A Human Intestine Chip Model to Combat Pediatric Malnutrition


Environmental enteric dysfunction (EED), a state of constant gut inflammation observed in children in low resource settings, causes damage to the gut that reduces delivery, absorption, and utilization of nutrients and impairs tissue and bone growth that ultimately leads to stunting and impaired development. In 2020, an estimated 149 million children worldwide under the age of 5 were classified as stunted, with a height-for-age significantly below the worldwide average. This crisis will likely be accelerated by the combination of global conflict, rising poverty and inequalities, climate change, and the COVID-19 pandemic that has already led to spikes in the incidence of wasting, a more acute form of malnutrition. Stunting in childhood also can lead to permanent physical and cognitive injury which significantly burdens individuals and societies contributing to a cycle of poverty. 
Development of a therapeutic intervention for EED has been identified as an urgent global health need. To this end, animal models have been developed to study how nutrition and microbes contribute to the development of this disease, but the lack of human-relevant models has hampered these efforts. Working as part of a global collaboration with investigators at Aga Khan University in Pakistan, University of Virginia, Washington University, Boston Children's Hospital, and Cincinnati Children's Hospital, our group at the Wyss Institute for Biologically Inspired Engineering at Harvard University has leveraged our organ-on-a-chip (Organ Chip) microfluidic culture technology to build a human model of EED to better understand what causes the disease and how to treat it. Our EED Chip is a miniaturized cell culture device that integrates living intestinal epithelial cells isolated from children with EED  that are then exposed to flowing cell culture medium that has been depleted of key nutrients. Using this approach, we can model key features of the intestinal damage that occurs in EED (e.g., increased barrier disruption and production of inflammatory cytokines). This use of globally collaborative science to address health in low- and middle-income country settings was supported by the unique combination of the Wyss Institute with its emphasis on translational science and the Bill and Melinda Gates Foundation, which has established a unique consortium of scientific and clinical experts working collectively to increase our understanding of EED and enable development of new interventions. 
The basis for our model is the human Intestine Chip, which is a two-channel microfluidic device lined by primary duodenal organoid-derived intestinal epithelium interfaced with human intestinal microvascular endothelium that experiences fluid flow and peristalsis like motions that drive its ability to recapitulate both intestinal form and function. The Intestine Chip supports the development of villus like structures and multi-lineage differentiation that mimic the architecture and transcriptional phenotype of the human small intestine in addition to forming a tight epithelial barrier, producing mucus, and carrying out digestion and absorption of nutrients. The Intestine Chip's design also allows control over the nutritional environment of the cells while enabling monitoring of their morphological and functional responses. 

Fig. 1.  A schematic representation of Small Intestine chips.

We initially modified the Intestine Chip to develop a human EED in vitro model by exposing the healthy intestinal cells to a nutritional deficiency that we created by perfusing the chip with a custom made culture medium that is deficient in niacinamide and tryptophan (-N/-T). These two nutrients were targeted because they play an important role in intestinal development, energy balance, and antimicrobial peptide secretion, and deficiencies in these nutrients have previously been implicated in the development of EED.  The results from our initial studies were striking: exposure to -N/-T medium resulted in villus blunting and compromise of barrier function, much as is seen in patients with EED. Moreover, numerous repeats of this study revealed the strength of this effect again and again.

Fig. 2 Immunofluorescence cross section micrographs showing effects of nutritional deficiency on villus-like structures in the Intestine Chips lined by healthy epithelial cells (top) versus cells isolated from EED patients (bottom) in complete medium (Con) or medium deficient in niacin and tryptophan (-N/-T) (yellow, phalloidin; blue, Hoechst; bar = 50 µm).Caption

Our next challenge was to incorporate intestinal epithelium isolated from children with EED so that we could better understand how their intestines respond to nutritional deficiency. Thanks to a fruitful collaboration with Asad Ali and Junaid Iqbal at Aga Khan University, we were able to obtain organoid cultures they generated from biopsy specimens collected during gastroscopies of children who failed to show improvement of EED symptoms following a nutritional intervention. We enzymatically disrupted these organoids and cultured the released intestinal epithelial cells in our Intestine Chips to create EED Chips. We then explored the functional, metabolic and transcriptional differences in EED versus healthy Intestine Chips when cultured with complete or nutritionally deficient medium.
In collaboration with Lee Denson (Cincinnati Children’s Hospital) and Sean Moore (University of Virginia), we were able to directly compare our transcriptional data obtained from our Organ Chips to clinical data they obtained from EED patients in Pakistan. Importantly, we found that when chips lined with EED intestinal epithelium are exposed to nutritional deficiency, they exhibit transcriptional changes similar to those seen in clinical biopsy specimens from EED patients in Pakistan as well as in Bangladesh and Zambia.  We saw a high degree of similarity between the transcriptomes including a 60% overlap within the top ten upregulated genes from the clinical signature. Having established that our model was useful, we then utilized COMPBIO, a unique contextual language processing program (Richard Head, Washington University), to analyze our results, which led to the identification of chemokines, brush border structural integrity, fatty acid uptake and amino acid transport as pathways affected by exposure of EED Intestine Chips to nutritional deficiency. 
Because the Intestine Chip offers the unique ability to analyze the contributions of multiple potential contributors to disease development, both alone and in combination, we were able to demonstrate that morphological and functional changes, including villus blunting, intestinal barrier disfunction, reduced fatty acid uptake, and mucus secretion were attributed mainly to nutritional deficiency exposure.  In contrast, EED Intestine Chips showed impaired amino acid transport and a reduced level of secreted cytokines at baseline that became significantly upregulated compared to healthy Intestine Chips when exposed to -N/-T deficiencies.

Fig. 3. Nutritionally deficient EED Chips recapitulate EED patient transcriptional signatures. Of the top 9 upregulated genes in the clinical EED signature, 6 were also upregulated when EED chips were exposed to -N/-T medium.

This novel in vitro human model of EED - a pediatric disorder affecting millions of children worldwide - may be leveraged to elucidate disease pathophysiology and enable the development of new prevention and/or therapeutic measures. In the studies described in this manuscript, we were able to directly associate phenotypic and genotypic changes seen in EED patients with this complex and multifactorial disease with nutritional deficiency and genetic/epigenetic changes that we observed in the intestinal epithelium.  Future studies could leverage additional Organ Chip models, such as co-culture of the EED Intestine Chip with complex gut microbiome derived from EED patients under physiologically relevant oxygen gradients, as well as with a Lymphoid Follicle Chip that recapitulates complex immune responses, to understand how these factors interplay so as to result in impaired response to oral vaccinations that are often observed in children with EED.