Celiac sprue (also known as celiac disease or gluten-sensitive enteropathy) is an inheritable disease characterized by a small intestine inflammatory reaction to gluten proteins from widely prevalent food sources such as wheat, rye and barley. Symptoms include fatigue, diarrhea, and malabsorption of nutrients which can lead to anemia or osteoporosis. Celiac sprue is a lifelong disease, and if untreated it is associated with increased mortality. Despite its high prevalence in most population groups (>1:200) and serious manifestations, there is no other approved therapy for celiac sprue other than complete, life-long gluten exclusion from the patient's diet. Maintaining a strict gluten-free diet is challenging, if not impossible, because gluten is an abundant food ingredient and additive. Thus, an alternative non-dietary therapy is urgently needed.

Gluten immunotoxicity is closely related to its digestive resistance in the gastrointestinal tract. The predominant amino acid residues in gluten are proline(15%) and glutamine(35%), a structure which makes gluten difficult to digest by physiological enzymes such as pepsin, trypsin, elastase and chymotrypsin. Therefore, a potentially attractive therapeutic option is to supplement the diet with an oral enzyme (or enzyme combination) that accelerates breakdown of immunotoxic gluten peptides and detoxifies gluten before it reaches the small intestine, the site of the inflammatory response. Various proteases have been studied to identify an optimal oral enzyme therapy, taking into account gluten specificity and resistance to the harsh physiological gastric conditions (pepsin and low pH). Two distinct two-enzyme combinations have been the focus of this study: (1) Combination of a glutamine-specific cysteine protease (EP-B2) from barley and a proline-specific bacterial endopeptidase from Sphingomonas capsulata (SC PEP), (2) Combination of an aspergillopepsin (ASP) and a fungal dipeptidyl peptidase IV (DPPIV)

The first option involves highly specific cleavage of the most abundant amino acid residues in gluten. Both EP-B2 and SC PEP cleave gluten under gastric conditions, namely the presence of pepsin and an acidic environment. However, the activity of SC PEP is sub-optimal under acidic conditions and is susceptible to pepsin. Thus, we used a combination of sequence and structure-based approach in addition to machine learning algorithms and recursive mutagenesis to identify improved variants of SC PEP. Mutants with significantly improved activity under simulated gastric conditions were identified and analyzed.

To evaluate the second option, we investigated the properties of selected proteases that have been previously approved for human consumption. While neither ASP nor DPPIV alone is unable to detoxify gluten under simulated gastric conditions, a defined dose combination of ASP and DPPIV was able to thoroughly detoxify moderate quantities of dietary gluten.