Immunoproliferative Small Intestinal Disease Associated with Campylobacter jejuni
http://www.100md.com
《新英格兰医药杂志》
ABSTRACT
Background Immunoproliferative small intestinal disease (also known as alpha chain disease) is a form of lymphoma that arises in small intestinal mucosa-associated lymphoid tissue (MALT) and is associated with the expression of a monotypic truncated immunoglobulin heavy chain without an associated light chain. Early-stage disease responds to antibiotics, suggesting a bacterial origin. We attempted to identify a causative agent.
Methods We performed polymerase chain reaction (PCR), DNA sequencing, fluorescence in situ hybridization, and immunohistochemical studies on intestinal-biopsy specimens from a series of patients with immunoproliferative small intestinal disease.
Results Analysis of frozen intestinal tissue obtained from an index patient with immunoproliferative small intestinal disease who had a dramatic response to antibiotics revealed the presence of Campylobacter jejuni. A follow-up retrospective analysis of archival intestinal-biopsy specimens disclosed campylobacter species in four of six additional patients with immunoproliferative small intestinal disease.
Conclusions These results indicate that campylobacter and immunoproliferative small intestinal disease are associated and that C. jejuni should be added to the growing list of human pathogens responsible for immunoproliferative states.
Immunoproliferative small intestinal disease, also known as alpha chain disease, is a mucosa-associated lymphoid-tissue (MALT) lymphoma characterized by infiltration of the bowel wall with a plasma-cell population that secretes a monotypic, truncated immunoglobulin heavy chain lacking an associated light chain.1,2,3 Lymphoid infiltration leads to malabsorption and protein-losing enteropathy. The disease can cause a spectrum of histopathological changes, ranging from seemingly benign lymphoid infiltration to malignant diffuse large-B-cell lymphoma.
Since its initial description,1 immunoproliferative small intestinal disease has largely been reported in the Mediterranean basin, the Middle East, the Far East, and Africa. In the Middle East, the most common site of extranodal lymphoma is the gastrointestinal tract: immunoproliferative small intestinal disease accounts for approximately one third of gastrointestinal lymphomas.4 The restricted geographic distribution of the disease has led to the hypothesis that environmental factors may have a pathogenic role.
Remarkably, early-stage immunoproliferative small intestinal disease typically responds to antibiotics, suggesting that it may be triggered by bacterial infection.5,6 However, previous attempts to identify a causative agent with the use of standard culture methods have failed.5,6,7 Immunoproliferative small intestinal disease shares histopathological features with gastric MALT lymphoma associated with Helicobacter pylori infection,2,3 and a case report suggested that H. pylori may be a potential causative agent.8 However, this proposal was not supported by the results of a subsequent retrospective study of 21 patients with immunoproliferative small intestinal disease.9 We used a molecular strategy similar to that used to identify the causative agents of bacillary angiomatosis (Bartonella henselae)10 and Whipple's disease (Tropheryma whipplei),11 in order to determine whether immunoproliferative small intestinal disease could be linked to specific bacterial species.
Case Reports
Index Patient
A 45-year-old woman from Cameroon was hospitalized for a 12-month-long illness characterized by chronic diarrhea and wasting. Histologic examination of endoscopic biopsy specimens from her duodenum and proximal jejunum revealed massive infiltration of the lamina propria, with a dimorphic population of plasma cells located in the superficial mucosa and centrocyte-like lymphocytes proliferating deeper in the mucosa, mostly around crypts (Figure 1A, Figure 1B, Figure 1C, and Figure 1D). Infiltration of the lamina propria was less pronounced in antral-biopsy specimens.
Figure 1. Histologic and Immunohistologic Studies of Endoscopic-Biopsy Specimens from the Index Patient.
In Panel A, staining with hematoxylin and eosin shows lymphoid infiltration of the lamina propria in a section of jejunum (x50). In Panel B, a section of the antral region of the stomach stained with hematoxylin and eosin shows lymphoid infiltration of the lamina propria (x100). In Panel C, which shows a higher-power view of the specimen in Panel A, the typical lymphoplasmacytic features of the superficial mucosa infiltrate are evident (hematoxylin and eosin, x1000). In Panel D, a higher-power view of the specimen shown in Panel A reveals centrocyte-like cells infiltrating the crypt epithelium (hematoxylin and eosin, x400). Sections of jejunum (Panel E) and stomach (Panel F) stained with primary antibodies directed against the B-cell marker CD20 (appears brown when stained with enzyme-linked secondary antibodies) and counterstained with hematoxylin (x100) show CD20-positive centrocyte-like lymphocytes pervading the lamina propria surrounding crypts (bottom insets show higher-power views of boxed areas, x300). CD20-positive centrocyte-like lymphocytes infiltrate the crypt epithelium and produce characteristic lymphoepithelial lesions (arrows in Panels E and F). In contrast, the superficial plasma-cell infiltrate is negative for CD20. The inset in the upper right-hand portion of Panel F shows a section stained with KL1 (which recognizes cytokeratin) and counterstained with hematoxylin (x300). Cross-sectioned crypts contain brown, cytokeratin-positive epithelial cells. Jejunal sections (x400) were incubated with monoclonal antibodies that recognize the heavy chain (Panel G), heavy chain (Panel H), light chain (Panel I), or light chain (Panel J) and were counterstained with hematoxylin. The results establish that the infiltrating lymphoid-cell population (lymphoplasmacytes and centrocyte-like cells) expresses the heavy chain without its associated light chain.
Immunohistochemical studies of these biopsy specimens showed that the plasma cells were negative for CD20, whereas the centrocyte-like lymphocytes were positive for CD20 (Figure 1E and Figure 1F). The cytoplasm of both plasma cells and centrocyte-like lymphocytes showed intense staining for immunoglobulin heavy chain, with no detectable expression of light chains or surface immunoglobulin in these cells (Figure 1G, Figure 1H, Figure 1I, and Figure 1J). Centrocyte-like lymphocytes were responsible for the lymphoepithelial lesions that are typical of MALT lymphomas (Figure 1D, Figure 1E, and Figure 1F).
Staining of gastric- and intestinal-biopsy specimens with Giemsa and silver stains and serologic assays were all negative for H. pylori. Tissue and stool cultures were negative for salmonella, yersinia, shigella, and campylobacter. An enzyme-linked immunosorbent assay and Western blot assays of serum were positive for Campylobacter jejuni (data not shown).
Analysis of blood samples disclosed 7560 lymphocytes per cubic millimeter, with the vast majority having morphologic features of lymphoplasmacytes. Analysis with the use of a fluorescence-activated cell sorter established that 82 percent of peripheral-blood lymphocytes were CD19+, CD20–, CD38+, CD138– B cells. Immunocytochemical studies of the sorted cells revealed high levels of cytoplasmic heavy chains without detectable light chains or surface immunoglobulin. Bone marrow biopsy disclosed extensive infiltration with lymphoplasmacytes. Evidence of the clonality of this population was provided by Southern blot analysis of peripheral-blood lymphocyte DNA, which showed discrete rearrangements of the genes for the immunoglobulin heavy-chain and light-chain loci (data not shown). The serum IgA level was increased to 10.9 g per liter (normal range, 0.76 to 3.90), whereas IgG and IgM levels were decreased (to 5.81 and 0.22 g per liter, respectively). Immunoelectrophoresis of serum, aspirated jejunal luminal contents, and urine from the patient showed a prominent arc of heavy chain precipitin (Figure 2A). Western blot analysis of serum revealed the presence of a truncated monotypic 1 heavy chain.
Figure 2. Resolution of Immunoproliferative Small Intestinal Disease in the Index Patient after Antimicrobial Therapy.
Panel A shows the immunoelectrophoresis of serum before and after antibiotic treatment; the arrow points to the prominent arc of heavy chain precipitin observed before treatment. Panel B shows the time course of the disappearance of the CD19+, CD38+, and heavy chain (HC)–positive leukemic cells, as assessed by analysis of peripheral-blood lymphocytes with the use of a fluorescence-activated cell sorter. Panel C shows the gradual decrease in the serum level of total IgA.
On the basis of these clinical, histopathological, and immunologic findings, a diagnosis of immunoproliferative small intestinal disease was made. Given the gastric extension of the disease, the phenotypic similarities between the disease and H. pylori–associated gastric MALT lymphoma,2,3 and a previous case report of regression of immunoproliferative small intestinal disease in a patient after the eradication of H. pylori infection,8 triple antimicrobial therapy was initiated (1 g of amoxicillin twice daily, 500 mg of metronidazole twice daily, and 500 mg of clarithromycin twice daily) in combination with a proton-pump inhibitor (20 mg of omeprazole twice daily). The diarrhea subsided within a week, the leukemic component disappeared within 10 weeks (Figure 2B), and there was progressive resolution of the paraproteinemia (Figure 2A and Figure 2C) and intestinal lymphoplasmacytic infiltrate (data not shown). Treatment was stopped after five months because the patient was asymptomatic, the leukemic and serum paraprotein abnormalities had disappeared, and lymphoplasmacytic infiltration of the intestine was barely detectable in follow-up endoscopic biopsy specimens from the jejunum. One year after the diagnosis had been established, the results of clinical examination and laboratory studies were normal.
Six Additional Patients
Immunoproliferative small intestinal disease is rare. Frozen intestinal samples from other patients with untreated immunoproliferative small intestinal disease were not available to us. Therefore, we retrieved archival paraffin-embedded excisional jejunal-biopsy specimens obtained at the time of laparoscopy 8 to 27 years earlier from six additional patients in whom immunoproliferative small intestinal disease had been diagnosed at a single hospital according to established histopathological and immunologic criteria.1,5,6,12
Methods
Tissue Samples
We analyzed tissue specimens from six sources. Frozen and formalin-fixed, paraffin-embedded specimens were obtained from gastric, duodenal, and jejunal biopsies performed endoscopically in the index patient one day before and eight days after the initiation of antimicrobial therapy. Archival jejunal-biopsy specimens, which had been fixed in Bouin's fluid and embedded in paraffin, were obtained from six additional patients who had received a diagnosis of immunoproliferative small intestinal disease. Formalin-fixed, paraffin-embedded jejunal-biopsy specimens were obtained endoscopically from a patient infected with human immunodeficiency virus (HIV) who had acute C. jejuni enteritis. Formalin-fixed, paraffin-embedded stomach samples were obtained from a patient with H. pylori gastritis. Frozen duodenal-biopsy specimens were obtained from 10 patients with chronic diarrhea of unknown origin (as defined by negative cultures for known enteropathogens). Formalin-fixed, paraffin-embedded duodenal-biopsy specimens were obtained from 10 patients who were being evaluated for anemia.
Bacterial Strains and Antibodies
We obtained C. jejuni reference strain CIP 70.2, H. pylori reference strain CIP 103995, and Escherichia coli reference strain CIP 7624 from the Institut Pasteur collection. Two mouse monoclonal antibodies were used for immunohistochemical studies: IgG2b NCL-C-JEJUNI (Novocastra), which recognizes a flagellar immunologic determinant common to C. jejuni and H. pylori,13 and IgM CP1/IIG10 (Biotrend), which is directed against a 43-kD, non–surface-associated H. pylori antigen that is not detectable in other helicobacter species or unrelated bacteria.
Polymerase Chain Reaction
Universal primers PB (5'TAACACATGCAAGTCGAACG3') and DG74 (5'AGGAGGTGATCCAACCGCA3') were used to amplify bacterial 16S ribosomal DNA (rDNA).14 Established primers directed at C. jejuni (CCCJ609F 5'AATCTAATGGCTTAACCATTA3' and CCJ1442R 5'GTAACTAGTTTAGTATTCCGG3')15 were used to generate an 852-bp 16S rDNA amplicon. Primers specific for H. pylori (5'fla 5'CCACGGTTAAAGCGTCTATTGG3' and 3'fla 5'GATCGCATTAGTCAACCTCCCG3')16 were used to amplify a 401-bp fragment from the flaB gene. Two pairs of primers — 72/96F (5'ACAGGAAGAAGCTTGCTTCTTTGC3') and 455/477R (5'GAGCAAAGGTATTAACTTTACTCCC3') and 455/477F (5'GGGAGTAAAGTTAATACCTTTGCTC3') and 1000/1025R (5'ACATTCTCATCTCTGAAAACTTCCG3') — were designed to amplify 16S rDNA from Enterobacteriaceae. Cycling conditions for all polymerase chain reactions (PCRs) were as follows: 7 minutes at 95°C, 35 cycles of denaturation for 30 seconds at 95°C, annealing for 20 seconds at 55°C, and 2 minutes of extension at 72°C and then 10 minutes at 72°C.
The specificity of the PCR assays was initially established with the use of DNA extracted from the three reference strains listed above (Table 1). For the index patient, amplicons from 16S ribosomal genes were subcloned, and the nucleotide sequences of 12 randomly chosen clones were determined on both strands. All 16S rDNA sequences that were at least 99 percent homologous were considered to belong to the same species.17,18
Table 1. Results of Polymerase-Chain-Reaction (PCR) Assays of Biopsy Specimens from the Index Patient with Immunoproliferative Small Intestinal Disease and Control Samples.
Fluorescence in Situ Hybridization
Cy3-tagged 5'ATTACTGAGATGACTAGCACCCC3' (Cj-490) was used to probe for the presence of C. jejuni,19 whereas Cy3-tagged 5'CACACCTGACTGACTATCCCG3' (Hpy-1) was used to detect H. pylori.20 Deparaffinized, formalin-fixed sections that were 5 μm thick were incubated for 10 minutes in lysozyme buffer (10 mg of lysozyme per milliliter , 100 mM TRIS–hydrochloric acid , and 50 mM EDTA) at room temperature and then for 2 hours at 35°C in hybridization chambers with the oligonucleotide probe (diluted to a final concentration of 4.5 ng per microliter in a solution of 30 percent formamide, 0.9 M sodium chloride, 20 mM TRIS–hydrochloric acid , and 0.01 percent sodium dodecyl sulfate). Tissue sections were washed twice (for 15 minutes at 37°C per cycle) in wash buffer (0.9 M sodium chloride, 20 mM TRIS–hydrochloric acid , 5 mM EDTA, and 0.01 percent sodium dodecyl sulfate) and once in distilled water, dried at room temperature, mounted, and observed by means of epifluorescence microscopy.
All tissue samples studied were collected as part of routine care. The National Ethics Committee of France requires neither approval by institutional review boards nor informed consent for this type of research to be conducted.
Results
Identification of C. jejuni DNA
DNA was extracted from frozen jejunal- and gastric-biopsy specimens from the index patient. Initial PCR assays of these DNAs were performed with the use of well-established universal primers known to generate amplicons from the 16S rDNA genes from most phyla in the bacteria superkingdom.14 Amplicons were produced in assays in which the DNA template was extracted from frozen jejunal samples harvested one day before treatment was initiated. In contrast, no amplicons were produced with the use of DNA prepared from material harvested eight days after therapy was started.
The nucleotide sequences of 8 of 12 cloned amplicons generated from the pretreatment preparation of jejunal DNA were 99.6 percent identical to 16S rDNA amplified from DNA of the C. jejuni reference strain. Computer searches of GenBank with the use of the BLAST program (http://www.ncbi.nlm.nih.gov/blast/) and the 16S rDNA data base (http://greengenes.llnl.gov/16S/) showed that the four remaining sequences were from abiotrophia, neisseria, lactococcus, and haemophilus. Members of these four genera are part of the normal oropharyngeal microbiota in humans.
Follow-up PCR studies with the use of primers directed at C. jejuni 16S rDNA yielded amplicons of the expected size from antral, jejunal, and fecal DNAs prepared from material obtained before the initiation of antimicrobial therapy (Table 1). Assays of the same DNAs with the use of primers directed at H. pylori 16S rDNA or primers that recognize 16S rDNA from members of the Enterobacteriaceae family were all negative (Table 1). In addition, the PCR assay was negative with the use of primers directed at C. jejuni and DNA prepared from jejunal- and gastric-biopsy specimens obtained from the index patient eight days after the initiation of antibiotic treatment and from DNA extracted from frozen proximal intestinal-biopsy specimens from 10 control patients with diarrhea of unknown origin (i.e., not associated with cultivatable enteropathogens) (Table 1).
The results of sequencing 16S rDNA amplicons obtained with universal bacterial 16S rDNA primers and the C. jejuni–directed PCR assays of endoscopic biopsy specimens from the index patient and controls provided initial evidence of an association between this campylobacter species and immunoproliferative small intestinal disease.
Detection of Campylobacter jejuni in Small Intestinal Tissue
To provide further evidence of the presence of C. jejuni in the index patient, we developed a fluorescence in situ hybridization (FISH) assay using an oligonucleotide probe directed at C. jejuni 16S rRNA. The sensitivity and specificity of the FISH assay were established as follows: staining with the Cy3-labeled oligonucleotide was readily detectable in a smear of cultured C. jejuni (Figure 3A), intraluminal C. jejuni were detected in a jejunal-biopsy specimen from an HIV-infected patient with culture-documented C. jejuni enteritis (Table 2), and no signal was noted in gastric-biopsy specimens from a patient with H. pylori gastritis (Table 2 and Figure 3B).
Figure 3. Fluorescence in Situ Hybridization (FISH) and Immunohistochemical Analysis of Tissue Specimens.
Panel A shows a positive control for FISH: the Cy3-labeled oligonucleotide probe directed against Campylobacter jejuni and a smear of cultured C. jejuni type strain (x400). Panels B and C show the specificity of the FISH assay; the C. jejuni–directed probe or a Helicobacter pylori–specific probe and sections of a gastric-biopsy specimen from a patient with H. pylori–associated gastritis were used for FISH (x400). The C. jejuni probe does not recognize H. pylori (Panel B), whereas the H. pylori probe produces a prominent signal (Panel C). Panels D, E, F, and G show the results of FISH analysis with the use of the C. jejuni probe and sections prepared from biopsy specimens of the proximal small intestine (Panel D) and antrum (Panel F) obtained from the index patient before the initiation of antimicrobial therapy. After hybridization, sections were demounted and stained with hematoxylin and eosin (Panels E and G). Prominent signals are seen in the lamina propria and surrounding blood vessels (arrows in Panels D, E, F, and G) (x400). Panels H, I, J, K, L, and M show immunohistochemical analysis of jejunal sections stained with NCL-C-JEJUNI monoclonal antibody (appears brown when stained with enzyme-linked secondary antibodies) and hematoxylin (x400). The arrows point to immunolabeled material shown at a higher magnification in the bottom insets in the six panels (x800). Panel H shows a specimen from a patient infected with human immunodeficiency virus who had acute C. jejuni enteritis. Bacteria are present in the lumen and are associated with enterocytes. Panel I shows a specimen from the index patient. Panels J, K, and L show archival biopsy specimens from three patients with immunoproliferative small intestinal disease (Patients 1, 2, and 3 in Table 2) that are positive for C. jejuni on FISH. The top insets in these three panels show sections with intraluminal immunolabeled bacteria (x400). Panel M shows a biopsy specimen from a patient with immunoproliferative small intestinal disease (Patient 6 in Table 2) that was negative for C. jejuni on FISH.
Table 2. Results of Fluorescence in Situ Hybridization and Immunohistochemical Assays of Biopsy Specimens from the Index Patient, Six Other Patients with Immunoproliferative Small Intestinal Disease, and Controls.
Subsequent FISH analysis with the C. jejuni probe revealed the organism in sections of jejunal- and stomach-biopsy specimens obtained from the index patient before the initiation of antimicrobial therapy. The organism was most prominent in the lamina propria, often in the vicinity of capillaries (Figure 3D, Figure 3E, Figure 3F, and Figure 3G), a finding consistent with the observed tropism of this organism in an animal model of infection.21 A control H. pylori–specific FISH probe (Figure 3C) did not produce a signal (Table 2), a finding that is consistent with the negative Giemsa and silver staining and the PCR results (Table 1).
Finally, immunohistochemical methods were used to confirm the presence of C. jejuni in the index patient. Two well-characterized, commercially available preparations of monoclonal antibody were used: one recognizes a flagellar immunologic determinant shared by C. jejuni and H. pylori (NCL-C-JEJUNI), and another is specific for an H. pylori epitope (CP1/IIG10). Control experiments that used sections of jejunum from an HIV-infected patient with culture-positive acute C. jejuni enteritis disclosed a prominent signal with NCL-C-JEJUNI and no signal with the H. pylori–specific monoclonal antibody (Figure 3H and Table 2). Assays of sections prepared from jejunal- and gastric-biopsy specimens obtained before the index patient started antimicrobial therapy disclosed readily detectable signals in the lamina propria with NCL-C-JEJUNI but not with the H. pylori–specific antibody (Figure 3I and Table 2). Together, these results support the conclusion that C. jejuni was present at the sites of gut abnormalities in the index patient and rule out the possibility of concomitant H. pylori infection.
Analysis of Archival Tissues
We were unable to recover amplifiable DNA from any of the archival samples from six patients with immunoproliferative small intestinal disease — that is, the PCRs for the human -actin gene as well as for C. jejuni 16S rDNA were all negative. This failure most likely reflects the method of fixation together with the age of the samples. However, because material fixed with Bouin's fluid is known to be suitable for FISH,22,23 these biopsy specimens were analyzed with the use of three probes: a universal bacterial 16S rDNA oligonucleotide (positive control for the presence of bacteria), the probe directed at C. jejuni, and the specific probe for H. pylori. The bacterial FISH probe was positive in samples from five of the six patients (Patients 1 through 5 in Table 2). The C. jejuni–directed probe produced a positive signal in three of the six biopsy specimens from these patients (Patients 1, 2, and 3 in Table 2).
In addition, the two monoclonal antibodies were used to confirm that C. jejuni immunologic determinants were present in all three FISH-positive samples (Figure 3J, Figure 3K, and Figure 3L), as well as in one FISH-negative sample (Figure 3M and Table 2). No signal was detected with these monoclonal antibodies in the 10 control duodenal samples. Moreover, the H. pylori–specific FISH probe and monoclonal antibody did not detect this species in any of the archival jejunal-biopsy specimens from the six patients with immunoproliferative small intestinal disease (Table 2).
Discussion
Previous attempts to link a bacterial species to immunoproliferative small intestinal disease with the use of culture-based approaches have failed.5,6,7 We used a molecular approach to establish a link between immunoproliferative small intestinal disease and C. jejuni infection in an index patient. This association is based on three observations. First, a PCR assay with the use of universal bacterial 16S rDNA primers and DNA templates prepared from biopsy specimens of the proximal small intestine obtained before the initiation of antimicrobial treatment in the index patient yielded amplicons encompassing the entire 16S rDNA gene that were identified as C. jejuni by sequencing. No other enteropathogen was detected by PCR. Second, FISH and immunohistochemical analyses showed C. jejuni in diseased biopsy specimens of the small intestine. Third, the eradication of C. jejuni with antimicrobial therapy was associated with rapid remission of the immunoproliferative small intestinal disease (i.e., resolution of both diarrhea and lymphoplasmacytic infiltration of the intestine and the disappearance of the leukemic component and the monotypic truncated immunoglobulin heavy chain in serum).
Further support for an association between C. jejuni and immunoproliferative small intestinal disease is provided by four additional observations. Our retrospective study of a monocentric series of archival biopsy specimens yielded four additional cases of campylobacter-associated disease among six patients with immunoproliferative small intestinal disease. Puri et al. described a patient in whom culture-positive C. jejuni diarrhea developed two to three days after the initiation of antineoplastic chemotherapy for immunoproliferative small intestinal disease24 — a sequence of events that suggests that chemotherapy exacerbated a preexisting C. jejuni infection. Immunoproliferative small intestinal disease occurs almost exclusively in developing countries where C. jejuni infection is hyperendemic, often chronic, and asymptomatic.5,25,26,27 Finally, antimicrobial regimens reported to be effective in treating the disease are also active against C. jejuni infection.5,6,7
Our FISH and immunohistochemical studies detected C. jejuni in biopsy specimens of the small intestine and stomach from the index patient but not in biopsy specimens of the intestinal lumen. This result is in agreement with the negative stool cultures and may account for the failure of previous studies to link immunoproliferative small intestinal disease with this organism with the use of standard culture-based studies.5,6,7 Moreover, C. jejuni is microaerophilic and may exist in a viable but noncultivatable state.28
Our results do not allow us to conclude that C. jejuni is the only bacterial species associated with immunoproliferative small intestinal disease. Nonetheless, C. jejuni should be added to the growing list of human pathogens responsible for chronic infection that are also implicated in antigen-driven immunoproliferative states.29,30,31 The association of C. jejuni with immunoproliferative small intestinal disease is reminiscent of the link between H. pylori infection and gastric MALT lymphoma.29,30 However, in contrast to a previous case report8 and in agreement with a more recent study based on a series of 21 cases,9 we found no evidence that H. pylori is involved in the development of immunoproliferative small intestinal disease.
C. jejuni has been shown to persist in Peyer's patches and mesenteric lymph nodes in a gnotobiotic mouse model32 and to secrete a toxin, CdtB, that mediates DNA damage.33 These properties could be critical in the pathogenesis of immunoproliferative small intestinal disease. C. jejuni can elicit a strong IgA mucosal response, and chronic infection with C. jejuni leads to sustained stimulation of the mucosal immune system.26 This persistent stimulation could eventually lead to the expansion of IgA-secreting clones and to the selection of a clone that secretes heavy chains that has eluded antibody-antigen Fc-dependent down-regulation.34,35 Eradication of the antigenic source with antimicrobial treatment may stop the proliferation of this lymphoplasmacytic population.
As is true for H. pylori–associated gastric MALT lymphomas, a better understanding of the role of C. jejuni in the pathogenesis of immunoproliferative small intestinal disease will require further analysis of mechanisms and the development of suitable animal models. The identification of an association between C. jejuni and immunoproliferative small intestinal disease may lead to improvements in the diagnosis, management, and prevention of this disease, at least in a subgroup of patients.
Supported by grants from the Programme de Recherche Fondamentale en Microbiologie, Maladies Infectieuses et Parasitaires, Ministère de l'Education Nationale de la Recherche et de la Technologie.
We are indebted to Hélène Poirel for Southern blot analysis, to Pierre Aucouturier for Western blot studies, to Jean-Louis Fauchère for C. jejuni and H. pylori serologic analyses, and to Jean-Claude Rambaud for assistance in retrieving archival biopsy materials and for helpful discussions.
Source Information
From H?pital Avicenne (M.L., A.M., F.S., J.F., L.G., O.L.), Institut Pasteur (M.L., O.L.), H?pital Necker (E.A., C.P., P.B.), Conservatoire National des Arts et Métiers (P.P.), H?pital Saint-Louis (D.B.), and H?pital Lariboisière (A.L.) — all in Paris; and Washington University School of Medicine, St. Louis (J.I.G.).
Address reprint requests to Dr. Lecuit at the Service des Maladies Infectieuses et Tropicales, H?pital Necker–Enfants Malades, Université Paris V, 149 rue de Sèvres, 75743 Paris CEDEX 15, France, or at mlecuit@pasteur.fr.
References
Seligmann M, Danon F, Hurez D, Mihaesco E, Preud'homme JL. Alpha-chain disease: a new immunoglobulin abnormality. Science 1968;162:1396-1397.
Isaacson P, Wright DH. Malignant lymphoma of mucosa-associated lymphoid tissue: a distinctive type of B-cell lymphoma. Cancer 1983;52:1410-1416.
Isaacson PG, Dogan A, Price SK, Spencer J. Immunoproliferative small-intestinal disease: an immunohistochemical study. Am J Surg Pathol 1989;13:1023-1033.
Salem P, Anaissie E, Allam C, et al. Non-Hodgkin's lymphomas in the Middle East: a study of 417 patients with emphasis on special features. Cancer 1986;58:1162-1166.
Rambaud J-C, Brouet J-C, Seligmann M. Alpha chain disease and related lymphoproliferative disorders. In: Ogra PL, Mestecky J, Lamm ME, Strober W, McGhee JR, Bienenstock J, eds. Handbook of mucosal immunology. San Diego, Calif.: Academic Press, 1994:425-33.
Isaacson PG, Norton AJ, eds. Extranodal lymphomas. London: Churchill Livingstone, 1994:340.
Harzic M, Girard-Pipau F, Halphen M, Ferchal F, Pérol Y, Rambaud J-C. étude bactériologique, parasitologique et virologique de la flore digestive dans la maladie des cha?nes alpha. Gastroenterol Clin Biol 1985;9:472-479.
Fischbach W, Tacke W, Greiner A, Konrad H, Müller-Hermelink. Regression of immunoproliferative small intestinal disease after eradication of Helicobacter pylori. Lancet 1997;349:31-32.
Malekzadeh R, Kaviani MJ, Tabei SZ, Abdolhadi B, Haghshenas M, Navab F. Lack of association between Helicobacter pylori infection and immunoproliferative small intestinal disease. Arch Iran Med 1999;2:1-4.
Relman DA, Loutit JS, Schmidt TM, Falkow S, Tompkins LS. The agent of bacillary angiomatosis: an approach to the identification of uncultured pathogens. N Engl J Med 1990;323:1573-1580.
Relman DA, Schmidt TM, MacDermott RP, Falkow S. Identification of the uncultured bacillus of Whipple's disease. N Engl J Med 1992;327:293-301.
Galian A, Lecestre MJ, Scotto J, Bognel C, Matuchansky C, Rambaud JC. Pathological study of alpha-chain disease, with special emphasis on evolution. Cancer 1977;39:2081-2101.
Newell DG. Monoclonal antibodies directed against the flagella of Campylobacter jejuni: production, characterization and lack of effect on the colonization of infant mice. J Hyg (Lond) 1986;96:131-141.
Greisen K, Loeffelholz M, Purohit A, Leong D. PCR primers and probes for the 16S rRNA gene of most species of pathogenic bacteria, including bacteria found in cerebrospinal fluid. J Clin Microbiol 1994;32:335-351.
Linton D, Lawson AJ, Owen RJ, Stanley J. PCR detection, identification to species level, and fingerprinting of Campylobacter jejuni and Campylobacter coli direct from diarrheic samples. J Clin Microbiol 1997;35:2568-2572.
Suerbaum S, Josenhans C, Labigne A. Cloning and genetic characterization of the Helicobacter pylori and Helicobacter mustelae flaB flagellin genes and construction of H. pylori flaA- and flaB-negative mutants by electroporation-mediated allelic exchange. J Bacteriol 1993;175:3278-3288.
Kwok AY, Su SC, Reynolds RP, et al. Species identification and phylogenetic relationships based on partial HSP60 gene sequences within the genus Staphylococcus. Int J Syst Bacteriol 1999;49:1181-1192.
Rossello-Mora R, Amann R. The species concept for prokaryotes. FEMS Microbiol Rev 2001;25:39-67.
Waller DF, Ogata SA. Quantitative immunocapture PCR assay for detection of Campylobacter jejuni in foods. Appl Environ Microbiol 2000;66:4115-4118.
Trebesius K, Panthel K, Strobel S, et al. Rapid and specific detection of Helicobacter pylori macrolide resistance in gastric tissue by fluorescent in situ hybridisation. Gut 2000;46:608-614.
Boosinger TR, Powe TA. Campylobacter jejuni infections in gnotobiotic pigs. Am J Vet Res 1988;49:456-458.
Weiss LM, Chen YY. Effects of different fixatives on detection of nucleic acids from paraffin-embedded tissues by in situ hybridization using oligonucleotide probes. J Histochem Cytochem 1991;39:1237-1242.
Montone KT, Litzky LA. Rapid method for detection of Aspergillus 5S ribosomal RNA using a genus-specific oligonucleotide probe. Am J Clin Pathol 1995;103:48-51.
Puri AS, Aggarwal R, Khan EM, et al. Explosive Campylobacter jejuni diarrhea in immunoproliferative small intestinal disease. Indian J Gastroenterol 1992;11:141-143.
Ponka A, Pitkanen T, Sarna S, Kosunen TU. Infection due to Campylobacter jejuni: a report of 524 outpatients. Infection 1984;12:175-178.
Skirrow MB, Blaser MJ. Campylobacter jejuni. In: Blaser MJ, Smith PD, Ravdin JI, Greenberg HB, Guerrant RL, eds. Infections of the gastrointestinal tract. New York: Raven Press, 1995:825-48.
Coker AO, Isokpehi RD, Thomas BN, Amisu KO, Obi CL. Human campylobacteriosis in developing countries. Emerg Infect Dis 2002;8:237-244.
Rollins DM, Colwell RR. Viable but nonculturable stage of Campylobacter jejuni and its role in survival in the natural aquatic environment. Appl Environ Microbiol 1986;52:531-538.
Wotherspoon AC, Doglioni C, Diss TC, et al. Regression of primary low-grade B-cell gastric lymphoma of mucosa-associated lymphoid tissue type after eradication of Helicobacter pylori. Lancet 1993;342:575-577.
Hussell T, Isaacson PG, Crabtree JE, Spencer J. The response of cells from low-grade B-cell gastric lymphomas of mucosa-associated lymphoid tissue to Helicobacter pylori. Lancet 1993;342:571-574.
Pagano JS. Viruses and lymphomas. N Engl J Med 2002;347:78-79.
Fauchere JL, Veron M, Lellouch-Tubiana A, Pfister A. Experimental infection of gnotobiotic mice with Campylobacter jejuni: colonisation of intestine and spread to lymphoid and reticulo-endothelial organs. J Med Microbiol 1985;20:215-224.
Lara-Tejero M, Galan JE. A bacterial toxin that controls cell cycle progression as a deoxyribonuclease I-like protein. Science 2000;290:354-357.
Yodoi J, Adachi M, Noro N. IgA binding factors and Fc receptors for IgA: comparative studies between IgA and IgE Fc receptor systems. Int Rev Immunol 1987;2:117-141.
Ogra PL, Lamm ME, Mestecky J, Strober W, Bienenstock J. Mucosal immunology. 2nd ed. Vol. 1. San Diego, Calif.: Academic Press, 1999.(Marc Lecuit, M.D., Ph.D.,)
Background Immunoproliferative small intestinal disease (also known as alpha chain disease) is a form of lymphoma that arises in small intestinal mucosa-associated lymphoid tissue (MALT) and is associated with the expression of a monotypic truncated immunoglobulin heavy chain without an associated light chain. Early-stage disease responds to antibiotics, suggesting a bacterial origin. We attempted to identify a causative agent.
Methods We performed polymerase chain reaction (PCR), DNA sequencing, fluorescence in situ hybridization, and immunohistochemical studies on intestinal-biopsy specimens from a series of patients with immunoproliferative small intestinal disease.
Results Analysis of frozen intestinal tissue obtained from an index patient with immunoproliferative small intestinal disease who had a dramatic response to antibiotics revealed the presence of Campylobacter jejuni. A follow-up retrospective analysis of archival intestinal-biopsy specimens disclosed campylobacter species in four of six additional patients with immunoproliferative small intestinal disease.
Conclusions These results indicate that campylobacter and immunoproliferative small intestinal disease are associated and that C. jejuni should be added to the growing list of human pathogens responsible for immunoproliferative states.
Immunoproliferative small intestinal disease, also known as alpha chain disease, is a mucosa-associated lymphoid-tissue (MALT) lymphoma characterized by infiltration of the bowel wall with a plasma-cell population that secretes a monotypic, truncated immunoglobulin heavy chain lacking an associated light chain.1,2,3 Lymphoid infiltration leads to malabsorption and protein-losing enteropathy. The disease can cause a spectrum of histopathological changes, ranging from seemingly benign lymphoid infiltration to malignant diffuse large-B-cell lymphoma.
Since its initial description,1 immunoproliferative small intestinal disease has largely been reported in the Mediterranean basin, the Middle East, the Far East, and Africa. In the Middle East, the most common site of extranodal lymphoma is the gastrointestinal tract: immunoproliferative small intestinal disease accounts for approximately one third of gastrointestinal lymphomas.4 The restricted geographic distribution of the disease has led to the hypothesis that environmental factors may have a pathogenic role.
Remarkably, early-stage immunoproliferative small intestinal disease typically responds to antibiotics, suggesting that it may be triggered by bacterial infection.5,6 However, previous attempts to identify a causative agent with the use of standard culture methods have failed.5,6,7 Immunoproliferative small intestinal disease shares histopathological features with gastric MALT lymphoma associated with Helicobacter pylori infection,2,3 and a case report suggested that H. pylori may be a potential causative agent.8 However, this proposal was not supported by the results of a subsequent retrospective study of 21 patients with immunoproliferative small intestinal disease.9 We used a molecular strategy similar to that used to identify the causative agents of bacillary angiomatosis (Bartonella henselae)10 and Whipple's disease (Tropheryma whipplei),11 in order to determine whether immunoproliferative small intestinal disease could be linked to specific bacterial species.
Case Reports
Index Patient
A 45-year-old woman from Cameroon was hospitalized for a 12-month-long illness characterized by chronic diarrhea and wasting. Histologic examination of endoscopic biopsy specimens from her duodenum and proximal jejunum revealed massive infiltration of the lamina propria, with a dimorphic population of plasma cells located in the superficial mucosa and centrocyte-like lymphocytes proliferating deeper in the mucosa, mostly around crypts (Figure 1A, Figure 1B, Figure 1C, and Figure 1D). Infiltration of the lamina propria was less pronounced in antral-biopsy specimens.
Figure 1. Histologic and Immunohistologic Studies of Endoscopic-Biopsy Specimens from the Index Patient.
In Panel A, staining with hematoxylin and eosin shows lymphoid infiltration of the lamina propria in a section of jejunum (x50). In Panel B, a section of the antral region of the stomach stained with hematoxylin and eosin shows lymphoid infiltration of the lamina propria (x100). In Panel C, which shows a higher-power view of the specimen in Panel A, the typical lymphoplasmacytic features of the superficial mucosa infiltrate are evident (hematoxylin and eosin, x1000). In Panel D, a higher-power view of the specimen shown in Panel A reveals centrocyte-like cells infiltrating the crypt epithelium (hematoxylin and eosin, x400). Sections of jejunum (Panel E) and stomach (Panel F) stained with primary antibodies directed against the B-cell marker CD20 (appears brown when stained with enzyme-linked secondary antibodies) and counterstained with hematoxylin (x100) show CD20-positive centrocyte-like lymphocytes pervading the lamina propria surrounding crypts (bottom insets show higher-power views of boxed areas, x300). CD20-positive centrocyte-like lymphocytes infiltrate the crypt epithelium and produce characteristic lymphoepithelial lesions (arrows in Panels E and F). In contrast, the superficial plasma-cell infiltrate is negative for CD20. The inset in the upper right-hand portion of Panel F shows a section stained with KL1 (which recognizes cytokeratin) and counterstained with hematoxylin (x300). Cross-sectioned crypts contain brown, cytokeratin-positive epithelial cells. Jejunal sections (x400) were incubated with monoclonal antibodies that recognize the heavy chain (Panel G), heavy chain (Panel H), light chain (Panel I), or light chain (Panel J) and were counterstained with hematoxylin. The results establish that the infiltrating lymphoid-cell population (lymphoplasmacytes and centrocyte-like cells) expresses the heavy chain without its associated light chain.
Immunohistochemical studies of these biopsy specimens showed that the plasma cells were negative for CD20, whereas the centrocyte-like lymphocytes were positive for CD20 (Figure 1E and Figure 1F). The cytoplasm of both plasma cells and centrocyte-like lymphocytes showed intense staining for immunoglobulin heavy chain, with no detectable expression of light chains or surface immunoglobulin in these cells (Figure 1G, Figure 1H, Figure 1I, and Figure 1J). Centrocyte-like lymphocytes were responsible for the lymphoepithelial lesions that are typical of MALT lymphomas (Figure 1D, Figure 1E, and Figure 1F).
Staining of gastric- and intestinal-biopsy specimens with Giemsa and silver stains and serologic assays were all negative for H. pylori. Tissue and stool cultures were negative for salmonella, yersinia, shigella, and campylobacter. An enzyme-linked immunosorbent assay and Western blot assays of serum were positive for Campylobacter jejuni (data not shown).
Analysis of blood samples disclosed 7560 lymphocytes per cubic millimeter, with the vast majority having morphologic features of lymphoplasmacytes. Analysis with the use of a fluorescence-activated cell sorter established that 82 percent of peripheral-blood lymphocytes were CD19+, CD20–, CD38+, CD138– B cells. Immunocytochemical studies of the sorted cells revealed high levels of cytoplasmic heavy chains without detectable light chains or surface immunoglobulin. Bone marrow biopsy disclosed extensive infiltration with lymphoplasmacytes. Evidence of the clonality of this population was provided by Southern blot analysis of peripheral-blood lymphocyte DNA, which showed discrete rearrangements of the genes for the immunoglobulin heavy-chain and light-chain loci (data not shown). The serum IgA level was increased to 10.9 g per liter (normal range, 0.76 to 3.90), whereas IgG and IgM levels were decreased (to 5.81 and 0.22 g per liter, respectively). Immunoelectrophoresis of serum, aspirated jejunal luminal contents, and urine from the patient showed a prominent arc of heavy chain precipitin (Figure 2A). Western blot analysis of serum revealed the presence of a truncated monotypic 1 heavy chain.
Figure 2. Resolution of Immunoproliferative Small Intestinal Disease in the Index Patient after Antimicrobial Therapy.
Panel A shows the immunoelectrophoresis of serum before and after antibiotic treatment; the arrow points to the prominent arc of heavy chain precipitin observed before treatment. Panel B shows the time course of the disappearance of the CD19+, CD38+, and heavy chain (HC)–positive leukemic cells, as assessed by analysis of peripheral-blood lymphocytes with the use of a fluorescence-activated cell sorter. Panel C shows the gradual decrease in the serum level of total IgA.
On the basis of these clinical, histopathological, and immunologic findings, a diagnosis of immunoproliferative small intestinal disease was made. Given the gastric extension of the disease, the phenotypic similarities between the disease and H. pylori–associated gastric MALT lymphoma,2,3 and a previous case report of regression of immunoproliferative small intestinal disease in a patient after the eradication of H. pylori infection,8 triple antimicrobial therapy was initiated (1 g of amoxicillin twice daily, 500 mg of metronidazole twice daily, and 500 mg of clarithromycin twice daily) in combination with a proton-pump inhibitor (20 mg of omeprazole twice daily). The diarrhea subsided within a week, the leukemic component disappeared within 10 weeks (Figure 2B), and there was progressive resolution of the paraproteinemia (Figure 2A and Figure 2C) and intestinal lymphoplasmacytic infiltrate (data not shown). Treatment was stopped after five months because the patient was asymptomatic, the leukemic and serum paraprotein abnormalities had disappeared, and lymphoplasmacytic infiltration of the intestine was barely detectable in follow-up endoscopic biopsy specimens from the jejunum. One year after the diagnosis had been established, the results of clinical examination and laboratory studies were normal.
Six Additional Patients
Immunoproliferative small intestinal disease is rare. Frozen intestinal samples from other patients with untreated immunoproliferative small intestinal disease were not available to us. Therefore, we retrieved archival paraffin-embedded excisional jejunal-biopsy specimens obtained at the time of laparoscopy 8 to 27 years earlier from six additional patients in whom immunoproliferative small intestinal disease had been diagnosed at a single hospital according to established histopathological and immunologic criteria.1,5,6,12
Methods
Tissue Samples
We analyzed tissue specimens from six sources. Frozen and formalin-fixed, paraffin-embedded specimens were obtained from gastric, duodenal, and jejunal biopsies performed endoscopically in the index patient one day before and eight days after the initiation of antimicrobial therapy. Archival jejunal-biopsy specimens, which had been fixed in Bouin's fluid and embedded in paraffin, were obtained from six additional patients who had received a diagnosis of immunoproliferative small intestinal disease. Formalin-fixed, paraffin-embedded jejunal-biopsy specimens were obtained endoscopically from a patient infected with human immunodeficiency virus (HIV) who had acute C. jejuni enteritis. Formalin-fixed, paraffin-embedded stomach samples were obtained from a patient with H. pylori gastritis. Frozen duodenal-biopsy specimens were obtained from 10 patients with chronic diarrhea of unknown origin (as defined by negative cultures for known enteropathogens). Formalin-fixed, paraffin-embedded duodenal-biopsy specimens were obtained from 10 patients who were being evaluated for anemia.
Bacterial Strains and Antibodies
We obtained C. jejuni reference strain CIP 70.2, H. pylori reference strain CIP 103995, and Escherichia coli reference strain CIP 7624 from the Institut Pasteur collection. Two mouse monoclonal antibodies were used for immunohistochemical studies: IgG2b NCL-C-JEJUNI (Novocastra), which recognizes a flagellar immunologic determinant common to C. jejuni and H. pylori,13 and IgM CP1/IIG10 (Biotrend), which is directed against a 43-kD, non–surface-associated H. pylori antigen that is not detectable in other helicobacter species or unrelated bacteria.
Polymerase Chain Reaction
Universal primers PB (5'TAACACATGCAAGTCGAACG3') and DG74 (5'AGGAGGTGATCCAACCGCA3') were used to amplify bacterial 16S ribosomal DNA (rDNA).14 Established primers directed at C. jejuni (CCCJ609F 5'AATCTAATGGCTTAACCATTA3' and CCJ1442R 5'GTAACTAGTTTAGTATTCCGG3')15 were used to generate an 852-bp 16S rDNA amplicon. Primers specific for H. pylori (5'fla 5'CCACGGTTAAAGCGTCTATTGG3' and 3'fla 5'GATCGCATTAGTCAACCTCCCG3')16 were used to amplify a 401-bp fragment from the flaB gene. Two pairs of primers — 72/96F (5'ACAGGAAGAAGCTTGCTTCTTTGC3') and 455/477R (5'GAGCAAAGGTATTAACTTTACTCCC3') and 455/477F (5'GGGAGTAAAGTTAATACCTTTGCTC3') and 1000/1025R (5'ACATTCTCATCTCTGAAAACTTCCG3') — were designed to amplify 16S rDNA from Enterobacteriaceae. Cycling conditions for all polymerase chain reactions (PCRs) were as follows: 7 minutes at 95°C, 35 cycles of denaturation for 30 seconds at 95°C, annealing for 20 seconds at 55°C, and 2 minutes of extension at 72°C and then 10 minutes at 72°C.
The specificity of the PCR assays was initially established with the use of DNA extracted from the three reference strains listed above (Table 1). For the index patient, amplicons from 16S ribosomal genes were subcloned, and the nucleotide sequences of 12 randomly chosen clones were determined on both strands. All 16S rDNA sequences that were at least 99 percent homologous were considered to belong to the same species.17,18
Table 1. Results of Polymerase-Chain-Reaction (PCR) Assays of Biopsy Specimens from the Index Patient with Immunoproliferative Small Intestinal Disease and Control Samples.
Fluorescence in Situ Hybridization
Cy3-tagged 5'ATTACTGAGATGACTAGCACCCC3' (Cj-490) was used to probe for the presence of C. jejuni,19 whereas Cy3-tagged 5'CACACCTGACTGACTATCCCG3' (Hpy-1) was used to detect H. pylori.20 Deparaffinized, formalin-fixed sections that were 5 μm thick were incubated for 10 minutes in lysozyme buffer (10 mg of lysozyme per milliliter , 100 mM TRIS–hydrochloric acid , and 50 mM EDTA) at room temperature and then for 2 hours at 35°C in hybridization chambers with the oligonucleotide probe (diluted to a final concentration of 4.5 ng per microliter in a solution of 30 percent formamide, 0.9 M sodium chloride, 20 mM TRIS–hydrochloric acid , and 0.01 percent sodium dodecyl sulfate). Tissue sections were washed twice (for 15 minutes at 37°C per cycle) in wash buffer (0.9 M sodium chloride, 20 mM TRIS–hydrochloric acid , 5 mM EDTA, and 0.01 percent sodium dodecyl sulfate) and once in distilled water, dried at room temperature, mounted, and observed by means of epifluorescence microscopy.
All tissue samples studied were collected as part of routine care. The National Ethics Committee of France requires neither approval by institutional review boards nor informed consent for this type of research to be conducted.
Results
Identification of C. jejuni DNA
DNA was extracted from frozen jejunal- and gastric-biopsy specimens from the index patient. Initial PCR assays of these DNAs were performed with the use of well-established universal primers known to generate amplicons from the 16S rDNA genes from most phyla in the bacteria superkingdom.14 Amplicons were produced in assays in which the DNA template was extracted from frozen jejunal samples harvested one day before treatment was initiated. In contrast, no amplicons were produced with the use of DNA prepared from material harvested eight days after therapy was started.
The nucleotide sequences of 8 of 12 cloned amplicons generated from the pretreatment preparation of jejunal DNA were 99.6 percent identical to 16S rDNA amplified from DNA of the C. jejuni reference strain. Computer searches of GenBank with the use of the BLAST program (http://www.ncbi.nlm.nih.gov/blast/) and the 16S rDNA data base (http://greengenes.llnl.gov/16S/) showed that the four remaining sequences were from abiotrophia, neisseria, lactococcus, and haemophilus. Members of these four genera are part of the normal oropharyngeal microbiota in humans.
Follow-up PCR studies with the use of primers directed at C. jejuni 16S rDNA yielded amplicons of the expected size from antral, jejunal, and fecal DNAs prepared from material obtained before the initiation of antimicrobial therapy (Table 1). Assays of the same DNAs with the use of primers directed at H. pylori 16S rDNA or primers that recognize 16S rDNA from members of the Enterobacteriaceae family were all negative (Table 1). In addition, the PCR assay was negative with the use of primers directed at C. jejuni and DNA prepared from jejunal- and gastric-biopsy specimens obtained from the index patient eight days after the initiation of antibiotic treatment and from DNA extracted from frozen proximal intestinal-biopsy specimens from 10 control patients with diarrhea of unknown origin (i.e., not associated with cultivatable enteropathogens) (Table 1).
The results of sequencing 16S rDNA amplicons obtained with universal bacterial 16S rDNA primers and the C. jejuni–directed PCR assays of endoscopic biopsy specimens from the index patient and controls provided initial evidence of an association between this campylobacter species and immunoproliferative small intestinal disease.
Detection of Campylobacter jejuni in Small Intestinal Tissue
To provide further evidence of the presence of C. jejuni in the index patient, we developed a fluorescence in situ hybridization (FISH) assay using an oligonucleotide probe directed at C. jejuni 16S rRNA. The sensitivity and specificity of the FISH assay were established as follows: staining with the Cy3-labeled oligonucleotide was readily detectable in a smear of cultured C. jejuni (Figure 3A), intraluminal C. jejuni were detected in a jejunal-biopsy specimen from an HIV-infected patient with culture-documented C. jejuni enteritis (Table 2), and no signal was noted in gastric-biopsy specimens from a patient with H. pylori gastritis (Table 2 and Figure 3B).
Figure 3. Fluorescence in Situ Hybridization (FISH) and Immunohistochemical Analysis of Tissue Specimens.
Panel A shows a positive control for FISH: the Cy3-labeled oligonucleotide probe directed against Campylobacter jejuni and a smear of cultured C. jejuni type strain (x400). Panels B and C show the specificity of the FISH assay; the C. jejuni–directed probe or a Helicobacter pylori–specific probe and sections of a gastric-biopsy specimen from a patient with H. pylori–associated gastritis were used for FISH (x400). The C. jejuni probe does not recognize H. pylori (Panel B), whereas the H. pylori probe produces a prominent signal (Panel C). Panels D, E, F, and G show the results of FISH analysis with the use of the C. jejuni probe and sections prepared from biopsy specimens of the proximal small intestine (Panel D) and antrum (Panel F) obtained from the index patient before the initiation of antimicrobial therapy. After hybridization, sections were demounted and stained with hematoxylin and eosin (Panels E and G). Prominent signals are seen in the lamina propria and surrounding blood vessels (arrows in Panels D, E, F, and G) (x400). Panels H, I, J, K, L, and M show immunohistochemical analysis of jejunal sections stained with NCL-C-JEJUNI monoclonal antibody (appears brown when stained with enzyme-linked secondary antibodies) and hematoxylin (x400). The arrows point to immunolabeled material shown at a higher magnification in the bottom insets in the six panels (x800). Panel H shows a specimen from a patient infected with human immunodeficiency virus who had acute C. jejuni enteritis. Bacteria are present in the lumen and are associated with enterocytes. Panel I shows a specimen from the index patient. Panels J, K, and L show archival biopsy specimens from three patients with immunoproliferative small intestinal disease (Patients 1, 2, and 3 in Table 2) that are positive for C. jejuni on FISH. The top insets in these three panels show sections with intraluminal immunolabeled bacteria (x400). Panel M shows a biopsy specimen from a patient with immunoproliferative small intestinal disease (Patient 6 in Table 2) that was negative for C. jejuni on FISH.
Table 2. Results of Fluorescence in Situ Hybridization and Immunohistochemical Assays of Biopsy Specimens from the Index Patient, Six Other Patients with Immunoproliferative Small Intestinal Disease, and Controls.
Subsequent FISH analysis with the C. jejuni probe revealed the organism in sections of jejunal- and stomach-biopsy specimens obtained from the index patient before the initiation of antimicrobial therapy. The organism was most prominent in the lamina propria, often in the vicinity of capillaries (Figure 3D, Figure 3E, Figure 3F, and Figure 3G), a finding consistent with the observed tropism of this organism in an animal model of infection.21 A control H. pylori–specific FISH probe (Figure 3C) did not produce a signal (Table 2), a finding that is consistent with the negative Giemsa and silver staining and the PCR results (Table 1).
Finally, immunohistochemical methods were used to confirm the presence of C. jejuni in the index patient. Two well-characterized, commercially available preparations of monoclonal antibody were used: one recognizes a flagellar immunologic determinant shared by C. jejuni and H. pylori (NCL-C-JEJUNI), and another is specific for an H. pylori epitope (CP1/IIG10). Control experiments that used sections of jejunum from an HIV-infected patient with culture-positive acute C. jejuni enteritis disclosed a prominent signal with NCL-C-JEJUNI and no signal with the H. pylori–specific monoclonal antibody (Figure 3H and Table 2). Assays of sections prepared from jejunal- and gastric-biopsy specimens obtained before the index patient started antimicrobial therapy disclosed readily detectable signals in the lamina propria with NCL-C-JEJUNI but not with the H. pylori–specific antibody (Figure 3I and Table 2). Together, these results support the conclusion that C. jejuni was present at the sites of gut abnormalities in the index patient and rule out the possibility of concomitant H. pylori infection.
Analysis of Archival Tissues
We were unable to recover amplifiable DNA from any of the archival samples from six patients with immunoproliferative small intestinal disease — that is, the PCRs for the human -actin gene as well as for C. jejuni 16S rDNA were all negative. This failure most likely reflects the method of fixation together with the age of the samples. However, because material fixed with Bouin's fluid is known to be suitable for FISH,22,23 these biopsy specimens were analyzed with the use of three probes: a universal bacterial 16S rDNA oligonucleotide (positive control for the presence of bacteria), the probe directed at C. jejuni, and the specific probe for H. pylori. The bacterial FISH probe was positive in samples from five of the six patients (Patients 1 through 5 in Table 2). The C. jejuni–directed probe produced a positive signal in three of the six biopsy specimens from these patients (Patients 1, 2, and 3 in Table 2).
In addition, the two monoclonal antibodies were used to confirm that C. jejuni immunologic determinants were present in all three FISH-positive samples (Figure 3J, Figure 3K, and Figure 3L), as well as in one FISH-negative sample (Figure 3M and Table 2). No signal was detected with these monoclonal antibodies in the 10 control duodenal samples. Moreover, the H. pylori–specific FISH probe and monoclonal antibody did not detect this species in any of the archival jejunal-biopsy specimens from the six patients with immunoproliferative small intestinal disease (Table 2).
Discussion
Previous attempts to link a bacterial species to immunoproliferative small intestinal disease with the use of culture-based approaches have failed.5,6,7 We used a molecular approach to establish a link between immunoproliferative small intestinal disease and C. jejuni infection in an index patient. This association is based on three observations. First, a PCR assay with the use of universal bacterial 16S rDNA primers and DNA templates prepared from biopsy specimens of the proximal small intestine obtained before the initiation of antimicrobial treatment in the index patient yielded amplicons encompassing the entire 16S rDNA gene that were identified as C. jejuni by sequencing. No other enteropathogen was detected by PCR. Second, FISH and immunohistochemical analyses showed C. jejuni in diseased biopsy specimens of the small intestine. Third, the eradication of C. jejuni with antimicrobial therapy was associated with rapid remission of the immunoproliferative small intestinal disease (i.e., resolution of both diarrhea and lymphoplasmacytic infiltration of the intestine and the disappearance of the leukemic component and the monotypic truncated immunoglobulin heavy chain in serum).
Further support for an association between C. jejuni and immunoproliferative small intestinal disease is provided by four additional observations. Our retrospective study of a monocentric series of archival biopsy specimens yielded four additional cases of campylobacter-associated disease among six patients with immunoproliferative small intestinal disease. Puri et al. described a patient in whom culture-positive C. jejuni diarrhea developed two to three days after the initiation of antineoplastic chemotherapy for immunoproliferative small intestinal disease24 — a sequence of events that suggests that chemotherapy exacerbated a preexisting C. jejuni infection. Immunoproliferative small intestinal disease occurs almost exclusively in developing countries where C. jejuni infection is hyperendemic, often chronic, and asymptomatic.5,25,26,27 Finally, antimicrobial regimens reported to be effective in treating the disease are also active against C. jejuni infection.5,6,7
Our FISH and immunohistochemical studies detected C. jejuni in biopsy specimens of the small intestine and stomach from the index patient but not in biopsy specimens of the intestinal lumen. This result is in agreement with the negative stool cultures and may account for the failure of previous studies to link immunoproliferative small intestinal disease with this organism with the use of standard culture-based studies.5,6,7 Moreover, C. jejuni is microaerophilic and may exist in a viable but noncultivatable state.28
Our results do not allow us to conclude that C. jejuni is the only bacterial species associated with immunoproliferative small intestinal disease. Nonetheless, C. jejuni should be added to the growing list of human pathogens responsible for chronic infection that are also implicated in antigen-driven immunoproliferative states.29,30,31 The association of C. jejuni with immunoproliferative small intestinal disease is reminiscent of the link between H. pylori infection and gastric MALT lymphoma.29,30 However, in contrast to a previous case report8 and in agreement with a more recent study based on a series of 21 cases,9 we found no evidence that H. pylori is involved in the development of immunoproliferative small intestinal disease.
C. jejuni has been shown to persist in Peyer's patches and mesenteric lymph nodes in a gnotobiotic mouse model32 and to secrete a toxin, CdtB, that mediates DNA damage.33 These properties could be critical in the pathogenesis of immunoproliferative small intestinal disease. C. jejuni can elicit a strong IgA mucosal response, and chronic infection with C. jejuni leads to sustained stimulation of the mucosal immune system.26 This persistent stimulation could eventually lead to the expansion of IgA-secreting clones and to the selection of a clone that secretes heavy chains that has eluded antibody-antigen Fc-dependent down-regulation.34,35 Eradication of the antigenic source with antimicrobial treatment may stop the proliferation of this lymphoplasmacytic population.
As is true for H. pylori–associated gastric MALT lymphomas, a better understanding of the role of C. jejuni in the pathogenesis of immunoproliferative small intestinal disease will require further analysis of mechanisms and the development of suitable animal models. The identification of an association between C. jejuni and immunoproliferative small intestinal disease may lead to improvements in the diagnosis, management, and prevention of this disease, at least in a subgroup of patients.
Supported by grants from the Programme de Recherche Fondamentale en Microbiologie, Maladies Infectieuses et Parasitaires, Ministère de l'Education Nationale de la Recherche et de la Technologie.
We are indebted to Hélène Poirel for Southern blot analysis, to Pierre Aucouturier for Western blot studies, to Jean-Louis Fauchère for C. jejuni and H. pylori serologic analyses, and to Jean-Claude Rambaud for assistance in retrieving archival biopsy materials and for helpful discussions.
Source Information
From H?pital Avicenne (M.L., A.M., F.S., J.F., L.G., O.L.), Institut Pasteur (M.L., O.L.), H?pital Necker (E.A., C.P., P.B.), Conservatoire National des Arts et Métiers (P.P.), H?pital Saint-Louis (D.B.), and H?pital Lariboisière (A.L.) — all in Paris; and Washington University School of Medicine, St. Louis (J.I.G.).
Address reprint requests to Dr. Lecuit at the Service des Maladies Infectieuses et Tropicales, H?pital Necker–Enfants Malades, Université Paris V, 149 rue de Sèvres, 75743 Paris CEDEX 15, France, or at mlecuit@pasteur.fr.
References
Seligmann M, Danon F, Hurez D, Mihaesco E, Preud'homme JL. Alpha-chain disease: a new immunoglobulin abnormality. Science 1968;162:1396-1397.
Isaacson P, Wright DH. Malignant lymphoma of mucosa-associated lymphoid tissue: a distinctive type of B-cell lymphoma. Cancer 1983;52:1410-1416.
Isaacson PG, Dogan A, Price SK, Spencer J. Immunoproliferative small-intestinal disease: an immunohistochemical study. Am J Surg Pathol 1989;13:1023-1033.
Salem P, Anaissie E, Allam C, et al. Non-Hodgkin's lymphomas in the Middle East: a study of 417 patients with emphasis on special features. Cancer 1986;58:1162-1166.
Rambaud J-C, Brouet J-C, Seligmann M. Alpha chain disease and related lymphoproliferative disorders. In: Ogra PL, Mestecky J, Lamm ME, Strober W, McGhee JR, Bienenstock J, eds. Handbook of mucosal immunology. San Diego, Calif.: Academic Press, 1994:425-33.
Isaacson PG, Norton AJ, eds. Extranodal lymphomas. London: Churchill Livingstone, 1994:340.
Harzic M, Girard-Pipau F, Halphen M, Ferchal F, Pérol Y, Rambaud J-C. étude bactériologique, parasitologique et virologique de la flore digestive dans la maladie des cha?nes alpha. Gastroenterol Clin Biol 1985;9:472-479.
Fischbach W, Tacke W, Greiner A, Konrad H, Müller-Hermelink. Regression of immunoproliferative small intestinal disease after eradication of Helicobacter pylori. Lancet 1997;349:31-32.
Malekzadeh R, Kaviani MJ, Tabei SZ, Abdolhadi B, Haghshenas M, Navab F. Lack of association between Helicobacter pylori infection and immunoproliferative small intestinal disease. Arch Iran Med 1999;2:1-4.
Relman DA, Loutit JS, Schmidt TM, Falkow S, Tompkins LS. The agent of bacillary angiomatosis: an approach to the identification of uncultured pathogens. N Engl J Med 1990;323:1573-1580.
Relman DA, Schmidt TM, MacDermott RP, Falkow S. Identification of the uncultured bacillus of Whipple's disease. N Engl J Med 1992;327:293-301.
Galian A, Lecestre MJ, Scotto J, Bognel C, Matuchansky C, Rambaud JC. Pathological study of alpha-chain disease, with special emphasis on evolution. Cancer 1977;39:2081-2101.
Newell DG. Monoclonal antibodies directed against the flagella of Campylobacter jejuni: production, characterization and lack of effect on the colonization of infant mice. J Hyg (Lond) 1986;96:131-141.
Greisen K, Loeffelholz M, Purohit A, Leong D. PCR primers and probes for the 16S rRNA gene of most species of pathogenic bacteria, including bacteria found in cerebrospinal fluid. J Clin Microbiol 1994;32:335-351.
Linton D, Lawson AJ, Owen RJ, Stanley J. PCR detection, identification to species level, and fingerprinting of Campylobacter jejuni and Campylobacter coli direct from diarrheic samples. J Clin Microbiol 1997;35:2568-2572.
Suerbaum S, Josenhans C, Labigne A. Cloning and genetic characterization of the Helicobacter pylori and Helicobacter mustelae flaB flagellin genes and construction of H. pylori flaA- and flaB-negative mutants by electroporation-mediated allelic exchange. J Bacteriol 1993;175:3278-3288.
Kwok AY, Su SC, Reynolds RP, et al. Species identification and phylogenetic relationships based on partial HSP60 gene sequences within the genus Staphylococcus. Int J Syst Bacteriol 1999;49:1181-1192.
Rossello-Mora R, Amann R. The species concept for prokaryotes. FEMS Microbiol Rev 2001;25:39-67.
Waller DF, Ogata SA. Quantitative immunocapture PCR assay for detection of Campylobacter jejuni in foods. Appl Environ Microbiol 2000;66:4115-4118.
Trebesius K, Panthel K, Strobel S, et al. Rapid and specific detection of Helicobacter pylori macrolide resistance in gastric tissue by fluorescent in situ hybridisation. Gut 2000;46:608-614.
Boosinger TR, Powe TA. Campylobacter jejuni infections in gnotobiotic pigs. Am J Vet Res 1988;49:456-458.
Weiss LM, Chen YY. Effects of different fixatives on detection of nucleic acids from paraffin-embedded tissues by in situ hybridization using oligonucleotide probes. J Histochem Cytochem 1991;39:1237-1242.
Montone KT, Litzky LA. Rapid method for detection of Aspergillus 5S ribosomal RNA using a genus-specific oligonucleotide probe. Am J Clin Pathol 1995;103:48-51.
Puri AS, Aggarwal R, Khan EM, et al. Explosive Campylobacter jejuni diarrhea in immunoproliferative small intestinal disease. Indian J Gastroenterol 1992;11:141-143.
Ponka A, Pitkanen T, Sarna S, Kosunen TU. Infection due to Campylobacter jejuni: a report of 524 outpatients. Infection 1984;12:175-178.
Skirrow MB, Blaser MJ. Campylobacter jejuni. In: Blaser MJ, Smith PD, Ravdin JI, Greenberg HB, Guerrant RL, eds. Infections of the gastrointestinal tract. New York: Raven Press, 1995:825-48.
Coker AO, Isokpehi RD, Thomas BN, Amisu KO, Obi CL. Human campylobacteriosis in developing countries. Emerg Infect Dis 2002;8:237-244.
Rollins DM, Colwell RR. Viable but nonculturable stage of Campylobacter jejuni and its role in survival in the natural aquatic environment. Appl Environ Microbiol 1986;52:531-538.
Wotherspoon AC, Doglioni C, Diss TC, et al. Regression of primary low-grade B-cell gastric lymphoma of mucosa-associated lymphoid tissue type after eradication of Helicobacter pylori. Lancet 1993;342:575-577.
Hussell T, Isaacson PG, Crabtree JE, Spencer J. The response of cells from low-grade B-cell gastric lymphomas of mucosa-associated lymphoid tissue to Helicobacter pylori. Lancet 1993;342:571-574.
Pagano JS. Viruses and lymphomas. N Engl J Med 2002;347:78-79.
Fauchere JL, Veron M, Lellouch-Tubiana A, Pfister A. Experimental infection of gnotobiotic mice with Campylobacter jejuni: colonisation of intestine and spread to lymphoid and reticulo-endothelial organs. J Med Microbiol 1985;20:215-224.
Lara-Tejero M, Galan JE. A bacterial toxin that controls cell cycle progression as a deoxyribonuclease I-like protein. Science 2000;290:354-357.
Yodoi J, Adachi M, Noro N. IgA binding factors and Fc receptors for IgA: comparative studies between IgA and IgE Fc receptor systems. Int Rev Immunol 1987;2:117-141.
Ogra PL, Lamm ME, Mestecky J, Strober W, Bienenstock J. Mucosal immunology. 2nd ed. Vol. 1. San Diego, Calif.: Academic Press, 1999.(Marc Lecuit, M.D., Ph.D.,)