As shown in Table 2, the G intestinalis–positive children have significantly higher urinary concentrations of F2-Iso (P < 0.001) and RBC GSR (P = 0.019). Correspondingly, there was a trend, with P values of <0.10, for the comparison of infected and uninfected subjects for higher RBC CAT (P = 0.062) and RBC SOD (P = 0.067). Only urinary 8-OHdG and RBC GPX failed to show any association with giardiasis status.

TABLE 2: Comparison for oxidation biomarkers between subjects with positive and negative diagnosis for Giardia intestinalis (n = 74)

With respect to intensity, using the O.D. absorbance values of the ELISA test as a proxy for intensity of infection, Table 3 provides the Spearman correlation coefficients for the crude absorbance data with the values for the respective urinary and enzymatic biomarkers. Significant positive associations were found between higher O.D. and urinary F2-Iso (<0.001), RBC CAT (P = 0.025), and RBC SOD (P = 0.014).

TABLE 3: Spearman correlations for Giardia intestinalis absorbance and oxidation biomarkers measured in urine and erythrocytes (n = 74)

DISCUSSION

The United Nations, through its operative agencies, is committed to addressing the challenge of “Zero Hunger”; highest among the operative priorities is the reduction in linear growth retardation (stunting) ⁽¹⁾. Guatemala ranks the highest in the prevalence of stunting in the Americas ⁽²⁾. As cited, a subset of studies in Guatemala ^(7,8) and elsewhere ^(9–19) provide evidence for an association between Giardia infections and inferior growth. As in our previous evaluations of the prevalence of asymptomatic intestinal colonization with this protozoa in Guatemalan schoolchildren ^(22,27), we found an abundant prevalence of antigen-positive samples among the stools collected in the 3 day care centers in the SOSEP system in the Quetzaltenango province. Moreover, the 3 study sites differed in their average height-for-age z score (HAZ) status, with stunting rates varying from 35% in center A to 63% in center B to 81% in center C (unpublished findings).

Our results are consistent with the findings of Basu et al ⁽²⁸⁾ who also documented a higher excretion of both biomarkers in adult male subjects. We speculate that behavioral traits of boys at this age may produce greater exposure to oxidative stress or that the susceptibility to oxidation is influenced by sex.

In a previous study at CeSSIAM, the use of the O.D. in the ELISA readings of the Giardia fecal antibody assay showed an inverse correlation with height. Because the reaction intensity for absorbance truncates at an O.D. of 3.0, we postulated with the expansion of the range, perhaps through dilution of high-response samples, an even better rank ordering could be achieved ⁽¹⁹⁾. Nevertheless, using the entire range that the ProSpecT-Giardia-EZ microplate permits, another detectable biological association, namely absorbance versus oxidation biomarkers, shows the potential validity of the former as an indicator of the intensity of Giardia infestation.

We find, heretofore, unrecognized associations between RBC antioxidant enzymes and urine oxidative stress biomarkers and the prevalence and intensity of asymptomatic giardiasis in our study population. With respect to oxidative exposures and urinary biomarkers, a greater excretion would logically signify a greater systemic oxidation. For enzyme activity, the interpretation is not so straightforward. On the one hand, lower activities of enzymes would argue for a depletion of antioxidant capacity, that is, making the subject more vulnerable to oxidative damage, while, on the other hand, higher enzymatic activity might signify a compensatory reaction, that is, adding protection in the face of persistent oxidative stress.

The observed interplay between Giardia intensity and urinary oxidative biomarkers could follow 1 or several causal scenarios. For example, the presence of varying numbers of protozoal pathogens in the intestinal lumen could result in the transmission of stimuli that would provoke oxidation locally within intestinal enterocytes or, with absorption, in somatic cells within the body. Alternatively, independent exposures producing states of low-level enhancement of cellular oxidation could alter immune defenses to a degree that permit the colonization by Giardia. The rank-order association of the phenomena would logically make the intensity of infestation driving a reactive oxidation into the more plausible scenario. It is axiomatic, however, that an association in no way proves causality. Indeed, a third (unmeasured) factor operating in the environment in a graded manner could simultaneously set the stage for both susceptibility to colonization by Giardia and expression of cellular oxidative stress. Insofar as no a priori hypothesis regarding the protozoa-oxidation interaction was part of the study design, our finding could be a unique and nonreproducible statistical association in this instance; only prospective investigation for confirmation in this and other settings of Giardia endemicity would establish the relation as a predictable phenomenon.

We acknowledge a series of strengths and weaknesses in the study. The major strength of the study was the convergence of data collected in a common group of preschool children that allowed a post hoc examination for an association having both a sample size of sufficient power and the possibility for scaling intensity. Moreover we had 2 fecal samples for diagnosis of each child's giardiasis status; it has been suggested that having ≥2 separate samples improves sensitivity ^(22,29,30). Insofar as the range of scaling is limited by the nature of the ELISA method, however, as discussed above, we may have been limited in ascertaining the true degree of association. The failure to have had the foresight to have diluted the fecal samples with the highest absorbance signals to extend the range is a limitation.

In conclusion, interesting and potentially important associations were found between the prevalence of asymptomatic giardiasis and with a proxy for intensity of infection and certain elements in an array of biomarkers related to systemic oxidation in Guatemalan preschool children across 3 settings. Although clear directional causality cannot be established, oxidation may represent a potential mediating factor in growth impairment associated with Giardia colonization of the intestine.

Acknowledgments

The authors thank SOSEP personnel, attendants, and parents for their participation and kind assistance. The authors also thank Raquel Campos, Sheny Romero, Carlos Tánchez, Victoria Pérez Lima, and Jeniece Alvey for their assistance in the sample collection. This work will be used in partial fulfillment of the doctoral requirements that will allow the graduate student, María José Soto-Méndez, to obtain the PhD degree in the University of Granada, Spain.