INTRODUCTION

Mucosal-associated lymphoid tissue (MALT) is the first line of defense against pathogens, in which secretory immunoglobulin A (S-IgA) plays an important role (¹). Animal studies have demonstrated a relationship between feeding (route of feeding) and fasting S-IgA levels (²-⁵). Modulation of the immune response by gut-associated lymphoid tissue (GALT) is based on S-IgA and lymphotoxin (LT) production. LT, a member of the tumor necrosis factor superfamily, is located on the surface of T and B cells as a heterotrimer composed of one alpha and two beta chains (LTa1b2). In addition to regulating the growth and function of lymphocytes, LT is important for developing secondary lymphoids (intestinal Peyer's patches) and intestinal immune responses. Preclinical studies have shown that feeding through the enteral route maintains the concentration of S-IgA, whereas total parenteral nutrition (TPN) significantly reduces its concentration and decreases the GALT size (⁵-¹⁶).

The most significant changes associated with the lack of enteral stimulation occur in the lamina propria; a decrease in S-IgA production and transportation to the intestinal lumen has been reported. Moreover, there is a decrease in the number of Th2 cells, which further reduces the levels of the interleukins (IL) IL-4 and IL-10, two cytokines necessary for S-IgA synthesis (¹⁷). In addition, there was a shift in the microbiome to gram-negative strains with an increase in circulating lipopolysaccharide (LPS), leading to an exacerbated pro-inflammatory response and impacting GALT function and development (¹⁶,¹⁸) (Fig. 1).

Figure 1. Changes produced during fasting and their effect on the development of the GALT. 

However, there are only a few reports in humans because of the complexity of GALT samples. GALT levels did not vary in healthy adults who volunteered to receive TPN for two weeks; biopsies from the proximal jejunum did not show changes in intraepithelial lymphocytes (IELs) or immunoglobulins (IgA, IgM, and IgG); nevertheless, the nutritional status was not assessed (¹⁹).

To our knowledge, no study has assessed the changes in S-IgA concentrations and LT expression in healthy individuals after acute fasting. Therefore, our study aimed to document whether S-IgA in parotid saliva and serum inflammation markers, including LPS, IL-10 and TNFα concentrations and LT expression, were affected in healthy volunteers who fasted for 72 hours.

METHODS

We performed a proof-of-concept study in July 2019 on a group of people who fasted for three days as part of a prehispanic ritual according to their cultural belief that fasting purifies their souls. The fasting includes no water or food for three days and the subjects were camping in a mountain. This study was conducted in accordance with the Declaration of Helsinki, and all procedures involving human subjects were approved by the Committee of Biomedical Research on Humans. Written informed consent was obtained from all the subjects. Anthropometric measurements, body composition, and handgrip strength were recorded. Subsequently, we collected parotid saliva and peripheral blood samples at the baseline (T0, no fasting) and on day 3 (T3, 72 h of fasting).

RESULTS

We studied 11 participants with a median age of 43 years (40-56), seven of whom were male. The median BMI was 27.3 kg/m² (24.9-29.1). The handgrip strength was normal in 8 participants (72.7 %), according to reference values for Mexicans(²⁵). The median phase angle was 6.7º (6.4-7.1), and the median extracellular total water/body total water ratio (ECW/BTW) was 0.371 (0.366-0.377). The lean body mass was 33.7 kg (30.9-36.5) in males and 28 kg (26.1-28) in females. The fat mass was 20.5 % (15-25.6) in males and 33.9 % (33.4-33.9) in females.

Ten participants had at least one of the following conditions: hypercholesterolemia, hypertriglyceridemia, overweight, or obesity. There was no effect on metabolic biomarkers after 3 days of fasting (glucose, TC, HDL-C, TG, CPR and LPS), except for LDL-C concentrations, which showed a significant increase (109.4 mg/dl vs 131 mg/dl, p = 0.010) (Table I).

Table I. Biochemical and inflammatory markers (n = 11). 

T0: basal time; T3: 72 hours of fasting; IR: interquartile range; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; CRP: C-reactive protein; LPS: lipopolysaccharide; S-IgA: secretory immunoglobulin A; LT: lymphotoxin; IL-10: interleukin 10; TNFα: tumor necrosis factor alfa. Median comparison by Wilcoxon's test.

^*p < 0.05.

There were no statistically significant differences in S-IgA concentrations, LT expression, or IL-10 levels between days 0 and 3. We did not observe any correlation between S-IgA and LT expression levels. However, we noted a significative reduction in TNFα concentration after the 3-day fasting (13.5 pg/mL vs 11.3 pg/mL, p = 0.005) (Fig. 2). We found no association among S-IgA, LT, or other cytokines.

Figure 2. Comparison between basal time (T0) and 72 hours (T3) concentrations: A. C-reactive protein (CRP). B. Lipopolysaccharide (LPS). C. Interleukin 10 (IL-10). D. Tumor necrosis factor alpha (TNFα). E. S-IgA. F. Lymphotoxin (LT) relative expression. Median comparison by Wilcoxon test. 

DISCUSSION

This is the first study performed in humans to assess the effects of fasting on S-IgA concentrations in saliva and LT α1ß2 mRNA transcripts as markers of mucosal immune response in healthy volunteers.

Animal studies, mostly in mice and rats, have shown that fasting (understood as the lack of enteral stimulation) has deleterious effects on GALT development and S-IgA production; however, when enteral stimulation is reintroduced, these changes are reversed (³,⁶,¹,¹⁷,²⁶-²⁸). All murine studies confirmed that S-IgA concentrations drastically decreased after the first day without enteral stimulation. At the same time, there was a decrease in the mucosal addressin cell adhesion molecule 1 (MadCAM-1) expression, needed for producing S-IgA and developing GALT. However, when enteral feeding was reestablished, MAdCAM-1 increased again. This suggests that damage to GALT is reversible (³,⁶,¹¹,²⁶-²⁸) (³,⁶,¹¹,²⁷-²⁹).

LT α1ß2 is lymphotoxin beta receptor's (LTßR) ligand, which plays a vital role in GALT differentiation and S-IgA production (³⁰). The receptor is located inside the Peyer patches and lamina propria, making it difficult to study in humans, mainly for ethical reasons. Therefore, our study investigated the ligand expression. Other studies have reported that blocking LTßR produces similar alterations in the mucosa to when TPN is administered to mice.

Our results showed a decrease in the gene expression, and there was no difference in the expression after 72 hours of fasting; maybe due to the small sample size, we couldn't observe changes. Different types of cytokines, such as IL-10, IL-4, and IL-6, are required to produce S-IgA (³¹,³²). We measured IL-10 production but did not observe any change over time.

In contrast, germ-free mice have low levels of S-IgA that normalize through bacterial colonization, and there is an activation of the Peyer's patches that increases the number of plasmatic B cells in the lamina propria (³³,³⁴). Furthermore, S-IgA interacts with commensal microbiota, inducing competition for the niche against pathogenic bacteria. Thus, IgA protects the host from pathogenic bacteria (³⁴). We measured LPS concentrations in serum as an indirect marker of bacterial type and also TNFα as an inflammatory marker, but we did not observe differences between days 0 and 3 in LPS and S-IgA.

There was a statistical difference in TNFα, consistent with other studies. In an animal study (³⁵), it has been observed how 30 days of intermittent fasting attenuates the inflammatory response of LPS by decreasing TNFα, IL-6, and IFN-γ, among other cytokines in the serum (³⁶,³⁷). Additionally, the study also demonstrated how rats that had 24 h of fasting could reduce the production of inflammatory cytokines, including TNFα, proposing fasting as a dietary measure to reduce inflammation in subjects with central adiposity.

We did not observe any changes in C-reactive protein (CRP) levels. It is an acute-phase protein associated with inflammation. It is associated with insulin resistance and the development of chronic diseases (³⁸-⁴⁰). Although the glucose levels in our study on fasting day 3 showed no changes, we hypothesized that the more days of fasting, the more inflammation, which may lead to a change in the intestinal mucosa barrier. In addition, we observed an increase in LPS concentration; however, this increase was not significant.

Our study had some limitations. First, we had a small sample size; however, performing studies with healthy subjects under fasting conditions was difficult. The subjects were camping in a mountain as part of the prehispanic ritual so that made difficult to obtain different kind of samples. Therefore, we could not obtain fecal samples to examine microbiota changes. We did not measure more cytokines needed for S-IgA production, such as IL-13 and IL-4. Our study's strengths are that this is one of the few studies performed in humans with complete fasting (no water and food for three days). We measured LT expression as a possible indicator of S-IgA production in the absence of LTßR expression when it is impossible to obtain an intestinal sample.

It is interesting to note that the results of previous studies may be due to the synergy between stress (injury, trauma, and diseases), malnutrition (undernutrition), and a lack of enteral stimulation. In our study, stress and malnutrition were not observed, which may explain why we could not detect changes in the mucosal immune response after fasting. Therefore, fasting (72 h) alone is not responsible for the decrease in the mucosal immunological response.

In this study, in people fasting for three days, we concluded that there were no differences in S-IgA concentrations, LT expression, LPS, and IL-10 concentrations. Further studies are needed to elucidate the role of ligand LT in the production of S-IgA and its relationship with nutritional status and inflammatory factors.