Indian Journal of Pathology and Microbiology

: 2013  |  Volume : 56  |  Issue : 4  |  Page : 342--348

Increased accumulation of dendritic cells in celiac disease associates with increased expression of autophagy protein LC3

Paramaguru Rajaguru1, Kim Vaiphei1, Biman Saikia2, Rakesh Kochhar3,  
1 Department of Histopathology, Post Graduate Institute of Medical Education and Research, Chandigarh, India
2 Department of Immunopathology, Post Graduate Institute of Medical Education and Research, Chandigarh, India
3 Department of Gastroenterology, Post Graduate Institute of Medical Education and Research, Chandigarh, India

Correspondence Address:
Kim Vaiphei
Department of Histopathology, PGIMER, Chandigarh - 160 012


Background: Celiac disease (CD) an immune-mediated disorder associates with accumulation of dendritic cell (DC) in duodenal mucosa. Autophagy has recently been implicated in autoantigen formation. However, its role in CD is still unknown. Aim: To examine role of autophagic protein LC3 expressed by activated DC in CD. Materials and Methods : Thirty CD patients were analyzed at initial presentation and after 6 months of gluten-free diet (GFD). Duodenal biopsies were studied for histological changes and CD11c, CD86, and MAP1LC3A expressions by double immunohistochemistry (IHC). Masson«SQ»s trichrome (MT) staining was used to assess basement membrane (BM) thickness and Oil Red O (ORO) staining for mucosal lipid deposit. Polymerase chain reaction (PCR) was performed for HLA-DQ system. Statistical analysis was done using paired and unpaired t test, chi-square test, Fisher«SQ»s exact test, and McNemar-Bowker test. A P-value <0.05 was considered statistically significant. Results: HLA-DQ2 and HLA-DQ8 alleles were present in all studied patients. Increased BM thickness was observed in 63% and 73% had ORO-positive lipid in surface lining epithelium. Pre-treatment biopsies showed increased DCs expressing LC3, which were significantly less in follow-up biopsies. The follow-up biopsies had shown significant reduction in BM thickness and ORO. Conclusion : Histological improvement in duodenal biopsies was associated with reduction in activated DCs expressing autophagic protein, which probably play important role in pathogenesis of an autoimmune disorder like CD.

How to cite this article:
Rajaguru P, Vaiphei K, Saikia B, Kochhar R. Increased accumulation of dendritic cells in celiac disease associates with increased expression of autophagy protein LC3.Indian J Pathol Microbiol 2013;56:342-348

How to cite this URL:
Rajaguru P, Vaiphei K, Saikia B, Kochhar R. Increased accumulation of dendritic cells in celiac disease associates with increased expression of autophagy protein LC3. Indian J Pathol Microbiol [serial online] 2013 [cited 2020 Nov 27 ];56:342-348
Available from:

Full Text


Celiac disease (CD) is an immune-mediated disorder resulting from permanent gluten intolerance which primarily involves the gastrointestinal tract. It affects only predisposed individuals, whose most susceptible genetic features are related to human leucocyte antigen class II, HLA-DQ2 (90%), and HLA-DQ8 (5-10%). [1] CD is characterized by the presence of chronic inflammation of the small bowel mucosa and submucosa and in addition to the classical histological features in CD, the importance of lipid accumulation within the surface lining epithelium [2] and increased BM thickness has also been demonstrated. [3] The causal agent gluten present in intestinal lumen is engulfed by the lamina propria dendritic cell (DCs) for an effective antigen processing and get presented to mucosal T cells. [4] Two distinct population of antigen-presenting cells (APCs) are noted in duodenal mucosa: 80% are CD68-positive macrophages and 20% are CD11c-positive DC. CD11c-positive DCs accumulate in duodenal mucosa of CD patients and expressed activated phenotype detectable by CD86 and DC lysosome-associated membrane protein (DC-LAMP). [5] Many studies have demonstrated altered autophagy property in DCs. [6] A recent study has demonstrated inhibition of autophagy in DCs inhibited by citrullinated peptides in rheumatoid arthritis patients. [7] Autophagic gene ATG16L1 polymorphism rs2241880 (T300A) has been associated with Crohn's disease susceptibility but failed to prove to have an association with CD. [8]

Since autophagy is more of recent concept in various autoimmune disorders and its function in CD is still not well defined and studied, we planned to examine the role of activated DC and their autophagic property in disease pathobiology correlating with HLA-DQ system in these select group of patients.

 Materials and Methods

The study protocol was approved by the Institute Ethics Committee and Informed Consent was obtained from all patients enrolled in the study. This study involved 30 patients suspected to have CD clinically and showing an elevated anti IgA tTG. These patients underwent detail clinical and laboratory investigations at initial presentation and were put on gluten-free diet (GFD). After 6 months of GFD all patients were re-evaluated. Patients who were already on GFD or on antibiotics, who were detected to have gastric and/or duodenal ulcer on endoscopy, and lack of consent were excluded from the study.

All patients underwent upper GI endoscopy and four biopsies from second part of duodenum were obtained at the time of initial visit and also at or after 6 months of follow-up on GFD. Five milliliter of venous blood in an EDTA vacutainer was obtained from these patients for HLA typing. Biopsies from second part of duodenum in ten autopsy patients who died of non-intestinal problem like sudden cardiac death, acute myocardial infarction, etc were included as normal controls. Serial fresh frozen sections were used for H and E, MT, Oil Red O (ORO), and double IHC staining. H and E and MT staining were performed using a standard protocol. For ORO staining, the slides were initially fixed in 70% alcohol for 3 minutes and then put into a glass jar containing ORO solution in 0.5% propylene glycol (Sigma-Aldrich, USA) for 1 hour, wet mounted with glycerine and examined immediately under light microscopy for lipid deposits.

Double IHC staining

Sections were fixed in pre-cooled acetone (4°C) for 20 minutes and kept at room temperature for ~20 minutes for drying. Antibodies such as CD11c (1:100 dilution, mouse monoclonal, Novacastra, New castle, UK), CD86 (1:100 dilution, rabbit monoclonal, Abcam, Cambridge, England), and anti-MAP1LC3A (LC3) antibody (1:100 dilution, mouse polyclonal, Sigma Aldrich, USA) in combinations of CD11c + LC3 and CD86 + LC3 were used. Double immunostaining kit i.e. EnVision G\2 Doublestain System, Rabbit/Mouse (Dako, Denmark) was used appropriately.


All biopsies on H and E stained sections were classified according to modified Marsh criteria. [9] MT stained sections were used to measure the BM using an ocular micrometer (magnification 1000X) and an Olympus CHS microscope (Olympus Optical Ltd, Japan). At least ten readings were taken at various sites choosing the thickest and thinnest ones and the mean of which was calculated. Based on a previous study by Vakiani et al., [3] we have taken 5 μm thickness as cut off value and we divided the biopsies into two groups i.e. those more than 5 μm and those with five 5 μm thickness or less. ORO-stained sections were used to identify accumulation of lipid within surface epithelium, lamina propria, and macrophages.

Double IHC: Percentages of CD11c-positive, CD11c+LC3 double-positive, CD86-positive activated, CD86+LC3 double-positive and LC3-positive DCs were calculated by counting 1000 lamina propria cells (magnification 400x), choosing fields having maximum number of positively stained cells.

HLA-DQ typing: DNA was extracted from the blood sample using QIAamp DNA mini kit (Qiagen, Hilden, Germany) and PCR was performed using HLA-Ready Gene Coeliac Disease (Inno-Train Diagnostik GMBH, Kronberg, Germany) for HLA-DQ A1*0501,05 and DQB1*0201,02 alleles of HLA-DQ2 locus and HLA-DQB1*0302 of HLA-DQ8 locus.

Statistical analysis

Percentage, mean, standard deviation, and median were calculated for all the parameters analyzed in the study. Paired and unpaired t test, chi-square/Fisher's exact test and McNemar-Bowker test were used wherever applicable. A P-value <0.05 was considered statistically significant.


Of the 30 patients we studied, 15 belonged to 13-19 years of age group and the remaining 15 patients were older (age ranged from 20 to 51 years). [Table 1] shows demographic, clinical, and laboratory parameters of CD patients and patients on GFD. Main symptom at presentation was predominantly diarrhea in 16 (53.3%) patients. Patients who were on GFD showed a significant increase in weight and hemoglobin values (P <0.0001) at or after 6 months of follow up.{Table 1}

HLA-DQ system: Well-formed bands of size 140bp for HLA-DQB1 alleles and 200bp for HLA-DQA1 alleles were seen in all the 25 cases studied.

Histological parameters: First set of biopsies showed normal villi in 70% of patients and were classifiable as type 2. In follow-up biopsies, 63% did not show any change in histological type and were labeled as histological non responders. Only 37% patients showed histological improvement and were labeled as partial responders. None of the patients acquired normalization of the biopsy on GFD [Figure 1] and [Figure 2]. [Table 2] shows distributions and statistical analysis of pre and post treatment biopsies according to modified Marsh criteria. [9] All pretreatment biopsies had patchy BM thickening, 19 (63%) had > 5 mm and 11 (37%) had ≤ 5 mm [Table 3]. All biopsies showed significant (P<0.0001) reduction in BM thickness on GFD; however, 23% of biopsies did not reach value of 5 mm or less. Mean BM thickness in histological partial responders and non-responders were 5.69 ± 1.90 mm and 6.07 ± 2.34 mm, respectively. No significant difference in mean value was noted between no responders and partial responders histologically (P = 0.45) and there was no association with clinical, laboratory, or histological parameters.{Figure 1}{Figure 2}{Table 2}{Table 3}

[Table 4] shows ORO staining in pre and post treatment biopsies and control, and 73% showed intracytoplasmic lipid along the surface lining epithelium and brush border, whereas 50% had extracellular lipid deposition in superficial lamina propria and in macrophages. Control biopsies did not reveal any stainable lipid along surface lining but showed extracellular lamina propria and macrophage lipid. The surface epithelial cell lipid accumulation did not show any significant association with clinical, laboratory, or histological parameters. A significantly reduced (P = 0.031) number of follow-up biopsies had epithelial lipid deposition. {Table 4}

IHC staining [Figure 3] and [Figure 4], [Table 5]: Increased accumulations of CD11c-positive DC and CD86-positive activated DC with a mean of 15.99 ± 4.35 and 13.7 ± 3.5 respectively were noted in pretreatment biopsies. Mean CD11c and CD86 was significantly (P<0.0001) higher in CD cases compared to control. Significant (P<0.0001) reductions in CD11c and CD86 were noted in follow-up biopsies. Mean CD11c + LC3 and CD86 +LC3 double-positive DCs were 12.09 ± 3.69 and 10.52±3.59 respectively and were significantly higher than control. CD86 + LC3-positive DC was not observed in control. Significant (P<0.0001) reductions in CD11c + LC3 and CD86 + LC3 were noted in follow-up biopsies. However, there was no significant difference between LC3 in pretreatment biopsies and control (P = 0.52) and also between pre and post treatment biopsies (P = 0.86) [Table 5].{Figure 3}{Figure 4}{Table 5}


CD is one of the major causes of malabsorption in northern part of India with a prevalence of 1 in 96. [10] In our study, nine patients presented in third decade of life or beyond, and three patients were in fifth decade of life or beyond. These profiles of age distributions imply importance of suspecting and diagnosing CD in elderly individual. This ice berg disease can have wide spectrum of clinical presentations. Bhadada et al. [11] have shown CD to be a common cause (15.3%) for short stature due to growth retardation in young children. Atypical clinical presentations were seen in 20% of our patients; short stature was one of the common presenting features in adolescence thereby emphasizing importance of screening and excluding CD in this group of patients. We observed anemia to be the most common extra intestinal manifestation and the reported prevalence of anemia ranged from 12% to 69%. [12] The characteristic feature of CD-associated iron deficiency anemia (IDA) is its refractoriness to oral iron therapy. [13] We also documented anemia in 87% of our patients, all with IDA. CD has also been observed frequently with other autoimmune disorders. [14] Ten of our patients had associated autoimmune disorders like type I diabetes mellitus (DM) in six which was the most common disorder. Hence, all CD patients need to be investigated for any kind of other associated autoimmune disorders or vice versa. Clinical responses to GFD may be variable, some may require more than 12 months for a complete clinical response while a few may be non-responsive at all. [15] We observed complete clinical response in 57% of our patients after 6 months of GFD. A good correlation was observed between the HLA-DQ2 positivity and other parameters in confirming the diagnosis of CD in the present study. In spite of a typical or atypical clinical presentation, high level of sensitivity of HLA-DQ2 and HLA-DQ8 alleles helps in identifying CD patient. Due to the associated high percentage of HLA-DQ alleles expressions even by the first degree relative of CD patients, who are at risk of developing the disease, makes it an important laboratory parameter to identify this group of potentially high risk individuals. [16]

In spite of significant histological improvements noted, none of our patients acquired total normalization in mucosal architecture after 6 months of GFD. Although majority of our patients showed good clinical response, 63% of them failed to show any improvement in histological types, similar to the observation made by Lanzini et al. [17] And they concluded in their study that reversal to normality of small intestinal histology was rarely achieved in CD patients during the course of prolonged and strict adherence to GFD despite symptomatic remission. CD has been known to have variable degree of BM thickening. Vakiani et al. [3] classified CD patients having BM thickness more than 5 mm as collagenous sprue, but Rubio-Tapia et al. [18] used a minimum thickness of 20 mm for the diagnosis of collagenous sprue. More than 60% of our patients had five or more mm thick BM, which were similar to observations made by Brown et al. [19] in 64% of their patients. Vakiani et al. [3] and Rubio-Tapia et al. [18] observed reduction in BM thickness in CD patients who were either on GFD or immunomodulators. Many of patients showed a significant reduction in BM thickness, but eight (27%) of the duodenal biopsies showed more than 5 mm thick BM. More importantly, four of these patients had partial clinical improvement. Possibly these four patients belonged to the group of CD patients who require immunomodulators. The BM status in our patients did not exhibit any association with other histological parameters but correlated with the clinical response; BM thickness may be used as a parameter assessing severity of the disease process, response to GFD, and also to identify patients who will be requiring immunomodulators.

We have also studied the patterns of lipid deposition in the duodenal mucosa using ORO and observed that 73% of the biopsies showed ORO-positive lipid within surface lining epithelium. Significant reduction in surface epithelial lipid deposition was noted in patients on GFD. Our observations were similar to the reports by Schenk et al. [2] and Variend et al. [20] Surface lining epithelial lipid deposition may be an indicator for CD because of its high sensitivity compared to other enteropathies. We failed to observe any significant association with histological features but a classic ORO-positive staining pattern may be a softer cue for CD.

DC are specialized APCs found throughout the intestine, including in lamina propria of intestine, lymphoid follicles, Peyer's patches, and mesenteric lymph nodes. The gluten antigenic proteins are taken up by DCs, processed, and then presented to T cells. CD11c DC is a major APC in normal duodenal mucosa and acquisition of antigen presenting capacity associated with CD86 expression had been observed in CD. [14] All our biopsies showed an increased CD11c and CD86 DCs compared to normal control. Raki et al. [5] had shown accumulated CD11c DCs expressing co-stimulatory molecule CD86 in CD patients' duodenal biopsies. Di Sabatino et al. [21] documented a significantly higher CD11c-, CD86-, and CD123-positive DCs in duodenal biopsy specimens. We also observed a significant decrease in CD11c and CD86 in follow-up post treatment biopsies done at or after 6 months of GFD. Our observation is comparable to the results published by Di Sabatino et al. [21] and Raki et al. [5] in follow-up biopsies. Hence, CD11c-positive DCs appear to be the pioneer APC in CD, which acquires activation marker CD86 after exposure to gluten.

Double IHC staining done in pretreatment biopsies revealed that majority of CD11c- and CD86-positive DCs expressed LC3, which were significantly higher than control and were significantly decreased in follow-up biopsies. Interestingly CD86 DC in control lacked LC3. Increased expressions for autophagy-related proteins have been demonstrated in various human diseases including Alzheimer's disease. [22] Post treatment biopsies exhibited reduction in LC3 in DCs positive for CD11c and CD86. Similar to the histological features, these reductions in number of positive DCs did not reach baseline values. Autophagy has been emphasized as a double-edged sword in many disease conditions and is believed to play important roles in degradation of aggregated unwanted cellular proteins. [22],[23],[24] In many autoimmune disorders, there are considerable evidences showing deregulated autophagy system as the underlying culprit in production of autoantigen. [8] These observations may be used to emphasize importance of a defective autophagy system in context to CD as well and may be the underlying abnormal pathway in handling gluten protein. Hence, the hypothetical proposal will be put forward as follows i.e. a defective autophagy system in DCs may result in accumulation of undigested intracellular gluten proteins which are bound to autophagic proteins. This in turn will enhance an unwanted but a persistent stimulation of the APCs in order to compensate the defective mechanism, which ultimately results in accumulation and aggregation of the activated DCs in duodenal mucosa and is associated with an uncontrolled mucosal inflammation.

To conclude, our findings suggest important role of autophagy system in disease pathobiology in the development of CD and the observations made by us may have important implication in disease management. However, more study is required to support our observations in more number of randomized controlled patients. Our study further highlights the importance of studying basement membrane status in duodenal mucosa in assessing response to GFD.


We like to thank the Indian Council of Medical Research, New Delhi, India, for their partial monetary support. We also like to express our gratitude to all patients who were willing to be a part of the study protocol.


1Husby S, Koletzko S, Korponay-Szabo IR, Mearin ML, Phillips A, Shamir R, et al. European Society for Pediatric Gastroenterology, Hepatology, and Nutrition guidelines for the diagnosis of coeliac disease. J Pediatr Gastroenterol Nutr 2012;54:136-60.
2Schenk EA, Samloff IM, Klipstein FA. Morphologic characteristics of jejunal biopsy in celiac disease and tropical sprue. Am J Pathol 1965;47:765-81.
3Vakiani E, Arguelles-Grande C, Mansukhani MM, Lewis SK, Rotterdam H, Green PH, et al. Collagenous sprue is not always associated with dismal outcomes: A clinicopathological study of 19 patients. Mod Pathol 2010;23:12-26.
4Heyman M, Menard S. Pathways of gliadin transport in celiac disease. Ann N Y Acad Sci 2009;1165:274-8.
5Raki M, Tollefsen S, Molberg O, Lundin KE, Sollid LM, Jahnsen FL. A unique dendritic cell subset accumulates in the celiac lesion and efficiently activates gluten-reactive T cells. Gastroenterology 2006;131:428-38.
6Zhou D, Kang KH, Spector SA. Production of interferon alpha by human immunodeficiency virus type 1 in human plasmacytoid dendritic cells is dependent on induction of autophagy. J Infect Dis 2012;205:1258-67.
7Ireland JM, Unanue ER. Autophagy in antigen-presenting cells results in presentation of citrullinated peptides to CD4 T cells. J Exp Med 2011;208:2625-32.
8Dema B, Fernandez-Arquero M, Maluenda C, Polanco I, Figueredo MA, de la Concha EG, et al. Lack of association of NKX2-3, IRGM, and ATG16L1 inflammatory bowel disease susceptibility variants with celiac disease. Hum Immunol 2009;70:946-9.
9Oberhuber G, Granditsch G, Vogelsang H. The histopathology of coeliac disease: Time for a standardized report scheme for pathologists. Eur J Gastroenterol Hepatol 1999;11:1185-94.
10Makharia GK, Verma AK, Amarchand R, Bhatnagar S, Das P, Goswami A, et al. Prevalence of celiac disease in the northern part of India: A community based study. J Gastroenterol Hepatol 2011;26:894-900.
11Bhadada SK, Bhansali A, Kochhar R, Menon AS, Sinha SK, Dutta P, et al. Does every short stature child need screening for celiac disease? J Gastroenterol Hepatol 2008;23:e353-6.
12Bottaro G, Cataldo F, Rotolo N, Spina M, Corazza GR. The clinical pattern of subclinical/silent celiac disease: An analysis on 1026 consecutive cases. Am J Gastroenterol 1999;94:691-6.
13Varma S, Malhotra P, Kochhar R, Varma N, Kumari S, Jain S, et al. Celiac disease presenting as iron-deficiency anemia in northern India. Indian J Gastroenterol 2001;20:234-6.
14Barton SH, Murray JA. Celiac disease and autoimmunity in the gut and elsewhere. Gastroenterol Clin North Am 2008;37:411-28.
15Dewar DH, Donnelly SC, McLaughlin SD, Johnson MW, Ellis HJ, Ciclitira PJ. Celiac disease: Management of persistent symptoms in patients on a gluten-free diet. World J Gastroenterol 2012;18:1348-56.
16Thapa BR, Rawal P, Sapra B, Vaiphei K, Nain CK, Singh K. Familial prevalence of celiac disease. J Trop Pediatr 2011;57:45-50.
17Lanzini A, Lanzarotto F, Villanacci V, Mora A, Bertolazzi S, Turini D, et al. Complete recovery of intestinal mucosa occurs very rarely in adult coeliac patients despite adherence to gluten-free diet. Aliment Pharmacol Ther 2009;29:1299-308.
18Rubio-Tapia A, Talley NJ, Gurudu SR, Wu TT, Murray JA. Gluten-free diet and steroid treatment are effective therapy for most patients with collagenous sprue. Clin Gastroenterol Hepatol 2010;8:344-9.
19Brown IS, Smith J, Rosty C. Gastrointestinal pathology in celiac disease: A case series of 150 consecutive newly diagnosed patients. Am J Clin Pathol 2012;138:42-9.
20Variend S, Placzek M, Raafat F, Walker-Smith JA. Small intestinal mucosal fat in childhood enteropathies. J Clin Pathol 1984;37:373-7.
21Di Sabatino A, Pickard KM, Gordon JN, Salvati V, Mazzarella G, Beattie RM, et al. Evidence for the role of interferon-alfa production by dendritic cells in the Th1 response in celiac disease. Gastroenterology 2007;133:1175-87.
22Ma JF, Huang Y, Chen SD, Halliday G. Immunohistochemical evidence for macroautophagy in neurones and endothelial cells in Alzheimer's disease. Neuropathol Appl Neurobiol 2010;36:312-9.
23Tung YT, Wang BJ, Hu MK, Hsu WM, Lee H, Huang WP, et al. Autophagy: A double-edged sword in Alzheimer's disease. J Biosci 2012;37:157-65.
24Chen N, Karantza V. Autophagy as a therapeutic target in cancer. Cancer Biol Ther 2011;11:157-68.