Pseudomyxoma peritonei is a disease of MUC2-expressing goblet cells

Abstract

Pseudomyxoma peritonei, a syndrome first described by Karl F. Rokitansky in 1842, is an enigmatic, often fatal intra-abdominal disease characterized by dissecting gelatinous ascites and multifocal peritoneal epithelial implants secreting copious globules of extracellular mucin. Although past interest in the syndrome has focused on the questions of the site of origin (appendix versus ovary), mechanisms of peritoneal spread (multicentricity, redistribution phenomenon, or metastasis), and the degree of malignant transformation present (adenoma, borderline tumor, or carcinoma), another important question is the mechanism behind the accumulation of extracellular mucin, the real cause of the disease's morbidity and mortality irrespective of the site of origin, mechanism of peritoneal spread, or transformed status of its epithelium. Taking advantage of the recently cloned human mucin genes, we decided to investigate this question. Our studies revealed that pseudomyxoma peritonei is a disease of MUC2-expressing goblet cells. These cells also express MUC5AC but the latter mucin is not specific for pseudomyxoma peritonei. MUC2 expression accounts for the voluminous deposits of extracellular mucin (mucin:cell ratios exceeding 10:1) and distinguishes pseudomyxoma peritonei secondarily involving the ovary from primary ovarian mucinous tumors with peritoneal implants. Because mucinous tumors of the appendix similarly express MUC2, the MUC2 expression profile also supports an appendiceal rather than ovarian origin for pseudomyxoma peritonei. Increased steady-state mRNA is observed in pooled cases of pseudomyxoma peritonei but does not occur on the basis of gene rearrangement or gene amplification. Primary epithelial cell cultures obtained from pseudomyxoma peritonei express MUC2 whose levels can be epigenetically regulated. These lines up-regulate MUC2 expression in response to both methylation inhibition by 5-azacytidine and exposure to Pseudomonas aeruginosa lipopolysaccharide, both of whose effects can be suppressed by genistein pretreatment. Both immunocytochemical as well as in situ hybridization studies with ancillary digital image analysis reveal that MUC2 expression in cases of pseudomyxoma peritonei is independent of the degrees of malignant transformation that are present and, in fact, reflects the constitutive levels of expression observed in normal goblet cells of the appendix. Extracellular mucin accumulates dramatically in pseudomyxoma peritonei because the number of MUC2-secreting cells dramatically increase and because this MUC2 has no place to drain. These studies suggest that pseudomyxoma peritonei should be regarded as a disease of MUC2-expressing goblet cells whose MUC2 expression might be susceptible to pharmacological targeting.

Figure 1.

The typical gross appearance of pseudomyxoma peritonei illustrates the so-called “jelly-belly” (A); the hallmark of the disease is copious accumulations of mucin deposits in which the intestinal lining cells are sometimes difficult to see (B); the cells do give intense signals and can be readily seen at all magnifications when the appropriate antibody or probe is used: for example, anti-MUC2 clearly delineates the intestinal lining cells responsible for this massive mucin secretion and accumulation. The extracellular mucin deposits paradoxically are not immunoreactive because their extensive glycosylation blocks antibody access to the core protein (C); anti-sense MUC2 in situ hybridization demonstrates intense MUC2 signals within this lining epithelium (D); similarly anti-MUC5AC also delineates the intestinal lining cells responsible for mucin secretion; the extracellular mucin deposits close to the cells are, however, focally immunoreactive presumably because their glycosylation status is less restrictive to antibody access than is MUC2 (E); anti-sense MUC5AC in situ hybridization demonstrates less intense but still significant MUC5AC signals within this lining epithelium (F). The extracellular mucin specifically was not recognized by these probes and antibodies and hence was difficult to visualize. To quantitate the mucin volumes and the mucin/cell ratios depicted in the subsequent Figure 5 ▶ , we used Alcian blue staining, which stained the extracellular mucin more intensely permitting accurate digital measurements. A, Gross photograph; Original magnifications, ×250, B, H&E; C, anti-MUC2, immunoperoxidase; D, MUC2 anti-sense riboprobe; E, anti-MUC5AC, immunoperoxidase; F, MUC5AC anti-sense riboprobe.

Figure 2.

The mucin immunocytochemical profile of a primary ovarian mucinous tumor, in this case a mucinous cystadenoma (A) contrasts with the mucin immunocytochemical profile of pseudomyxoma peritonei with secondary ovarian involvement (B); the primary ovarian mucinous cystadenoma is MUC2 nonimmunoreactive (C) whereas the pseudomyxoma peritonei ovarian metastasis is strongly MUC2 immunoreactive (D); both the primary ovarian mucinous cystadenoma (E) and the pseudomyxoma peritonei metastasis (F) exhibit MUC5AC immunoreactivity. On close inspection some of the MUC5AC immunoreactivity appears extracellularly. Original magnifications: ×250 (A–E); ×450 (F). A and B, H&E; C and D, anti-MUC2, immunoperoxidase; E and F, anti-MUC5AC, immunoperoxidase.

Figure 3.

A: Southern blot of 10 pooled cases of pseudomyxoma peritonei (PP) compared to Southern blots of the Colo205 and HT29 colonic carcinoma cell lines. High molecular weight DNA (20 μg) was separately digested with the 6-cutters, EcoR1, BamHI, and PstI and probed with MUC2. No gene rearrangements or amplifications were observed in pseudomyxoma peritonei. B: Northern blot of 10 pooled cases of pseudomyxoma peritonei (PP) compared to Northern blots of the Colo205, HT29, and human mammary epithelial cells (HMEC). Equal amounts (10 μg) of total RNA were added to each lane and probed with MUC2. Only pseudomyxoma peritonei showed high steady-state mRNA levels.

Figure 4.

An origin of pseudomyxoma peritonei within the appendix is demonstrated. Here, a mucinous borderline tumor of the appendix is observed arising in situ from a normal appendix (A). The mucinous tumor is present at the top right and the normal appendix is present in the bottom left. B: Anti-sense MUC2 in situ hybridization reveals strong MUC2 signals in the corresponding areas of the mucinous borderline tumor (top right) compared to weak to absent signals in the normal appendix (bottom left). C: MUC2 immunocytochemistry reflects intense but focal MUC2 immunoreactivity limited to the goblet cells of the normal appendix compared to intense and homogeneous immunoreactivity in all of the proliferating epithelial cells of the mucinous borderline tumor of the appendix (D); this observation was further reflected by the anti-sense MUC2 in situ hybridization findings of intense but focal signals in the goblet cells of the normal appendix (E) compared to intense and homogeneous signals in all of the proliferating epithelial cells of the mucinous borderline tumor of the appendix (F). Original magnifications: ×150 (A, B); ×250 (C–F). A, H&E; B, MUC2 anti-sense riboprobe; C and D, anti-MUC2, immunoperoxidase; E and F, MUC2 anti-sense riboprobe.

Figure 5.

A: The extracellular mucin:cell ratio is quantitated in five individual cases of mucinous tumors of the appendix, primary ovarian mucinous tumors, and pseudomyxoma peritonei (PP). In each, results depict mean ± SD of 10 medium-power microscopic fields (×250). Higher mucin:cell ratios averaging 10:1 are in evidence in pseudomyxoma peritonei and in mucinous tumors of the appendix, suggesting that these two entities are related. In primary ovarian mucinous tumors, which included ovarian carcinomas with peritoneal implants, the mucin:cell ratio was much less and hovered around unity. B: The MUC2 and MUC5AC anti-sense in situ hybridization signals [fold increase over sense (background)] are quantitated in various mucinous lesions. Here mean signals from all of the individual cases were pooled and an overall mean ± SD was calculated. Mucinous tumors of the appendix and pseudomyxoma peritonei (PP) both exhibit strong signals with MUC2 more than MUC5AC. Primary ovarian mucinous tumors, in contrast, exhibit no MUC2 but slightly higher MUC5AC. C: Anti-sense MUC2 and MUC5AC in situ hybridization signals are quantitated in pseudomyxoma peritonei lesions showing varying degrees of transformation (benign, borderline, and malignant). Here mean signals from all of the individual cases of a given histology were pooled and an overall mean ± SD was calculated. The signals for MUC2 and MUC5AC do not vary with the degrees of transformation. D: Anti-sense MUC2 and MUC5AC in situ hybridization signals in pseudomyxoma peritonei compared to signals in goblet and nongoblet cells in the normal appendix are quantitated. Signals are increased in both pseudomyxoma peritonei as well as in normal goblet cells; digital image analysis of the in situ hybridization data from pseudomyxoma peritonei (E) and goblet cells in the normal appendix (F) confirm that on a per cell basis, there is no difference in the constitutive levels of expression (D).A–D, Comparative histograms; Original magnifications, ×250 (E), ×450 (F), E and F, MUC2 anti-sense riboprobe.

Figure 6.

A: Short-term cultures of pseudomyxoma peritonei grow as monolayers in which prominent cytoplasmic vacuoles are observed; B, these vacuoles are mucin vacuoles on ultrastructural analysis; C, anti-sense MUC2 in situ hybridization of a cell pellet of these cultured cells confirms intense MUC2 signals. Original magnifications: ×200 (A); ×3500 (B); ×150 (C). A, Phase contrast; B, uranyl acetate, lead citrate; C, MUC2 anti-sense riboprobe.

Figure 7.

Northern blots (10 μg of total RNA per lane) of pseudomyxoma peritonei cell cultures pretreated and treated with various agents. A: Treatment with an inhibitor of DNA methylation, 5-azacytidine (10 μmol/L), for 0 to 10 days increased steady-state MUC2 mRNA levels. B: Treatment with P. aeruginosa LPS (5 μg/ml), for 0 to 10 hours increased steady-state MUC2 mRNA levels approximately threefold to fivefold. Genistein (100 μg/ml), a tyrosine kinase inhibitor, when used to pretreat the cells, blocked the up-regulation of MUC2 × 5-azacytidine (C) as well as LPS (D).

Figure 8.

Schematic depicts our hypothesis that pseudomyxoma peritonei is a disease of MUC2-expressing goblet cells. In the normal colon (A), goblet cells comprise a minority of intestinal lining cells and constitutively express high levels of MUC2. This extracellular mucin is washed out in the colonic lumen and does not accumulate. Pseudomyxoma peritonei is a disease in which these MUC2-expressing goblet cells transform and proliferate, so instead of representing only a minority of the intestinal lining cells now represent all of the lining cells (B). The overall MUC2 expression and secretion rise dramatically in pseudomyxoma peritonei but, on a per cell basis, there is no difference in the constitutive levels of expression of MUC2, reflecting only the levels expressed by normal goblet cells. These MUC2-expressing goblet cells spread to the omentum and the mesentery either through appendiceal rupture or invasion, subsequent redistribution, or metastasis. In the omentum and mesentery the extracellular mucin cannot wash out the colonic lumen and instead accumulates as the gelatinous deposits of pseudomyxoma peritonei.