Article Text
PDF
Abstract
Aims To audit the cellular expression of the innate antibacterial enzyme lysozyme in colonic biopsies from a cohort of patients having microscopic colitis (MC—collagenous colitis (CC) or lymphocytic colitis (LC)). Results were compared with those recorded in patients with inflammatory bowel disease (IBD) of the colon (ulcerative colitis (UC) or Crohn's colitis).
Methods Fifty-five consecutive cases having biopsies from the left colon were investigated: 27 MC (14 CC and 13 LC) and 28 IBD (14 UC and 14 Crohn's colitis). Sections were stained with antilysozyme antibody. Twelve cases (3 CC, 3 LC, 3 UC and 3 Crohn's colitis) were challenged with the macrophage marker CD68 (clone PG-M1).
Results In MC, marked lysozyme expression in the colonic crypts was recorded in CC (p<0.05). The number of cases with metaplastic Paneth cells was higher in CC than in LC (p<0.05). In IBD, only active Crohn's colitis displayed marked lysozyme expression in the colonic crypts. Marked lysozyme immune-reactivity in subepithelial lamina propria mucosa (lpm) macrophages was found in LC (LC vs CC p<0.05).
Conclusions The increased production of the antibacterial enzyme lysozyme in CC and LC supports a bacterial aetiology for these two diseases. Lysozyme upregulation in different cell types (epithelial vs macrophages) supports the notion that CC and LC might be two different maladies.
- Microscopic colitis
- lysozyme expression
- collagenous colitis
- diagnosis
- immunocytochemistry
Statistics from Altmetric.com
Introduction
Microscopic colitis (MC) is a clinical syndrome epitomised by chronic watery diarrhoea, normal colonoscopy with characteristic microscopic findings.1 Two main subsets of MC have been recognised, collagenous colitis (CC) and lymphocytic colitis (LC).2 In 1976, Lindström3 reported in a patient having chronic watery diarrhoea and grossly normal colonoscopy, the presence of a subepithelial amorphous band in the colonic mucosa. He called this setting collagenous colitis. In the same year, Freeman et al4 found a similar clinical–histological malady in two patients. Since the observations by Lindström3 and Freeman et al4 included only three cases and the fourth was reported 4 years later,5 CC was considered a rare disease during the 1980s5 and the 1990s.6
Subsequently, new cases were detected and CC is today considered a common disease, particularly at hospitals treating a large number of patients with gastrointestinal (GI) disorders. The reported incidence of CC is approximately 0.6–2.3/100 000 per year with a prevalence of 10–15.7/100 000.7 In patients subjected to endoscopic biopsies due to chronic diarrhoea, the estimated prevalence varies from 0.5% to 42%.8
In 1989, Lazenby et al6 found increased number of intraepithelial lymphocytes in the superficial epithelium of the colon in patients having watery diarrhoea and grossly normal colonoscopy. The term lymphocytic colitis was proposed for this type of MC. Later studies indicate that the incidence of LC is 3.1/100 000 inhabitants per year and the prevalence is 14.4/100 000.9 In Sweden, the incidence of MC is on the rise.10
As early as 1922, Alexander Fleming11 discovered during a deliberate search for antibiotics, a natural defence substance against infection that he called lysozyme. Lysozyme, also known as muramidase or N-acetylmuramide glycanhydrolase, is a family of enzymes (EC 3.2.1.17) which damages bacterial cell walls by catalysing hydrolysis of 1,4-β-linkages between N-acetylmuramic acid and N-acetyl-d-glucosamine residues in a peptidoglycan and between N-acetyl-d-glucosamine residues in chitodextrins.12 13 Lysozyme, encoded by the LYZ gene 13, is an innate enzyme in the GI tract with potent non-immunological antibacterial properties.14 15 This enzyme is upregulated in surface epithelial cells and neck cells, in the oxyntic mucosa of the stomach and in the mucus glands in the antropyloric mucosa.16 Lysozyme is also upregulated in goblet cells (GC) seen in gastric intestinal metaplasia (IM) and Barrett's mucosa.17 Colorectal adenomas have also been found to produce lysozyme.18
In this study, lysozyme expression was assessed in a cohort of patients having chronic watery diarrhoea, normal gross colonoscopy and microscopic colitis (CC or LC) in colonic biopsies. Results were compared with those found in colonic biopsies from patients having inflammatory bowel disease (IBD) in the colon.
Material and methods
From the database of the Department of Pathology, 55 consecutive cases (27 with MC and 28 with IBD of the colon) were retrieved.
Out of 27 cases with MC, 14 had CC and 13 had LC and out of 28 cases with IBD, 14 had ulcerative colitis (UC) and 14 had Crohn's colitis.
Following the recommendations of Ayata et al19 only biopsies from the left colon (descending colon, sigmoid or rectum) were included since one of the goals of this study was to assess lysozyme expression in metaplastic Paneth cells. These cells are not normally present in the left colon.19
Definitions (H&E-stained sections)
Collagenous colitis
CC is defined as a histological constellation characterised by the presence of a thick subepithelial collagenous band, distorted superficial cell arrangement, areas of epithelial denudation and inflammatory cells in superficial epithelium and in lamina propria mucosa (lpm).20
Lymphocytic colitis
LC is defined as the presence of colonic mucosa showing ≥20 intraepithelial lymphocytes per 100 surface epithelial cells. The intraepithelial lymphocytosis concur with chronic inflammation in the lpm,6 mainly lymphocytes and plasma cells.
Ulcerative colitis
The diagnosis of UC is based on a combination of clinical, endoscopic and histological features and the exclusion of an infectious aetiology. The most common histological parameters are continuous, chronic diffuse mucosal inflammation, with or without active inflammation, architectural crypt irregularities, with or without villous formations or mucosal atrophy.21
Crohn's colitis
The diagnosis of Crohn's colitis is based on a combination of several histological features such as: discontinuous chronic mucosal inflammation, with or without active inflammation, aphtoid ulceration, epitheloid cell granuloma22 and focal chronic inflammation in the submucosa.
Paneth cells metaplasia is defined as the presence of Paneth cells at the base of the colonic crypts in biopsies located distal to the transverse colon.19 The cytoplasm of these Paneth cells displayed coarse eosinophilic autofluorescent23 granules in H&E-stained sections.
Sections were stained with H&E and with antihuman lysozyme antiserum (Dako A0099, DAKO, Glostrup, Denmark), dilution 1:1600 and incubation time 5 min on a Leica Bond XT (Leica Microsystems, Wetzlar, Germany).
Twelve cases (3 CC, 3 LC, 3 UC and 3 Crohn's colitis) were also immunohistochemically challenged with CD68, clone PG-M1 (Dako Cytomation, Glostrup, Denmark) at 1:200 dilution. CD68 is a highly glycosylated lysosomal membrane protein that is expressed in cytoplasmic granules, being particularly restricted to macrophages.
Immunohistochemical evaluation
Lysozyme expression was searched for in epithelial cells of the crypts, in metaplastic Paneth cells and in mononuclear cells in lpm. By the aid of an ocular microscale, lysozyme-expressing crypt cells and mononuclear cells in the lpm were topographically recorded in the basal or superficial halves of the crypts. Lysozyme and CD68-expressing mononuclear cells in the lpm underneath the superficial (luminal) epithelium were recorded separately. The presence of Paneth cells was double-checked by observing the respective H&E-stained section in a fluorescent microscope. This simple procedure renders all metaplastic Paneth cells autofluorescent.23
The intensity of lysozyme stain was semi-quantitatively classified into negative (0), slight (+), moderate (++) or marked (+++). The highest degree of lysozyme stain was used to classify cases showing variations in immune-reactivity.
Statistical analysis
The Kruskal–Wallis test was used to compare difference between groups. Statistical significance was defined as p<0.05.
The study was approved by the Regional Ethical Committee.
Results
Collagenous colitis
Crypts
Table 1 shows that in CC cases, moderate-marked lysozyme-expressing GC were found in the deep half of the crypts in 71% (10/14) of patients and marked expression in the superficial half of the crypts in 38% (n=5) of patients (figure 1).
Lysozyme expression in distal colonic biopsies from 55 patients: 27 with MC (14 CC and 13 LC) and 28 with IBD (14 UC and 14 Crohn's colitis)*
Collagenous colitis showing marked lysozyme immune-reactivity in goblet cells in the basal and superficial halves of the crypts (lysozyme immunostain, ×40).
Paneth cell metaplasia
Paneth cells were recorded in 71% (10/14) of the cases; all Paneth cells showed marked lysozyme immune-reactivity. The remaining four cases had no Paneth cells (figure 2).
Collagenous colitis showing marked lysozyme immune-reactivity in metaplastic Paneth cells at bottom of crypts (lysozyme immunostain, ×20).
Lamina propria mucosa
Moderate lysozyme-expressing cells were found in the deep half of the lpm in 36% (5/14) of the cases and moderate-marked-expressing cells in the superficial half of the lpm in 28% (4/14) of the cases. Marked lysozyme-expressing macrophages underneath the superficial epithelium were found in 28% (4/14) of the CC cases (figure 3).
Collagenous colitis. Note marked lysozyme immune-reactivity in lamina propria mucosa macrophages at both sides of the collagenous band (at arrows) (lysozyme immunostain, ×40).
Lysozyme expression in crypts, in metaplastic Paneth cells and in lpm did not correlate with the thickness of the collagenous band.20
Lymphocytic colitis
Crypts
Moderate lysozyme-expressing GC were detected in the deep half of the crypts in 23% (3/13) of the cases (figure 4). In the superficial half of the crypts, only slight lysozyme-expressing GC were found.
Lymphocyic colitis showing moderate lysozyme expression in the basal aspect of the crypts (lysozyme immunostain, ×20).
Paneth cell metaplasia
Paneth cells were recorded in 15% (2/13) of the cases: all Paneth cells displayed marked lysozyme expression. The remaining 11 cases contained no Paneth cells.
Lamina propria mucosa
Moderate lysozyme-expressing cells were seen in the deep half of the lpm in 31% (4/13) of the cases and in the superficial half in 23% (3/13) of the cases. Marked lysozyme-expressing macrophages underneath the superficial epithelium were found in 85% (11/13) of the LC cases (figures 5 and 6).
Lymphocytic colitis showing marked lysozyme expression in lamina propria mucosa macrophages juxtaposing the superficial epithelium (lysozyme immunostain, ×40).
Lymphocytic colitis showing marked CD68 expression in lamina propria mucosa macrophages juxtaposing the superficial epithelium (CD68, clone PG-M1 immunostain, ×40).
Ulcerative colitis (active)
Crypts
Moderate lysozyme-expressing cells were found in the deep half of the crypts in 25% (2/8) of cases with active UC and in the superficial half of the crypts, only slight lysozyme-expressing cells were present.
Paneth cell metaplasia
Paneth cells, all with marked lysozyme immune-reactivity were found in 25% (2/8) of the cases. The remaining six cases showed no Paneth cells.
Lamina propria mucosa
Moderate lysozyme-expressing cells in the deep half of the lpm were found in 12% (1/8) of the cases and in the superficial half in 25% (2/8) of the cases. In 12% (1/8) of the cases, marked lysozyme expression was seen in macrophages underneath the superficial epithelium.
Ulcerative colitis (in remission)
Crypts
All six cases (0%) having UC in remission showed none to slight lysozyme expression in the deep or superficial halves of the crypts.
Paneth cell metaplasia
Paneth cells with marked lysozyme immune-reactivity were found in 16% (1/6) of the cases. The remaining five cases showed no Paneth cells.
Lamina propria mucosa
Moderate lysozyme-expressing cells in the deep and superficial halves of the crypts were found in 17% (1/6) of the cases. In 33% (2/6) of cases, moderate lysozyme-expressing cells were found in macrophages underneath the superficial epithelium.
Crohn's colitis (active)
Crypts
Moderate-marked lysozyme-expressing cells in the deep and superficial halves of the crypts were found in 43% (3/7) of the cases with active Crohn's colitis.
Paneth cell metaplasia
Paneth cells showing marked lysozyme immune-reactivity were recorded in 28% (2/7) of the cases. The remaining five cases showed no Paneth cells.
Lamina propria mucosa
Marked-moderate lysozyme-expressing cells in the deeper half of the lpm were found in 57% (4/7) of the cases, in the superficial half in 43% (3/7) and underneath the superficial epithelium in 43% (3/7) of the cases.
Crohn's colitis (unaffected areas)
Crypts
Moderate lysozyme expression in the deep half of the crypts was recorded in 71% (5/7) of the cases and in the superficial half in 57% (4/7) of the cases.
Paneth cell metaplasia
Paneth cells with marked lysozyme immune-reactivity were found in 14% (1/7) of the cases. The remaining six cases contained no Paneth cells.
Lamina propria mucosa
Moderate lysozyme-expressing cells in the deep half of the lpm were seen in 71% (5/7) of the cases, in the superficial half in 44% (4/7) and underneath the superficial epithelium in 44% (4/7) of the cases.
Comparing lysozyme immune-reactivity in CC, LC and IBD
Crypts
In MC, marked lysozyme immune-reactivity in the deep and superficial halves of the colonic crypts was found only in CC (p<0.05). In IBD, only few cases with active Crohn's colitis displayed marked lysozyme expression in the deep and superficial halves of the colonic crypts.
Paneth cells
The number of cases having lysozyme-rich Paneth cells was significantly higher in CC than in LC (p<0.05). In MC, the number of cases with lysozyme-rich Paneth cells was significantly higher than in IBD (p<0.05).
Lamina propria mucosa macrophages
The number of cases with lysozyme-rich macrophages underneath the superficial epithelium was significantly higher in LC than in CC or IBD (p<0.05).
Discussion
The review of biopsies from the left colon having MC revealed that only CC had marked lysozyme-expressing cells in the colonic crypts. Although the cause for this phenomenon remains elusive, it would appear that a high neo-production of lysozyme is required in cells of the colonic crypts in patients with CC than in those with LC. In IBD, only some cases with active Crohn's colitis showed marked lysozyme-expressing cells in the superficial half of the crypts.
Studying H&E-stained sections from the distal colon, Ayata et al19 found Paneth cell metaplasia in 44% of cases with CC but only in 14% of those with LC. This difference was considered ‘perhaps related to as-yet unidentified difference in the aetiology of these two diseases’.19 We found that the number of cases with lysozyme-rich metaplastic Paneth cells were significantly more common in CC than in LC. In fact, 83% of the MC cases with lysozyme-rich metaplastic Paneth cells were CC.
Tanaka et al24 found in H&E biopsies of distal sites from IBD patients, increased Paneth cell metaplasia. These authors24 suggested that repair and regeneration might be the most potent stimuli causing distal Paneth cell metaplasia in IBD patients. Their studies did not include MC cases. We found in IBD biopsies from the left colon, lysozyme-rich metaplastic Paneth cells in 21% of the cases. It would appear that in patients with CC and to a lower degree with LC and IBD, lysozyme-rich metaplastic Paneth cells evolve in the left colon perhaps to boost lysozyme neo-production.
The lpm macrophages underneath the superficial epithelium demonstrated marked lysozyme neo-production,25 26 mainly in LC patients. The occurrence of these lysozyme-positive/CD68-positive macrophages has not been reported previously in LC.6–8 10 19 27 In addition, we found in CC lysozyme-rich macrophages both above and underneath the collagen band. This novel finding suggests the possibility that macrophages might be engulfing the collagenous material, thus offering an alternative explanation for the disappearance of the collagenous material in subsequent control colonoscopic biopsies in some patients. Although the biological significance of lpm macrophages in CC, LC and IBD remains unclear, the possibility that they might be instrumental in orchestrating defence mechanism(s) against currently elusive pathogens cannot be totally rejected. If this were the case then the lpm macrophage would emerge as a promising candidate to search for specific pathogens in these diseases.
The true aetiology of MC continues to remains unidentified.8–10 27 On the other hand, the increased production of the antibacterial enzyme lysozyme in CC and LC supports a bacterial aetiology for these two diseases. The different mucosal cell types displaying lysozyme upregulation (epithelial vs macrophages) substantiates the notion that CC and LC might be two different maladies.
To conclude, it should be mentioned that MC is not a condition specific to humans; both CC and LC have also been demonstrated in non-human primates having protracted intractable diarrhoeas.28
Take-home messages
The expression of lysozyme, an innate enzyme with potent non-immunological antibacterial properties, was investigated in biopsies from the left colon in patients with microscopic colitis (MC—collagenous colitis (CC) and lymphocytic colitis (LC)) and with inflammatory bowel disease (IBD) of the colon (ulcerative colitis (UC) and Crohn's colitis).
Marked lysozyme immune-reactivity in colonic crypts was recorded in CC and in active Crohn's colitis, whereas LC showed marked lysozyme immune-reactivity in macrophages underneath the superficial epithelium.
The number of cases with metaplastic Paneth cells was higher in CC than in LC.
The increased production of this antibacterial enzyme in CC and LC supports a bacterial aetiology for these two diseases.
Lysozyme upregulation in different mucosal cells (epithelial vs macrophages) supports the notion that CC and LC might be two different maladies.
References
Footnotes
Competing interests None.
Ethics approval This study was conducted with the approval of the Ethics Committee, Karolinska University Hospital.
Provenance and peer review Not commissioned; externally peer reviewed.