What Is Known/What Is New
What Is Known
- Polyethylene Glycol 3350 is a commonly used laxative to treat constipation in children.
- Concerns that this laxative may lead to neuropsychiatric symptoms in treated children have been reported.
- Behavioral disorders are common in children with constipation.
- No prior studies have assessed the effects of polyethylene glycol 3350 on behavior in humans or animals.
What Is New
- Daily administration of Polyethylene Glycol 3350 for 2 weeks changes microbiome composition in murine model.
- Polyethylene Glycol 3350 does not affect anxiety-like behavior in mice.
See “Raising the Bar on Translational Studies to Test the Neurobehavioral Effects of Laxatives” by Arbizu and Rao on page 427.
Polyethylene Glycol (PEG) 3350 is a water-soluble organic polymer that is minimally absorbed from the gastrointestinal tract and is used as an osmotic laxative. Although the Food and Drug Administration (FDA) has approved PEG3350 to treat constipation in those 18 years and older, it is commonly used “off label” in children (1–3). Studies in children show that PEG3350 is efficacious in the treatment of constipation and superior to other laxatives (4–8). The North American and European Societies for Pediatric Gastroenterology, Hepatology, and Nutrition recommend PEG3350 as the first-line therapy for treatment of constipation in children (9).
The FDA Adverse Events Reporting System received reports of neuropsychiatric symptoms in children treated with PEG3350, such as anxiety, aggression, and obsessive compulsive behaviors (10). Although previous studies that assessed the efficacy and tolerability of PEG3350 did not report any adverse behavioral effects, these studies were not designed to specifically evaluate behavior (4,5,8,11). Moreover, behavioral/neurological/psychological disorders are prevalent in children with constipation (12–14). Thus, determining whether adverse behavioral events are because of medication or associated with a pre-existing condition is difficult to ascertain.
One of the potential mechanisms by which PEG3350 may alter psychological and neurological functions, is through its effect on the brain-gut microbiome axis, which has been recognized to impact behavior in both preclinical and clinical studies (15–18). The influence that gut microbes have on the brain and behavior is particularly evident in anxiety disorders, wherein patients with generalized anxiety disorders show significant differences in the gut microbiota compared with healthy controls (19). In addition, transplanting fecal microbiota from irritable bowel syndrome (IBS) patients with comorbid anxiety leads to anxiety-like behaviors in recipient mice (20). PEG-based laxatives have the potential to change the gut microbiome (21–23).
Anxiety and anxiety-like behaviors are related to many of the neuropsychiatric symptoms reported to the FDA in children using PEG3350 (10). Thus, the main goal of our study was to determine if daily administration of PEG3350 leads to anxiety-like behavior in male and female mice. Additionally, to evaluate potential mechanisms of behavioral changes (if any), the diversity and composition of the gut microbiome was evaluated.
See (text, Supplemental Digital Content 1, for more details about this study methods, https://links.lww.com/MPG/C411).
CD-1 IGS mice ages 6 to 8 weeks were obtained from Charles Rivers Laboratories and housed 3 per cage under alternating 12 : 12 hours light-dark cycle. Mice had ad libitum access to water and standard chow. All study procedures were approved by the Institutional Animal Care and Use Committee (IACUC) at Nationwide Children's Hospital.
Polyethylene Glycol (PEG) 3350 (Perrigo Company plc., Allegan, Michigan, USA) was used at a dosage of 4 g/kg (designated as Hi PEG) or 1 g/kg (designated as Lo PEG). PEG3350 powder was dissolved in water and each mouse received 0.25 mL via oral gavage.
Magnesium citrate Spectrum Chemical MFG. Corp., Gardena, California, USA) was used to determine if any potential behavioral changes were specific to PEG3350 or to laxative use, in general. Two dosages were selected for use, 600 mg/kg (designated as Hi Mag) and 450 mg/kg (designated as Lo Mag). Magnesium citrate powder was dissolved in water and each mouse received 0.25 mL via oral gavage.
Two experiments were performed. Experiment 1 was designed to investigate whether administration of laxatives was associated with changes in behavior. Five groups (n = 9 males; n = 9 females) were randomly assigned to the Lo or Hi PEG group, the Lo or Hi Mag group, or vehicle (water) group, which consisted of daily oral gavage for 14 days. A sixth group (the control group) remained minimally handled. Mice were housed 3 per cage. Behavioral tests were conducted at 3 time-points: baseline, after 14 days of daily laxative or vehicle administration (designated as post-gavage), and after a 14 days washout period (ie, postwashout) (Fig. 1A).
Experiment 2 was designed to determine if behavior was affected by mice habituating to repeat behavioral tests. Six groups of mice (n = 12 males; n = 12 females) underwent the same allocation of drug administration. Behavioral tests were only performed once at the post-gavage time point (Fig. 1B).
Stool samples were collected by placing each mouse in an empty cage. Collected stool was given a score for consistency on a scale from 0 to 3. A score of 0 was a normal hard pellet, 1 was a stool that had a normal pellet shape, but easily dented, 2 was mushy stool without shape, or 3 was watery stool. This was done daily 4 to 6 hours after treating mice via oral gavage to determine effectiveness of laxatives, and twice weekly during the washout period.
Open Field Exploration
Mice were placed in the open field enclosure and allowed to explore for 300 seconds. Behavior was scored using a 16 × 16 photo-beam configuration analyzed by San Diego Instrument's Photo-beam Activity System version 2 software.
The light/dark test apparatus entailed dividing the open field box into light and dark compartments with black plexiglass. An opening allowed mice to move freely between the 2 compartments. Movements were recorded for 300 seconds.
Elevated Plus Maze
Mice were placed in the junction between the open and closed arms of the elevated plus maze (EPM). Behavior was scored using infrared I/R photo-beam activity tracker and MED PC IV software for 300 seconds.
16s rRNA Gene Sequencing
Stool samples from experiment 2 collected on days 0, 14, and 28, were selected randomly from both male and female mice in 4 groups (Hi PEG, Lo PEG, Hi Mag, and vehicle). The QIAamp Fast DNA Stool mini kit (Qiagen, Valencia, CA) was used for DNA extraction. Illumina MiSeq 2 × 250 paired end sequencing using NEXTERA unique dual index primers was used to sequence the V4-V5 16s rRNA variable region. Demultiplexing, quality filtering, calling of amplicon sequence variants (ASVs) by DADA2, and taxonomic assignment (SILVA database) of 16S rRNA gene sequencing data was conducted using the open-source, community-supported software program Quantitative Insights Into Microbial Ecology 2 (QIIME2) (24).
Differences in bacterial alpha diversity were determined using the Kruskal-Wallis nonparametric ANOVA. Unweighted and weighted UniFrac distances and permutational multivariate analysis of variance (PerMANOVA) were used to monitor differences in microbial community beta-diversity. Differences in bacterial taxa were determined using Kruskal-Wallis nonparametric tests. P values were corrected for multiple comparisons using the Bonferroni correction. For diversity and differential abundance analyses, samples were rarefied to 12,100 sequences (the lowest number of reads among the samples).
For behavioral analyses, a 2-factor analysis of variance (ANOVA) was used with treatment and time-point used as the 2 factors (time-point was used as a repeated factor). For experiment 2, a 1-factor ANOVA was utilized. Significant main or interaction effects were followed up with the Tukey post hoc test. Statistical analyses were conducted using IBM SPSS version 26. Differences were considered statistically significant with P < 0.05. Males and females were analyzed separately.
Body Weight and Stool Consistency
The percentage of body weight gained was significantly affected by gavage -including vehicle (water) gavage group—compared with the control group (Fig. 2A and B, Supplemental Digital Content, Tukey post hoc P < 0.05, https://links.lww.com/MPG/C412). By day 28, all groups had gained similar percentages of weight.
At baseline, the mean stool consistency score was near 0 for all mice. Stool consistency scores were significantly higher in mice treated with laxatives compared with the vehicle-gavaged group (Fig. 1C and D; P < 0.05). Stool consistency scores returned to near 0 by day 28 in all the groups.
In the light/dark test, open field and EPM tests, we found that there were no statistically significant effects of the laxatives on any of the measured parameters at any time-point (Fig. 2 and Figs. 3–5, Supplemental Digital Content, https://links.lww.com/MPG/C413, https://links.lww.com/MPG/C414, https://links.lww.com/MPG/C415). For the light/dark exploration, time spent in the dark (Fig. 2A and B), latency to enter the dark chamber, and number of transitions between chambers (Fig. 3A--D, Supplemental Digital Content, https://links.lww.com/MPG/C413) were not different for treated and untreated mice. Likewise, time spent in the periphery (Fig. 2C and D), distance travelled, and number of transitions (Fig. 4A--D, Supplemental Digital Content, https://links.lww.com/MPG/C414) did not differ at any time point. Behaviors on the EPM including time spent in the closed arm (Fig. 2E and F), time spent in the junction, and number of closed arm entrances (Fig. 5A--D, Supplemental Digital Content, https://links.lww.com/MPG/C415) also did not differ between the treatment groups.
We did observe significant effects of time-point on behavioral measures in the light/dark and open field tests. For example, latency to enter the dark (light/dark test) or distance travelled (open field) were significantly decreased in all groups with repeated testing (Fig. 3 and C, Supplemental Digital Content, https://links.lww.com/MPG/C413 and Fig. 4A and C; Ps < 0.05, https://links.lww.com/MPG/C414). As this difference occurred in control animals as well as laxative-treated animals, a second experiment was conducted to determine whether differences in behavior were because of repeated testing.
Body Weight and Stool Consistency
Mice receiving daily gavage did not gain weight during gavage period, (P < 0.05; Fig. 2C and D, Supplemental Digital Content, https://links.lww.com/MPG/C412). By day 28, body weight change in all groups was similar.
Stool consistency scores were significantly affected by laxatives (group × time-point interaction, P < 0.05; Fig. 1E and F). Stool scores were significantly higher in all groups compared with the vehicle-gavaged and control group (P < 0.05) and returned to approximately 0 in all groups during the washout period.
None of the behaviors assessed in Experiment 2 were found to be significantly different between any of the groups in the light/dark exploration (Fig. 3A and B and Fig. 6, Supplemental Digital Content, https://links.lww.com/MPG/C416), open field task (Fig. 3C and D and Fig. 7, Supplemental Digital Content, https://links.lww.com/MPG/C417), and EPM (Fig. 3E and F and Fig. 8, Supplemental Digital Content, https://links.lww.com/MPG/C418), confirming that the differences found between the time-points in experiment 1 were because of repeated behavioral testing and not because of any treatment.
At baseline, both alpha diversity and beta diversity were similar among all mice (Fig. 4). After 2 weeks of treatment, Hi PEG significantly affected alpha diversity as indicated by a decrease in Faith phylogenetic diversity (PD) in female mice (Fig. 4A; P < 0.05 vs. day 0), which persisted throughout the washout period (P < 0.05, Fig. 4A). Hi PEG did not significantly affect Faith PD in males (Fig. 4B). Pielou evenness was not significantly different in any of the groups (Fig. 9A and B, Supplemental Digital Content, https://links.lww.com/MPG/C419).
Beta diversity was significantly different after Hi PEG administration compared with baseline (P < 0.05, Fig. 4C and D). Unweighted and weighted UniFrac distances were significantly different at day 14 in both females and males administered Hi PEG compared with all other groups at all other time-points (P < 0.05; Fig. 4C and D and Fig. 9C and D, Supplemental Digital Content, https://links.lww.com/MPG/C419). The Hi PEG had persistent effects on beta diversity with both unweighted and weighted UniFrac distances being different on day 28 compared with day 0 (P < 0.05). Although less pronounced on the principal coordinate analysis (PCoA) plots (Fig. 4C and D), stool samples from mice given Lo PEG had significantly different unweighted and weighted UniFrac distances on day 14 compared with day 0 (P < 0.05) (Fig. 10, Supplemental Digital Content, https://links.lww.com/MPG/C420). Unweighted UniFrac distances remained different at day 28 compared with day 0, but were not different from day 14 indicating that effects on unweighted UniFrac distances persisted throughout the washout period. Weighted UniFrac distances, on the other hand, were different on day 28 compared with day 0 and day 14 (P < 0.05), suggesting that the effects of Lo PEG on weighted UniFrac distances did not completely persist during the washout period. Hi Mag also affected beta diversity, with unweighted UniFrac distances being significantly different on day 14 and day 28 compared with day 0 (P < 0.05, Fig. 4C and D, and Fig. 11, Supplemental Digital Content, https://links.lww.com/MPG/C421); however, Hi Mag did not significantly affect weighted UniFrac distances (Fig. 11, Supplemental Digital Content, https://links.lww.com/MPG/C421).
We tested whether there were bacterial genera that were significantly different in day 14 stool samples from mice given Hi PEG. A total of 13 genera were statistically significant in females treated with Hi PEG compared with vehicle control on Day 14 (Fig. 12, Supplemental Digital Content, https://links.lww.com/MPG/C422) including Alistipes, Bacteroides, and Akkermansia, 4 unclassified genera within Ruminococcaceae, and 5 within Ruminoclostridium (Fig. 12, Supplemental Digital Content, https://links.lww.com/MPG/C422). In males treated with Hi PEG, 10 genera were statistically significant compared with vehicle controls on day 14. These genera were Akkermansia, Bilophila, Parasutterella, Parabacteroides, Anaerofustis, Bacteroides, Lachnoclostridium, Eubacterium nodatum group, an unclassified Ruminococcaceae and an unclassified Lachnospiraceae (Fig. 13, Supplemental Digital Content, https://links.lww.com/MPG/C423).
To determine whether the relative abundance of bacterial taxa was related to murine behavior, Spearman rank correlation analyses were conducted for behavioral parameters indicative of anxiety-like behavior and the relative abundances of behavioral taxa that were found in at least half of the samples. None of the bacterial relative abundances were significantly correlated with murine anxiety-like behavior (not shown).
Daily administration of PEG3350 in mice did not lead to any differences in anxiety-like behavior in the light-dark exploration, open field task, or EPM that are well-validated behavioral tests capturing different components of locomotor activity and anxiety-like behavior (25–29). Our first approach utilized a repeated measures design used in behavioral studies previously (30,31). We, however, observed that some behaviors were significantly different at the follow-up time-points (in all mice, including control mice) compared with the baseline measures. Although there were no significant differences between any of the treatment groups, this change in behavior over time suggested that our repeated measure design affected behavior (32,33). Thus, to be certain that the lack of significant group effects was not because of our experimental design, we conducted a second experiment, in which mice were only tested at a single time-point post-gavage. This experiment confirmed that none of the laxatives affected behavior.
Despite the lack of any effects on behavior, both PEG3350 and Mag had significant laxative effects in mice, as indicated by changes in stool consistency. To model clinical administration, we gave mice PEG3350 or Mag via a single daily dose. Administration of the Hi PEG had the strongest effect on stool consistency but lower doses of PEG3350 and Mag also led to significant differences in stool consistency. Importantly, the lower dose of PEG3350 used in this study (1 g/kg) is similar to doses used in clinical practice and pediatric research studies (0.8–1 g/kg) (3,6–8). Thus, laxatives administered at clinically relevant doses, as well as at higher doses, led to significant differences in mouse stool consistency, but did not affect any of the behaviors tested. Another observation in the study, was failure of the mice in the 5 gavage groups to gain weight during gavage period compared with the minimally handled sixth group. This was noticed even in the vehicle gavage group; thus, it is unlikely to be related to laxatives. This is likely related to frequent handling and gavage process itself, which could be stressful. All mice eventually gained similar weight by the end of washout period, and weight change was not statistically different amongst all groups.
Multiple preclinical studies have demonstrated a link between the gut microbiome and behavior through communication via the gut-brain axis (15,16,18). Anxiety, in particular, is related to differences in the microbiome in both laboratory animals and humans. For example, Jiang et al (19) in 2018 found that compared with healthy controls, patients with generalized anxiety disorder (GAD) had significant differences in bacterial diversity and relative abundances. Although this may merely be an association between the gut microbiome and anxiety, studies also demonstrate causal links between the gut microbiome and behavioral responses. For example, in a study by De Palma et al (20) in 2017, germfree mice were colonized with fecal microbiota from IBS patients that also had comorbid anxiety exhibited higher anxiety-like behavior compared with mice that received fecal microbiota from IBS patients that did not have anxiety. Thus, we hypothesized that differences in microbiome composition may be 1 way in which PEG3350 could indirectly affect behavior.
In our study, PEG3350 led to differences in the microbiome, marked by significant effects in both alpha and beta diversity. Most noticeably, the Hi PEG, and to a lesser extent the Lo PEG and Hi Mag, significantly decreased alpha diversity after 2 weeks of treatment. The lower alpha diversity was not surprising, as a study by Tropini et al in 2018 also demonstrated that PEG3350-induced osmotic diarrhea in mice colonized with human microbiota led to significantly lower alpha diversity.
We also observed significant differences in bacterial beta diversity in mice treated with Hi PEG, and to a lesser extent with Lo PEG. These effects were most predominant postgavage but differences in bacterial beta diversity were still evident postwashout in mice treated with PEG3350. This suggests that PEG3350 can have effects on the microbiome that persist beyond the treatment period. This is consistent with previous studies (34) but further studies are needed to determine how long differences in the microbiome are evident.
The relative abundances of a number of bacterial taxa were significantly altered in mice treated with PEG3350. Interestingly, Akkermansia relative abundance was significantly increased in mice treated with PEG3350. This microbe has been linked to differences in anxiety-like behavior (35,36). But, in our study, none of the bacterial relative abundances were statistically correlated to anxiety-like behavior, suggesting that although PEG3350 leads to differences in stool consistency and differences in gut bacteria, these differences do not lead to changes in behavior.
We acknowledge that our study has limitations, for instance, we only assessed specific aspects of anxiety-like behavior in mice, and the duration of laxative use was relatively short. Thus, studies should address other behavioral dimensions and chronic laxative use.
It is well documented that behavioral and psychiatric disorders are more prevalent in people with chronic constipation (14,37). This overlap has made it difficult to interpret whether neuropsychiatric symptoms in children taking PEG3350 are secondary to laxative use or naturally co-occur with the underlying constipation. Although multiple studies have reported on the safety of PEG3350 (4–6,8,11), parents are still reluctant to use PEG3350 (38). Although we cannot rule out that laxatives affect behavior in children with chronic constipation, our study found no evidence that PEG3350 (or magnesium citrate) significantly affects anxiety-like behavior in mice tested on standard tasks.
1. Schiller LR, Emmett M, Santa Ana CA, et al. Osmotic effects of polyethylene glycol. Gastroenterology
2. Brady CE 3rd, DiPalma JA, Morawski SG, et al. Urinary excretion of polyethylene glycol 3350
and sulfate after gut lavage with a polyethylene glycol electrolyte lavage solution. Gastroenterology
3. Alper A, Pashankar DS. Polyethylene glycol: a game-changer laxative for children. J Pediatr Gastroenterol Nutr
4. Treepongkaruna S, Simakachorn N, Pienvichit P, et al. A randomised, double-blind study of polyethylene glycol 4000 and lactulose in the treatment of constipation in children. BMC Pediatr
5. Walia R, Steffen R, Feinberg L, et al. Tolerability, safety, and efficacy of PEG 3350 as a 1-day bowel preparation in children. J Pediatr Gastroenterol Nutr
6. Loening-Baucke V. Polyethylene glycol without electrolytes for children with constipation and encopresis. J Pediatr Gastroenterol Nutr
7. Modin L, Walsted AM, Dalby K, et al. Polyethylene glycol maintenance treatment for childhood functional constipation: a randomized, placebo-controlled trial. J Pediatr Gastroenterol Nutr
8. Jarzebicka D, Sieczkowska-Golub J, Kierkus J, et al. PEG 3350 versus lactulose for treatment of functional constipation in children: randomized study. J Pediatr Gastroenterol Nutr
9. Tabbers MM, DiLorenzo C, Berger MY, et al. European Society for Pediatric Gastroenterology, Hepatology, and Nutrition, North American Society for Pediatric Gastroenterology. Evaluation and treatment of functional constipation in infants and children: evidence-based recommendations from ESPGHAN and NASPGHAN. J Pediatr Gastroenterol Nutr
11. Bekkali NLH, Hoekman DR, Liem O, et al. Polyethylene glycol 3350
with electrolytes versus polyethylene glycol 4000 for constipation: a randomized, controlled trial. J Pediatr Gastroenterol Nutr
12. van Dijk M, Benninga MA, Grootenhuis MA, et al. Prevalence and associated clinical characteristics of behavior problems in constipated children. Pediatrics
13. Philips EM, Peeters B, Teeuw AH, et al. Stressful life events in children with functional defecation disorders. J Pediatr Gastroenterol Nutr
14. Kuizenga-Wessel S, Koppen IJN, Vriesman MH, et al. Attention deficit hyperactivity disorder and functional defecation disorders in children. J Pediatr Gastroenterol Nutr
15. Neufeld KA, Kang N, Bienenstock J, et al. Effects of intestinal microbiota on anxiety-like behavior. Commun Integr Biol
16. Diaz Heijtz R, Wang S, Anuar F, et al. Normal gut microbiota modulates brain development and behavior. Proc Natl Acad Sci USA
17. Bravo JA, Julio-Pieper M, Forsythe P, et al. Communication between gastrointestinal bacteria and the nervous system. Curr Opin Pharmacol
18. Cryan JF, O’Riordan KJ, Cowan CSM, et al. The microbiota-gut-brain axis. Physiol Rev
19. Jiang H-y, Zhang X, Yu Z-h, et al. Altered gut microbiota profile in patients with generalized anxiety disorder. J Psychiatr Res
20. De Palma G, Lynch MD, Lu J, et al. Transplantation of fecal microbiota from patients with irritable bowel syndrome alters gut function and behavior in recipient mice. Sci Transl Med
21. Drago L, Toscano M, De Grandi R, et al. Persisting changes of intestinal microbiota after bowel lavage and colonoscopy. Eur J Gastroenterol Hepatol
22. Jalanka J, Salonen A, Salojarvi J, et al. Effects of bowel cleansing on the intestinal microbiota. Gut
23. Gorkiewicz G, Thallinger GG, Trajanoski S, et al. Alterations in the colonic microbiota in response to osmotic diarrhea. PLoS One
24. Bolyen E, Rideout JR, Dillon MR, et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat Biotechnol
25. Crawley JN. Neuropharmacologic specificity of a simple animal model for the behavioral actions of benzodiazepines. Pharmacol Biochem Behav
26. Bourin M, Hascoet M. The mouse light/dark box test. Eur J Pharmacol
27. Walf AA, Frye CA. The use of the elevated plus maze as an assay of anxiety-related behavior in rodents. Nat Protoc
28. Prut L, Belzung C. The open field as a paradigm to measure the effects of drugs on anxiety-like behaviors: a review. Eur J Pharmacol
29. Crawley JN. Exploratory behavior models of anxiety in mice. Neurosci Biobehav Rev
30. Galeano P, Martino Adami PV, Do Carmo S, et al. Longitudinal analysis of the behavioral phenotype in a novel transgenic rat model of early stages of Alzheimer's disease. Front Behav Neurosci
31. Marrocco J, Mairesse J, Ngomba RT, et al. Anxiety-like behavior of prenatally stressed rats is associated with a selective reduction of glutamate release in the ventral hippocampus. J Neurosci
32. Voikar V, Vasar E, Rauvala H. Behavioral alterations induced by repeated testing in C57BL/6J and 129S2/Sv mice: implications for phenotyping screens. Genes Brain Behav
33. McIlwain KL, Merriweather MY, Yuva-Paylor LA, et al. The use of behavioral test batteries: effects of training history. Physiol Behav
34. Tropini C, Moss EL, Merrill BD, et al. Transient osmotic perturbation causes long-term alteration to the gut microbiota. Cell
35. McGaughey KD, Yilmaz-Swenson T, Elsayed NM, et al. Relative abundance of Akkermansia spp. and other bacterial phylotypes correlates with anxiety- and depressive-like behavior following social defeat in mice. Sci Rep
36. Chen YH, Bai J, Wu D, et al. Association between fecal microbiota and generalized anxiety disorder: severity and early treatment response. J Affect Disord
37. Nehra V, Bruce BK, Rath-Harvey DM, et al. Psychological disorders in patients with evacuation disorders and constipation in a tertiary practice. Am J Gastroenterol
38. Hussain SZ, Belkind-Gerson J, Chogle A, et al. Probable neuropsychiatric toxicity of polyethylene glycol: roles of media, internet and the caregivers. GastroHep