It is now more than 50 years since Dicke observed that wheat and wheat products were harmful to coeliac patients. Soon afterwards, van de Kamer and Weijers showed that the gluten fraction of wheat was toxic, whereas the starch and albumin fractions were not. The protein fraction of gluten that was soluble in aqueous alcohol, called gliadin, was found to be the most toxic, whereas the insoluble glutenin fraction was described as either nontoxic or ‘possibly’ toxic .
These early observations were clinical observations made in patients, showing the beneficial effects of gluten withdrawal on symptoms, growth in children and steatorrhoea. Since then, the benefits of a gluten-free diet have been universally accepted for the management of coeliac disease.
Many questions remain to be answered about gluten and how it precipitates the coeliac lesion. During the past 50 years much effort has been spent in investigating gluten in a variety of ways. Most of this effort has been directed towards gliadin and its fractions, whereas the role of glutenin has been largely ignored.
Nomenclature and classification of gluten proteins
Gluten, which is responsible for the excellent baking properties of wheat, is defined as the rubbery, cohesive mass that remains when wheat dough is washed in water to remove starch granules and other soluble constituents. It has classically been divided into two fractions that are either soluble or insoluble in aqueous alcohol. The alcohol-soluble proteins (prolamins) in wheat were termed gliadins and the insoluble proteins (glutelins) in wheat, glutenins. Similar fractions were described in the closely related cereals, rye and barley. More recent studies have shown, however, that the extraction of flour or gluten with aqueous alcohol does not lead to clear-out fractions. Glutenin subunits are therefore found in the soluble gliadin fraction and gliadins are also present in the insoluble glutenin fraction [1,2].
The most important criterion for classifying these protein mixtures is the primary structure. Gluten proteins have been extensively investigated by highly efficient techniques (e.g. two-dimensional electrophoresis, reversed-phase high-performance liquid chromatography) and by analysis of their amino acid composition and sequences. A more satisfactory classification into three groups, based on molecular weight, has also been described  (see Fig. 1).
The three proteins groups of gluten
The high molecular weight (HMW) group contains two different protein types, the x-HMW and the y-HMW subunits. They are characterized by relatively high proportions of glycine and tyrosine. Glutamine, glycine and proline account for 65–70% of the total amino acid residues. The glutenin proteins are typically 600–800 amino acids in length, comprising nonrepetitive N-terminal and C-terminal domains and a long repetitive central domain.
The medium molecular weight group contains ω-gliadins, which contain high amounts of glutamine, proline and phenylalanine, accounting for 80% of the total amino acid composition. The sulphur-containing amino acids, cysteine and methionine, are absent; these were therefore termed the S-poor prolamins. These proteins consist almost entirely of repetitive sequences and are typically 400–500 amino acids in length.
The low molecular weight (LMW) group consists of α-type and γ-type gliadins and the LMW subunits of glutenin. The amino acid sequences of these three protein types form five domains. Apart from the N-terminal domains, the others consist of repetitive sequences, all being rich in glutamine and proline. The LMW proteins are approximately 250–300 amino acids long.
It should be noted that in the many types of proteins described there are high degrees of homology with many similar repetitive sequences, particularly containing glutamine and proline. There is less homology of the repeat motifs of HMW prolamins than of those in medium molecular weight and LMW prolamins.
It should also be noted that more than 100 different proteins from each of the types described exist, which reflect the varieties of wheat and also the growing conditions. There will be 90% homology between varieties, and any heterogeneity is due to substitutions, insertions or deletions of single residues or short oligopeptides.
From this summary of gluten chemistry, it will be apparent that there are many proteins in gluten. The classical gliadin and glutenin fractions are not as distinct as originally thought, and there are many similarities. One would therefore expect that all the proteins may be involved in precipitating coeliac disease and should be excluded from a gluten-free diet.
Assessment of glutenin toxicity in coeliac disease
Mainly because van de Kamer and colleagues  suggested that glutenin was probably not toxic for coeliac patients, this belief has largely persisted. This has been supported by the fact that glutenin proteins are difficult to separate and purify, and therefore to test in coeliac disease. The early assessment of glutenin toxicity by van de Kamer et al.  and Sheldon  was by a symptomatic response and the glutenin fraction used was most probably impure and contaminated with gliadins.
Since then, as our understanding of the chemistry of these proteins has increased, it has been possible to produce more consistent and purified proteins and polypeptides for assessment. The means of testing have also advanced.
In the 1970s–1980s, organ culture of mucosal biopsies was introduced and has been an extremely useful means of in-vitro assessment of protein toxicity  and of investigating coeliac pathogenesis . Our understanding of the mucosal immune system and the immunopathology of coeliac disease has also increased, however, so that the immunological effects of specific peptides have been extensively assessed by a variety of test systems [8–10].
There is, of course, a dichotomy between a protein or peptide being toxic and being immunogenic. Hence, lymphocyte-based systems are assessing immunostimulatory properties, whereas in-vitro biopsy systems may be assessing direct toxicity, indirect toxicity and also immunological effects.
Ultimately, the clinical and mucosal effects must be tested in vivo. Protocols to do this have been developed [11,12].
As a result of these developments some evidence has accumulated about the possible toxicity and immunogenicity of glutenins in coeliac disease.
Immunostimulatory effects of glutenin proteins
In the past 10 years, Koning's and Sollid's groups have used their highly specific and sensitive techniques to test a number of glutenin-derived and gliadin-derived peptides as regards their immunostimulatory effects in coeliac disease.
In 1999 van de Wal et al.  showed that a specific glutenin peptide stimulated coeliac-derived mucosal T cells and that this peptide was repetitively present in many glutenin proteins. The same group showed that there was mucosal T-cell reactivity towards homologous gliadin and glutenin peptides in coeliac children . Molberg et al.  showed that a range of highly purified HMW glutenins were stimulatory to coeliac mucosal T cells. In 2005 Koning's group  demonstrated that there was wide variation in the amount of T-cell-stimulating peptides in both gliadins and glutenins, indicating that some wheat varieties may be less stimulatory than others.
There are important conclusions from these observations. First, glutenin peptides have similar immunostimulatory effects in coeliac disease as gliadin peptides and, secondly, these peptides are repeated throughout the glutenin proteins, including those of HMW. The consequences of these findings are that it is highly likely that glutenins are toxic in coeliac disease, and any attempt to breed wheat varieties with reduced gliadins but unaltered glutenins as a form of dietary modification will be extremely difficult.
In-vivo assessment of glutenin proteins
The ultimate assessment of any gluten protein regarding its toxicity or otherwise in coeliac disease has to be in patients. Ciclitira's group has recently published some observations along these lines . They have previously published their method of in-vivo testing [11,12] and have now shown that within 4–6 h of exposure to HMW x-glutenin and y-glutenin subunits, three treated patients had mucosal architectural changes of reduced villous height:crypt depth ratios and decreased enterocyte height. In two patients there was also a signifcant increase in intraepithelial lymphocytes. Immunofluorescent staining for interleukin-15 increased, particularly in the epithelium, in all three patients within 2 h of peptide exposure. The authors also showed that in mucosal T-cell-stimulatory assays, 11 of 17 coeliac patients' lymphocytes were stimulated by HMW glutenin subunits. Interestingly, this was not dependent upon prior deamidation with tissue transglutaminase.
The authors conclude from these observations that many coeliac patients may be sensitive to glutenin proteins and certainly further work on the specific peptides is necessary. This is important both for the development of novel dietary treatments and for gluten testing in food. They also suggest that their observations add to the growing body of evidence that the innate immune system is involved in the early response to gluten peptides in the evolution of the coeliac mucosal immune response.
Criticisms of such work usually revolve around the purity of the peptides being tested, the use of a nonspecific ‘control’ protein (i.e. nongluten peptides), and the relevance of acute direct instillation into the duodenum compared with long-term dietary ingestion of the proteins under test. Ciclitira's group is well aware of these problems and seeks to address them. Their observations will require confirmation, but nevertheless they are an important step towards increasing our understanding of the extent of gluten toxicity in coeliac disease.
We now know that gluten is a highly complex mixture of heterogeneous wheat proteins, but that the old classification into gliadins and glutenins is no longer helpful in understanding toxicity in coeliac disease. When classified according to their primary amino acid structure, the gliadins and glutenins, while being very different and heterogeneous, do have partially homologous repetitive sequences that make them similar. The relevance of these to coeliac disease is demonstrated by their ability to be stimulatory in assays of immune activity in coeliac patients.
These findings suggest that the previous belief that glutenins were nontoxic in coeliac disease is incorrect. Not only are glutenin peptides immunostimulatory, similar to gliadin peptides, but evidence is accumulating that glutenin subunits also have direct mucosal toxicity in coeliac disease. One has to conclude, therefore, that components in gluten, both in gliadin and in glutenin, are responsible for precipitating the abnormalities characteristic of coeliac disease.
Conflict of interest
The author was the sole contributor to this leading article.
1. Wieser H. The precipitating factor in coeliac
disease. Balliere's Clin Gastroenterol 1995; 9:191–207.
2. Shewry PR, Tatham AS, Kasarda DD. Cereal proteins and coeliac
disease. In: Marsh MN editor. Coeliac
disease. Oxford: Blackwell Scientific Publications; 1992. pp. 305–348.
3. Shewry PR, Tatham AS, Forde J, Kreis M, Miflin BJ. The classification and nomenclature of wheat gluten
proteins: a reassessment. J Cereal Sci 1986; 4:97–106.
4. van de Kamer JH, Weijers HA, Dicke WK. Coeliac
disease. IV. An investigation into the injurious constituents of wheat in connection with their action on patients with coeliac
disease. Acta Pediatr Scand 1953; 42:223–231.
5. Sheldon W. Coeliac
disease. Lancet 1955; ii:1097–1103.
6. Howdle PD, Corazza GR, Bullen AW, Losowsky MS. Gluten
sensitivity of small intestinal mucosa in vitro: quantitative assessment of histologic change. Gastroenterology 1981; 80:442–450.
7. Maiuri L, Ciacci C, Ricciardelli I, Vacca L, Raia V, Auricchio S, et al. Association between innate response to gliadin
and activation of pathogenic T cells in coeliac
disease. Lancet 2003; 362:30–37.
8. Molberg O, Kett K, Scott H, Thorsby E, Sollid LM, Lundin KF. Gliadin
specific, HLA-DQ2-restricted T cells are commonly found in small intestinal biopsies from coeliac
patients, but not from controls. Scand J Immunol 1997; 46:103–109.
9. van de Wal Y, Kooy YM, Van Veelen PA, Pena SA, Mearin LM, Molberg O, et al. Small intestinal T cells of celiac patients recognise a natural pepsin fragment of gliadin
. Proc Natl Acad Sci USA 1998; 95:10050–10054.
10. Anderson PR, Degano P, Godkin AJ, Jewell DP, Hill AV. In vivo antigen challenge in celiac disease identifies a single transglutaminase-modified peptide as the dominant A-gliadin
T-cell epitope. Nat Med 2000; 6:337–342.
11. Ciclitira PJ, Evans DJ, Fagg NL, Lennox ES, Dowling RH. Clinical testing of gliadin
fractions in coeliac
patients. Clin Sci (London) 1984; 66:357–364.
12. Sturgess R, Day P, Ellis HJ, Lundin KF, Gjertsen HA, Kontakou M, et al. Wheat peptide challenge in coeliac
disease. Lancet 1994; 343:758–761.
13. van de Wal Y, Kooy YM, van Veelen P, Vader W, August SA, Drijfhout JW, et al. Glutenin
is involved in the gluten
-derived mucosal T cell response. Eur J Immunol 1999; 29:3133–3139.
14. Vader W, Kooy Y, van Veelen P, De Ru A, Harris D, Benckhuijsen W, et al. The gluten
response in children with celiac disease is directed toward multiple gliadin
peptides. Gastroenterology 2002; 122:1729–1737.
15. Molberg O, Solheim Flaete N, Jensen T, Lundin KE, Arentz-Hansen H, Anderson OD, et al. Intestinal T-cell responses to high-molecular-weight glutenins in celiac disease. Gastroenterology 2003; 125:337–344.
16. Spaenij-Dekking L, Kooy-Winkelaar Y, van Veelen P, Drijfhout JW, Jonker H, van Soest L, et al. Natural variation in toxicity of wheat: potential for selection of non-toxic varieties for celiac disease patients. Gastroenterology 2005; 129:797–806.
17. Dewar DH, Amato M, Ellis HJ, Pollock EL, Gonzalez-Cinca N, Engel W, et al. The toxicity of high molecular weight glutenin
subunits of wheat to patients with coeliac
disease. Eur J Gastroenterol Hepatol 2006; 18:483–491.