Distinct gene expression profiles characterize the histopathological stages of disease in Helicobacter-induced mucosa-associated lymphoid tissue lymphoma - PubMed (original) (raw)

Distinct gene expression profiles characterize the histopathological stages of disease in Helicobacter-induced mucosa-associated lymphoid tissue lymphoma

Anne Mueller et al. Proc Natl Acad Sci U S A. 2003.

Abstract

Long-term colonization of humans with Helicobacter pylori can cause the development of gastric B cell mucosa-associated lymphoid tissue lymphoma, yet little is known about the sequence of molecular steps that accompany disease progression. We used microarray analysis and laser microdissection to identify gene expression profiles characteristic and predictive of the various histopathological stages in a mouse model of the disease. The initial step in lymphoma development is marked by infiltration of reactive lymphocytes into the stomach and the launching of a mucosal immune response. Our analysis uncovered molecular markers of both of these processes, including genes coding for the immunoglobulins and the small proline-rich protein Sprr 2A. The subsequent step is characterized histologically by the antigen-driven proliferation and aggregation of B cells and the gradual appearance of lymphoepithelial lesions. In tissues of this stage, we observed increased expression of genes previously associated with malignancy, including the laminin receptor-1 and the multidrug-resistance channel MDR-1. Finally, we found that the transition to destructive lymphoepithelial lesions and malignant lymphoma is marked by an increase in transcription of a single gene encoding calgranulin AMrp-8.

PubMed Disclaimer

Figures

Figure 1

Figure 1

Histopathological and CLUSTER analysis of BALB/c mice infected with Helicobacter. (A and B) Mice were infected up to 24 mos with H. pylori, H. felis, or H. heilmannii. One-half of each stomach was fixed, stained with hematoxylin/eosin, and examined for the presence of LELs. (A) Cross section of a _H. heilmannii-_infected stomach (Upper) displaying marked lymphocytic infiltration (dark-blue staining) and partial destruction of gastric gland epithelium (Inset). (Lower) Cross section of an age-matched uninfected stomach. (B) Virulence comparison of the three Helicobacter species. Infected stomachs were graded on a 0–3 scale, as described in Methods. The percentage of animals with moderate (grade 2) or severe (grade 3) pathology is indicated. Note: at ≈18 mos postinfection (p.i.), the majority of animals show severe signs of illness, and thus group sizes dwindle due to infection-induced and age-related deaths (only 80% of uninfected and 50–70% of infected animals survive for 24 mo, depending on the infecting Helicobacter species). This finding explains the increased percentage of animals with severe LELs at 18 mos p.i. compared with the later time points. (C) Overview of the pathology observed in all 41 animals used in this study. Thirteen control (uninfected) animals and 28 _H. heilmannii-_infected animals (contr and Hh) from three different time points [12, 18, and 22 mos (12M, 18M, 22M)] were classified on a 0–3 scale with respect to the presence of lymphocytic infiltration and LELs, as described in Methods. An overall pathology classification was assigned to every animal on the basis of these grades. The animals were assembled into four groups, as indicated by the color coding. Animals with mild pathology are depicted in green; animals with moderate and severe pathology are depicted in black; and animals with a lymphoma diagnosis are depicted in red. Control animals are shown in blue. The color code is maintained throughout the paper. (D) The dendrogram reveals the relatedness of the gene expression profiles of the mice as determined by CLUSTER analysis. The data were filtered with respect to spot quality and data distribution before clustering.

Figure 2

Figure 2

Molecular signature of _Helicobacter-infected murine stomachs. (A) Overview of all 250 genes that are differentially regulated in the infected vs. the uninfected animals and have P values <0.001. Data are a measure of relative gene expression and represent the quotient of the hybridization of the fluorescent cDNA probe prepared from each stomach sample compared with its reference pool. Red and green represent high and low experimental sample/reference ratios, respectively (see scale bar). Gray signifies technically inadequate or missing data. Rows represent genes, and columns represent arrays. An enlarged section of the overview is shown in B. The color coding of array names is described in Fig. 1_C. Selected genes are designated. The symbols next to the gene names specify the laser microdissection fraction (as indicated in the legend) containing the respective transcript. See Fig. 5, which is published as supporting information on the PNAS web site,

www.pnas.org

, for the complete cluster.

Figure 3

Figure 3

Molecular signature of the histologically distinct stages of Helicobacter-induced lymphoproliferative disease. (A) Overview of all 300 genes that are differentially regulated in the mild pathology vs. the moderate and severe pathology cohorts of infected animals and have P values <0.001. Five enlarged sections are shown in B. Selected genes are designated and the color coding of array names is as described in Fig. 1_C. The symbols next to the gene names specify the laser capture fraction (as indicated in the legend) containing the respective transcript. See Fig. 6 for the complete cluster. (C and D) Predictive power of the signature gene lists shown in Fig. 2 and A and B. Gene expression profiles were generated for an additional set of 29 animals. Arrays were clustered by using the 250 and 300 genes constituting the signature gene lists. The dendrogram in C reflects the relatedness of all 29 specimens when the 300 gene list distinguishing mild from moderate and severe pathology is used for clustering. The color coding of arrays is described in Fig. 1_C. (D) Comparison of the pathology predictions that were made on the basis of gene and array clustering shown in C and Fig. 7, with the results of the histopathological classification of the stomachs (as indicated by the color coding). Only animal B1 was placed in the wrong pathology group.

Figure 4

Figure 4

(A) The diagram depicts the result of the SAM algorithm applied pairwise to cohorts of mice grouped according to their pathology (as indicated in the squares). The 15 most significantly differing genes are listed, and the gene order reflects the significance score assigned by SAM (highest scores at the top). (B) The cellular origin of selected transcripts. Frozen tissue sections cut from stomachs of six animals were stained and subjected to laser microdissection. Both lymphocytic and mucosal fractions were isolated from each stomach with two exceptions: only the mucosal fraction could be harvested from the uninfected stomach contr15M-1, because it did not display lymphocytic aggregation. In the case of stomach Hh24M-5, the mucosa was so severely destroyed and interspersed with large aggregates of lymphocytes that no contamination-free mucosal fraction could be obtained. The gene expression data were filtered with respect to spot quality (spots with regression correlations <0.6 were omitted) and data values (genes whose log2 of red/green normalized ratio is more than two in at least four arrays were selected) before clustering. Only genes for which information was available for >80% of arrays were included. Two representative clusters are shown: (Left) Genes present in the lymphocytic fractions (“lymphocyte signature”); (Right) mucosal genes (“mucosal signature”). The gene expression pattern for calgranulin/Mrp-8 is shown separately, because it did not fall into either cluster. Genes of interest are designated, and those genes that were also identified in the whole-stomach screen are marked with an asterisk. The overall pathology of all six animals is listed in C.

Similar articles

Cited by

References

    1. Parsonnet J, Friedman G D, Vandersteen D P, Chang Y, Vogelman J H, Orentreich N, Sibley R K. N Engl J Med. 1991;325:1127–1131. - PubMed
    1. Blaser M J. Br Med J. 1998;316:1507–1510. - PMC - PubMed
    1. Wyatt J I, Rathbone B J. Scand J Gastroenterol. 1988;142:44–49. - PubMed
    1. Stolte M, Eidt S. J Clin Pathol. 1989;42:1269–1271. - PMC - PubMed
    1. Genta R M, Hamner H W, Graham D Y. Hum Pathol. 1993;24:577–583. - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources