Our study demonstrates that microplastic feeding affects both composition and diversity of colonic microbial communities. To date, the studies on this field have been focused on the effect of MPs on the gut microbial communities of soil animals or mice 14,15,36,37, and to our knowledge, this is the first report about the modifications and potential effects of microplastics on human colonic microbiota. Moreover, the doses of MPs tested in previous studies with animal models are higher than those detected in edible foods and beverages, but lower than the estimated human daily intake 14,16,36,38− 42. Our approach was to expose human colonic microbiota to a concentration of microplastics closer to reality, so we selected 166 mg/intake of PET MPs, corresponding to the estimated daily intake in humans, according to the study of Senathiraja et al. 3. We focused our study on the changes during the gastrointestinal digestion and fermentation processes both of PET microparticles’ morphology and structure and of the colonic microbial communities responsible for metabolic bioconversions in the large intestine. For this reason, PET MPs underwent different treatments to simulate the oral, gastric and small intestinal phases of the digestion before getting in contact with previously stabilized human colonic microbiota in the simgi® dynamic simulator, which mimicked the overall impact of digested microplastics on the complex microbial intestinal ecosystem19,25.
PET thermoplastic, as well as other polymers materials, show viscoelastic behavior during grinding, which results in a plastic deformation. This inhibited crack initiation, and hence a break-up did not happen. Thus, the PET pellets were grinded in liquid nitrogen, and this allowed cracking them into smaller particles of ca. 60 µm. PET MPs surface evidenced the appearance of brittle fractures with abundant river marks, associated with the compression side during cryogrinding, and few ductile fracture regions. In agreement, the Raman spectra of the MPs suggest a slight loss of crystallinity with respect to the net pellet. Regarding the effect of gastrointestinal digestion and colonic fermentation on PET MPs, FESEM images revealed an evolution of the surface after in vitro intestinal digestion, indicative of a slight interaction with the media, which resulted, according to Raman characterization, with a relative amorphization of the PET structure, but no remarkable morphological changes. PET MPs average size was large enough to avoid disintegration during digestion, allowing us to monitor the MPs at the different intestinal regions. Salt and organic matter deposits were observed on the particle surfaces after gastrointestinal digestion, as previously reported in the work of Stock et al 12. Crystalline deposits were more abundant, fact that can be related to the increase in salts concentration due to simulated gastric fluid (SGF) and simulated intestinal fluid (SIF) addition during gastrointestinal digestion. Besides, some organic deposits, probably forming the so-called (protein-) corona 12, were also found on PET MPs surfaces at this point, possibly due to the presence of enzymes from the digestion simulation. MPs morphologies show the same appearance after gastrointestinal digestion agreeing with Stock 12 results, despite particle size differences. It has been suggested that the particles’ size, more than their chemical structure, as well as their possible deformation or degradation, and the protein corona formation during the digestive process are crucial for the bioavailability, and thus for the intestinal uptake rate of the particles and the subsequent toxicological impact and health risks 12.
After colonic fermentation, PET MPs presented remarkable organic deposits on their surface, which seemed to be colonized by some members of the colonic microbiota, and the surface roughness evolved towards a globular surface. Moreover, the crystalline structure of the PET MPs is affected after the colonic fermentation, which confirms the biotransformation of the polymer by the human digestion process already observed after the gastric step. The FESEM micrographs show in some cases damaged or broken bacteria adhered to the MPs surface, not allowing to observe an intact microbiota. This may be due to the fact that the cells were not subjected to a fixation process just after sampling and prior to the treatment for their observation in the FESEM, so as not to alter the structure of the PET MPs. Even so, different adhered bacterial cells and biofilm-like structures were observed in colonic samples, supporting the hypothesis that some members of the colonic microbiota could adhere to MPs surface. In relation to the MPs colonization, different human gut microbial species, such as Escherichia coli, Pseudomonas aeruginosa and Staphylococus epidermidis, have shown the ability to adhere and even form biofilms on diverse plastic material surfaces such as polyethylene, polypropylene and polystyrene, among others 8,43,44. Besides, a recent work has shown the adhesion of the gut microbiota of honey bees to the surface of polystyrene MPs 45. Bacterial communities present on colonic microbiota adhere and colonize the gut mucosa, forming biofilms essential to their cross-feeding relationships, nutrient availability and protection against toxins (i.e. antibiotics), mechanical damage and shear caused by fluid flow 43. In our study, the possible adhesion of some members of the colonic microbiota to the PET MPs surface could be due to the absence of a mucus layer or intestinal epithelium to adhere to in the simgi®, promoting the formation of biofilms on MPs with the aim to protect themselves and establish their relationships and functions as a community. Another possibility could be the adhesion of some bacterial species able to metabolize and degrade the PET MPs. In this regard, although it has been reported that PET is resistant to biodegradation, various bacterial hydrolases from environmental samples, such as cutinases, lipases, carboxylesterases, and esterases, have been shown to degrade PET to different extents 46. In general, polymer biodegradation processes include different degrees of material decomposition, which can start with non-enzymatic hydrolysis that promotes fragmentation, followed by assimilation by microorganisms that further involves the enzymatic degradation 47. In this sense, the Ideonella sakaiensis PETase depolymerizes PET, liberating soluble products 48. Besides, some gut bacterial strains from earthworms have displayed the ability to reduce significantly the particle size of low-density polyethylene 44. In our work, Raman results showed progressive amorphization of PET MPs during gastrointestinal digestion, possibly related to oxidative amorphization of PET 33, and more evident after colonic fermentation. These structural changes in the PET MPs particles suggest a potential biodegradation probably driven by colonic microbiota, supporting the existence of an interaction between the colonic microbiota and PET MPs particles. However, although we cannot rule out the presence of PET degrading activities in some specific members of the human colonic microbiota, so far there are no studies or evidence of the existence of human gut bacterial species able to degrade MPs.
Further research is needed to analyze the possible PET MPs colonization by members of human colonic microbiota after ingestion and the mechanisms and functions underlying this adhesion, as well as to explore the possible presence of bacterial activities on human gut microbiome susceptible to biodegrade PET MPs or affect its morphology and structure. Additionally, longer duration experiments should be performed to observe if bacteria from colonic microbiota can degrade plastic over time when faced with a labile C limitation. Another factor for future consideration is the comparison with different MPs polymers and sizes, and the potential degradation depending on the polymer type. Moreover, in the present study a net PET without polymer processing additives was used. The effect of such polymer additives must be also further evaluated because are currently used in commercial products.
Regarding the effect of the intervention with PET MPs on the colonic microbial communities stabilized in simgi®, plate counts and relative abundances of different bacterial groups revealed changes after 72 h of fermentation. The microbiota has undergone a stabilization process prior to the intervention for 14 days followed to a control period of 5 days (120 h) with stable microbial populations presenting a microbial profile of the similar to time 0 h of PET MPs exposure, so we assume that without any disturbance, colonic microbiota should maintain their levels and remain stable for the 72 h of PET MPs exposure. This fact is supported by bacterial counts results, that showed stable levels of different bacterial groups as lactic acid bacteria, Staphylococcus and Enterococcus during the 72 h.
Bacterial counts show reductions in total aerobic and anaerobic bacteria, as well as in Bifidobacterium spp. and Clostridium spp. This suggests that PET MPs and/or their potential resulting constituent monomers exert a negative effect against colonic microbiota, decreasing the levels of total viable bacteria, and to a greater extent of certain microbial groups such as Bifidobacterium, Clostridium and enterobacteria. Even though the bacterial counts method is widely used in simulators to routinely monitor how the colonic microbiota is evolving during the process, this technique provides information only about the viable and cultivable bacteria, allowing to detect about 20% of the present microbial communities 49. Hence, it is not the best approximation to deepen into the real microbial changes produced in response to the intervention with PET MPs in simgi®. Accordingly, 16S rRNA gene metagenomic analysis was performed to delve in the microbial communities’ changes, obtaining different trends for the bacterial biodiversity and for different taxonomic levels.
PET MPs intervention revealed changes in the biodiversity and the relative abundances of different taxa in the three compartments of simgi®. Regarding biodiversity, PET MPs promote a decrease in the alpha diversity indices in terms of Observed species and Shannon index, more evident for the TC and DC compartments. This decrease in the alpha biodiversity indices has been previously reported in animal models after a exposure to MPs, as well as the changes in the beta diversity 14,38,39,41,45, indicating that the structure of the gut microbiota is altered after MPs exposure. Bacteroidetes levels showed an important decrease after PET MPs intervention, being this effect more powerful in the case of TC and DC compartments, reaching levels below 5% of relative abundance. This result also agrees with those reported in other intervention studies with MPs in different animal models that detected an important drop in the relative abundance of the members of this phylum 16,38,39,41. This trend is also supported by our results at family and genus levels, detecting lower levels of Bacteroides and Parabacteroides, in agreement with other studies 50,51, and suggesting that MPs could have an important antibacterial effect on these key members of the gut microbiota. Bacteroides and Parabacteroides species include many important opportunistic pathogens, but as essential members of a balanced microbiota, they are considered to be health-maintaining. These two groups have the ability to reinforce the epithelial barrier and ameliorate inflammation by producing anti-inflammatory molecules such as polysaccharide A (PSA), sphingolipids and outer membrane vesicles (OMVs) for the transport of these molecules to the epithelium, and produce antibacterial molecules to prevent the colonization and invasion of exogenous bacteria 52,53. So, the decrease detected in the proportions of these taxa in vitro after PET MPs intervention is expected be reproduced in vivo in the human gut, which would mean the decrease or even loss of two essential groups not only for the maintenance of the correct balance of the gut microbial communities, but also for the intestinal immune homeostasis and barrier function, which imbalance has been associated with systemic and intestinal diseases, such as inflammatory bowel disease (IBD) and irritable bowel syndrome (IBS), among others 52,54. Firmicutes phylum members increased in the AC compartment, although this trend was not detected for the other two colonic compartments. The increase in the proportions of Firmicutes phyla after MPs intervention is a trend previously reported in other animal models such as zebrafish and mice 16,38,39,41. The changes in the proportions of these two phyla promotes an increment in Firmicutes/Bacteroidetes ratio, known to be of significant relevance for the human gut microbiota status, and often used as a biomarker in connection with human physiology, being its imbalance related to different metabolic disorders such as obesity and diabetes 55− 58. Moreover, the increase detected in Proteobacteria proportions in the TC and DC compartments also agrees with other studies in different animal models 16,40,42, being in our work this rise mainly driven by the increase in the proportions of Escherichia/Shigella and Bilophila genus members. Although a decrease in bacterial counts of Enterobacteria group is detected, 16S rRNA gene metagenomic analysis revealed an increase in the proportions of Enterobacteriaceae family and Escherichia/Shigella genus members. These techniques provide different approaches, not comparable but complementary, being bacterial counts a quantitative approach and the 16S rRNA sequencing a relative abundance result relative to the total number of microorganisms. In addition, the 16S rRNA gene analysis is capable of detecting not only cultivable microorganisms, but also microbial groups that cannot be cultivated under laboratory conditions, providing an image more complete and closer to the real situation. Bearing this in mind, this contradictory result could be due to the fact that despite a decrease in the total counts of members of this group is detected, also observed for the total of aerobes and anaerobes in general, there seems to be an increase in their relative proportion within the microbiota as a whole. A similar effect was also reported by Lu and colleagues, that observed a decrease in α-Proteobacteria levels by qPCR, and an increase in their relative abundance by 16S rRNA sequencing 15. Proteobacteria is a microbial signature of inflammation in the gut 59, and some members belonging to this phylum, such as Escherichia/Shigella and Bilophila,, have been widely associated with a pro-inflammatory effect, as for example Bilophila wadsworthia, that promotes pro-inflammatory TH1 immunity and exacerbates colitis in IBD-prone Il10−/− mice 60. The rise in the proportions of these pro-inflammatory bacterial groups has been related to different diseases, such as IBD, colorectal cancer, and coronary artery disease 61− 64. Finally, Synergistetes phylum also reveals an important increase in response to PET MPs exposure in simgi®, being the rise in Cloacibacillus genus the main responsible of this effect. Members belonging to this genus have been detected in elevated levels in patients with type 2 diabetes, colorectal cancer and even in Parkinson’s disease 65− 67, and have also been positively correlated to a pro-inflammatory effect 68.
In summary, our results reveal that the exposure of the human colonic microbiota to PET MPs clearly affected the microbial communities present, as reported for other micro- and nanoplastics 13,37,69, and other nanoparticles 29. This might negatively affect human health, by: i) the decrease of microbial groups essential for the correct balance of the gut microbiota and for the intestinal immune homeostasis and barrier function; and ii) the increase of different well-known pro-inflammatory and disease-related bacterial groups, fact that could lead to a chronic gut inflammation state, damaging the epithelial barrier and intestinal permeability, and increasing the risk of developing not only local diseases, such as IBD, but also systemic disorders. Besides, this work is the first study that evaluates the potential bidirectional effect of the PET MPs in an in vitro standardized gastrointestinal digestion static model and gut-microbial dynamic fermentation. The simgi® simulator allows to recreate and control in vitro a situation closer to reality than other in vitro models in terms of physiological conditions, human gastrointestinal microbiota populations and metabolic activity, providing a more real image of the interactions between PET MPs and human gut microbial communities. Our findings agree with trends reported in animal models that analyse the effect of different MPs on gut microbial communities, and have biological relevance from a microbial and health point of view. However, to date there are only a few works focused on this field, and none of them has been carried out in a gastrointestinal simulator that allows to mimic the physiological conditions of human gut microbiome and digestion process, fact that makes it difficult to compare the results. Furthermore, our data are not enough to perform a powerful statistical analysis. Nevertheless, even considering these limitations, our results suggest a clear interaction between the PET MPs and the human colonic microbiota, and establish a knowledge basis of the potential PET MPs effects on the bacterial communities present in the human intestine, and the possible effect of these microbial populations on PET MPs, useful for future investigations oriented to unravelling the real effect of these environmental contaminants on human health.
On the other hand, considering that the presence of microplastics is generalized and that the amount and variety of microplastics in the environment may rise exponentially with the massive increase in the use of polymers 70, initiatives to diminish the utilization of single-use plastics are important to reduce their impact. However, the vast majority of plastics are far from a circular economy process that would undoubtedly minimize the uncontrolled increase of microplastics in the environment. This study, which considered the impact of PET primary MPs, covered only one of the key possible impact points of MP ingestion on gut microbiota. Other critical aspect is the large number of process additives present in polymers, with unsuspected effects on gut microbiota. Polymer materials used in food packaging follow a strict regulation 71 that authorizes just hundreds of polymer additives, but those that are prohibited for food packaging, that exceed the number of approved ones, are used in polymers for other applications that could partly end up as MPs and form part of the diet, albeit undesirably. Moreover, MP ingestion could generate additional risk factors, since MPs coming from the environment could also act as vectors for possible pathogens or contaminants, which could directly or indirectly impact in gut microbiota and be related with gut dysbiosis 13,72. Hence, further research is needed to elucidate the effect of MPs intake on the human gut microbiome homeostasis, and thus be able to assess the risk that MPs ingestion through diet has on human health.