Establishment and characterization of a novel human myoepithelial cell line and matrix-producing xenograft from a parotid basal cell adenocarcinoma

Expression of Class II β-Tubulin by Proliferative Myoepithelial Cells in Canine Mammary Mixed Tumors

Benign mammary mixed tumors in dogs resemble human salivary pleomorphic adenomas with regard to their histogenesis, including the occurrence of cartilaginous or bony metaplasia as well as the expression pattern of cytoskeletal proteins in proliferative myoepithelial cells. Recently, a monoclonal antibody specific for class II β-tubulin has been developed. The epitope it recognizes was determined to be the hepta-peptide Glu-Glu-Glu-Glu-Gly-Glu-Asp, which is the common sequence found among the canine, rat, mouse, and human class II β-tubulin-specific regions. We carried out immunohistochemical studies on mammary mixed tumors obtained from three female dogs using this the monoclonal antibody. The antibody to class II β-tubulin reacted intensely with proliferative myoepithelial cells in canine mammary mixed tumors, whereas staining was barely detectable in normal myoepithelial cells surrounding alveoli and alveolar ducts within the tumor and adjacent normal tissue. Proliferative myoepithelial cells also expressed vimentin, but α-smooth muscle actin (αSMA) staining was barely detectable. Immunoblot analysis showed that class II β-tubulin and vimentin were expressed in myoepithelial cell lines prepared from the three mammary mixed tumors. On the other hand, only one cell line, which was negative for αSMA, produced cartilage-specific type II collagen. These results suggest that class II β-tubulin could be a new molecular marker of proliferating myoepithelial cells in canine mammary mixed tumors and that differential expression of cytoskeletal components is associated with cartilaginous metaplasia of proliferative myoepithelial cells in mixed mammary tumors.

Breast carcinomatous tumoral emboli can result from encircling lymphovasculogenesis rather than lymphovascular invasion

The canonical view of the origin of tumor lymphovascular emboli is that they usually originate from lymphovascular invasion as part of a multistep metastatic process. Recent experimental evidence has suggested that metastasis can occur earlier than previously thought and we found evidence that tumor emboli formation can result from the short-circuiting step of encircling lymphovasculogenesis. Experimentally, we used a xenograft of human inflammatory breast cancer (MARY-X), a model that exhibited florid tumor emboli, to generate tumoral spheroids in vitro. In observational studies, we chose human breast carcinoma cases where there appeared to be a possible transition of in situ carcinoma to lymphovascular emboli without intervening stromal invasion. These cases were studied by morphometry as well as IHC with tumor proliferation (Ki-67) and adhesion (E-cadherin) markers, myoepithelial (p63), as well as endothelial (podoplanin [D2-40], CD31, VEGFR-3, Prox-1) markers. Unlabelled spheroids coinjected with either GFP or RFP-human myoepithelial cells or murine embryonal fibroblasts (MEFs) gave rise to tumors which exhibited GFP/RFP immunoreactivity within the cells lining the emboli-containing lymphovascular channels. In vitro studies demonstrated that the tumoral spheroids induced endothelial differentiation of cocultured myoepithelial cells and MEFs, measured by real time PCR and immunofluorescence. In humans, the in situ clusters exhibited similar proliferation, E-cadherin immunoreactivity and size as the tumor emboli (p =.5), suggesting the possibility that the latter originated from the former. The in situclusters exhibited a loss (50%-100%) of p63 myoepithelial immunoreactivity but not E-cadherin epithelial immunoreactivity. The tumor emboli were mainly present within lymphatic channels whose dual p63/CD31, p63/D2-40 and p63/VEGFR-3 and overall weak patterns of D2-40/CD31/VEGFR-3 immunoreactivities suggested that they represented immature and newly created vasculature derived from originally myoepithelial-lined ducts. Collectively both experimental as well as observational studies suggested the possibility that these breast cancer emboli resulted from encircling lymphovasculogenesis rather than conventional lymphovascular invasion.

The lymphovascular embolus of inflammatory breast cancer exhibits a Notch 3 addiction

Inflammatory breast carcinoma (IBC) is characterized by exaggerated lymphovascular invasion (LVI), recapitulated in our human xenograft, MARY-X. This model exhibited lymphovascular emboli in vivo and corresponding spheroids in vitro. Owing to the morphological and gene profile resemblance of these spheroids to embryonal blastocysts, we wondered whether they might exhibit embryonic stem cell signaling. Specifically we investigated Notch and observed selective Notch 3 activation by expression profiling, reverse transcriptase- and real-time PCR, western blot and immunofluorescence in vitro, and immunohistochemistry in vivo. Notch 3 intracellular domain (N3icd) and six target genes, HES-5, HEY-1, c-Myc, Deltex-1, NRARP and PBX1, markedly increased in MARY-X. In addition, a significant percentage of MARY-X cells expressed aldehyde dehydrogenase (ALDH), a stem cell marker. Only the ALDH(+) cells were capable of secondary spheroidgenesis, tumorigenicity and self-renewal. Inhibiting Notch 3 activation in vitro with γ-secretase inhibitors (GSIs) or small interfering RNA resulted in a downregulation of Notch target genes, including CD133, and an induction of caspase 3-mediated apoptosis. Transfection of N3icd but not Notch 1 intracellular domain into normal human mammary epithelial cells resulted in increased expression of Notch target genes and induction of spheroidgenesis. GSI in vivo resulted in inhibitory but diffusion-limited effects on Notch 3 signaling, resulting in xenograft growth reduction. The lymphovascular emboli of human IBC exhibited dual N3icd and ALDH1 immunoreactivities independently of molecular subtype. This Notch 3 addiction of lymphovascular emboli might be exploited in future therapeutic strategies.

The lymphovascular embolus of inflammatory breast cancer expresses a stem cell-like phenotype

Inflammatory breast carcinoma (IBC) is a particularly lethal form of breast cancer characterized by exaggerated lymphovascular invasion, which is a phenotype recapitulated in our human xenograft MARY-X. MARY-X generated spheroids in vitro that resemble the embryonal blastocyst. Because of the resemblance of the spheroids to the embryonal blastocyst and their resistance to traditional chemotherapy/radiotherapy, we hypothesized that the spheroids expressed a stem cell-like phenotype. MARY-X spheroids expressed embryonal stem cell markers including stellar, rex-1, nestin, H19, and potent transcriptional factors, oct-4, nanog, and sox-2, which are associated with stem cell self-renewal and developmental potential. Most importantly, MARY-X spheroids expressed a cancer stem cell profile characterized by CD44(+)/CD24(-/low), ALDH1, and most uniquely, CD133. A significant percentage of single cells of MARY-X exhibited distinct proliferative and morphogenic potencies in vitro. As few as 100 cells derived from single-cell clonogenic expansion were tumorigenic with recapitulation of the IBC phenotype. Prototype stem cell signaling pathways such as notch3 were active in MARY-X. The stem cell phenotype exhibited by MARY-X also was exhibited by the lymphovascular emboli of human IBC cases independent of their molecular subtype. This stem cell-like phenotype may contribute to the aggressive nature of IBC but also may lend itself to selective targeting.

Mechanisms of disease: breast tumor pathogenesis and the role of the myoepithelial cell

Breast cancer and precancer cells are influenced by important paracrine regulation from the breast microenvironment, which might be as great a determinant of breast cancer behavior as the specific oncogenic or tumor-suppressive alterations occurring within malignant breast cells. Myoepithelial cells exert profound effects on breast tumor cell behavior, and lie in juxtaposition to abnormally proliferating breast epithelial cells in precancerous disease states such as ductal carcinoma in situ (DCIS). Myoepithelial cells also form a natural border separating breast epithelial cells from stromal angiogenesis. These anatomical relationships suggest that myoepithelial cells might inhibit both the progression of DCIS to invasive breast cancer, and carcinoma-induced angiogenesis. Our ability to study myoepithelial cells has been fostered by recent technical advances in cell selection and sorting procedures, improved selective media, and high throughput technologies, which are able to assess the gene and protein expression profiles within cells. In addition, the establishment of a number of immortalized cell lines and xenografts of myoepithelial cells derived from benign human myoepithelial tumors of diverse sources has provided a self-renewing cell source through which to study the phenotype of myoepithelial cells. Studies of primary and immortalized myoepithelial cell lines indicate that these cells exhibit a natural tumor suppressor function. Functional studies show that these cells have anti-invasive and antiangiogenic phenotypes.

Do myoepithelial cells hold the key for breast tumor progression?

Mammary myoepithelial cells have been a neglected facet of breast cancer biology, largely ignored since they have been considered to be less important for tumorigenesis than luminal epithelial cells from which most of breast carcinomas are thought to arise. In recent years as our knowledge of stem cell biology and the cellular microenvironment has been increasing, myoepithelial cells are slowly starting to gain more attention. Emerging data raise the hypothesis whether myoepithelial cells play a key role in breast tumor progression by regulating the in situ to invasive carcinoma transition and that myoepithelial cells are part of the mammary stem cell niche. Paracrine interactions between myoepithelial and luminal epithelial cells are known to be important for regulation of cell cycle progression, establishing epithelial cell polarity, and inhibiting cell migration and invasion. Based on these functions, normal mammary myoepithelial cells have been called "natural tumor suppressors." However, during tumor progression myoepithelial cells seem to loose these properties, and eventually this cell population diminishes as tumors become invasive. Better understanding of myoepithelial cell function and their role in tumor progression may lead to their exploitation for cancer therapeutic and preventative measures.

Myoepithelial cells: autocrine and paracrine suppressors of breast cancer progression

Host cellular paracrine regulation of tumor progression is an important determinant of tumor biology but one cell that has been ignored in this regulation is the myoepithelial cell. Myoepithelial cells surround normal ducts and precancerous lesions, especially of the breast and form a natural border separating proliferating epithelial cells from proliferating endothelial cells (angiogenesis). Myoepithelial cells may thus negatively regulate tumor invasion and metastasis. Whereas epithelial cells are susceptible targets for transforming events, myoepithelial cells are resistant. Therefore, it can be said that myoepithelial cells function as both autocrine as well as paracrine tumor suppressors. Our laboratory has found that myoepithelial cells secrete a number of suppressor molecules including high amounts of diverse proteinase inhibitors and angiogenic inhibitors but low amounts of proteinases and angiogenic factors compared to common malignant cell lines. This observation has been made in vitro, in mice, and in humans and suggests that myoepithelial cells exert pleiotropic suppressive effects on tumor progression. The gene expression profile of myoepithelial cells may explain the pronounced anti-invasive and anti-angiogenic effects of myoepithelial cells on carcinoma cells and may also account for the reduced malignancy of myoepithelial tumors, which are devoid of appreciable angiogenesis and invasive behavior.

Myoepithelial mRNA expression profiling reveals a common tumor-suppressor phenotype

A series of myoepithelial cell lines and xenografts derived from benign human myoepithelial tumors of diverse sources (salivary gland, breast, and lung) exhibit common mRNA expression profiles indicative of a tumor-suppressor phenotype. Previously established myoepithelial cell lines and xenografts (HMS-#; HMS-#X) were compared to nonmyoepithelial breast carcinoma cells (MDA-MB-231 and MDA-MB-468, and inflammatory breast carcinoma samples, IBCr, and IBCw), a normal mammary epithelial cell line (HMEC) and individual cases of human breast cancer (zcBT#T), and matched normal human breast tissues (zcBT#N) (overall samples = 22). The global gene expression profile (22,000 genes) of these individual samples was examined using Affymetrix Microarray Gene Chips and subsequently analyzed with both Affymetrix and DChip algorithms. The myoepithelial cell lines/xenografts were distinct and very different from the nonmyoepithelial breast carcinoma cells and the normal breast and breast tumor biopsies. Two hundred and seven specifically selected genes represented a subset of genes that distinguished (P < 0.05) all the myoepithelial cell lines/xenografts from all the other samples and which themselves exhibited hierarchical clustering. Further analysis of these genes revealed increased expression in genes belonging to the classes of extracellular matrix proteins, angiogenic inhibitors, and proteinase inhibitors and decreased expression belonging to the classes of angiogenic factors and proteinases. Developmental genes were also differentially expressed (either over or underexpressed). These studies confirm our previous impression that human myoepithelial cells express a distinct tumor-suppressor phenotype.

Establishment of an immortalized cell line and transplantable xenograft from a bronchioloalveolar lung carcinoma of a cat

OBJECTIVE

To establish an immortalized cell line and transplantable xenograft of feline bronchioloalveolar lung carcinoma (BAC).

SAMPLE POPULATION

Pleural effusion from a 12-year-old Persian male cat with BAC.

PROCEDURE

Tumor cells from the pleural effusion were grown in monolayer cell culture and injected into severe combined immunodeficient (SCID) mice to establish an immortalized cell line as well as a transplantable xenograft.

RESULTS

Both the primary lung carcinoma, the derived cell line, and the transplantable xenograft had evidence of a type-II pneumocyte origin expressing lamellar bodies ultrastructurally and thyroid transcription factor-1 and surfactant immunocytochemically. All 3 also expressed nuclear p53 immunoreactivity. A metaphase spread of the cell line (SPARKY) probed with fluorescein-labeled genomic feline DNA gave evidence of its feline origin. Flow cytometric studies indicated aneuploidy with a DNA index of 1.6. An R-banded karyotype revealed a modal number of 66 including the feline Y chromosome. The cell line had a doubling time of 16 hours. The xenograft (SPARKY-X) reached a diameter of 1 cm in 3 weeks in SCID mice. Deoxyribonucleic acid fingerprint analysis revealed that SPARKY and SPARKY-X were novel and strongly matched each other, except for the murine component found in SPARKY-X. Interestingly, SPARKY-X manifested the characteristic lepidic growth pattern of pulmonic BAC.

CONCLUSIONS

Both the cell line and xenograft retained their autochthonous BAC phenotype, making them useful for the subsequent dissection of molecular abnormalities in feline BAC and in vitro screening of chemotherapeutic agents.

Myoepithelial-specific CD44 shedding is mediated by a putative chymotrypsin-like sheddase

Our previous studies have demonstrated that myoepithelial cells, which surround incipient carcinomas in situ of the breast and other organs, exert antiinvasive and antiangiogenic effects in vitro through the elaboration of a number of different suppressor molecules among which include the shed membrane CD44. The present study addresses the mechanism of this myoepithelial CD44 shedding. This CD44 shedding is enhanced by PMA pretreatment, is specific for myoepithelial CD44, and inhibited by chymotrypsin-like inhibitors (chymostatin, alpha(1)-antichymotrypsin, TPCK, and SCCA-2) but not by trypsin-like inhibitors (TLCK), nor papain-like inhibitors (SCCA-1) nor hydroxamate-based or general metalloproteinase inhibitors (BB2516 (marimastat), 1,10-phenanthroline, and TIMP-1). The effect of PMA can be mimicked by exogenous chymotrypsin but not by other proteases. The CD44 shedding activity cannot be transferred by conditioned media, cell-cell contact, peripheral membrane, or integral membrane fractions. However, cell-free purified integral plasma membrane fractions obtained from myoepithelial cells pretreated with PMA also exhibit CD44 shedding which is inhibited by chymotrypsin-like inhibitors. These findings support the presence and activation of a putative chymotrypsin-like sheddase as the mechanism of CD44 shedding in myoepithelial cells.

Myoepithelial-specific CD44 shedding contributes to the anti-invasive and antiangiogenic phenotype of myoepithelial cells

Myoepithelial cells surround incipient ductal carcinomas of the breast and exert anti-invasive and antiangiogenic effects in vitro through the elaboration of suppressor molecules. This study examines one putative molecule, solubilized CD44 produced by myoepithelial shedding of membrane CD44. Studies with different human myoepithelial cell lines demonstrate that myoepithelial cells express and shed both the 85-kDa standard (CD44s) and the 130-kDa epithelial (CD44v8-10) isoforms, findings further confirmed by the use of isoform-specific antibodies. PMA pretreatment enhances CD44 shedding detected by two different methods at different time points: a reduction in surface CD44 at 2 h by flow cytometry and a marked decrease in both total cellular CD44 and plasma membrane CD44 at 12 h by Western blot. This shedding is both specific for CD44 and specific for myoepithelial cells. This shedding is inhibited by the chymotrypsin inhibitors chymostatin and alpha(1)-antichymotrypsin but not by general metallo-, cysteine, or other serine proteinase inhibitors. Myoepithelial-cell-conditioned medium and affinity-purified solubilized CD44 from this conditioned medium block hyaluronan adhesion and migration of both human carcinoma cell lines and human umbilical vein endothelial cells.

Tamoxifen enhances myoepithelial cell suppression of human breast carcinoma progression in vitro by two different effector mechanisms

Our previous studies have indicated that myoepithelial cells surrounding ductal and acinar epithelium of glandular organs, such as the breast, exert multiple paracrine suppressive effects on incipient and developing cancers that arise from this epithelium. Myoepithelial cells and derived cell lines (HMS 1-6) exert these effects through the secretion of a number of different effector molecules that exert anti-invasive, anti-proliferative, and anti-angiogenic activities. Since previous basic and clinical studies have examined the role of estrogen agonists and antagonists on human breast cancer cells and because issues of hormone replacement therapy (HRT) and tamoxifen chemoprevention are such timely issues in breast cancer, we wondered whether or not hormonal manipulations might affect myoepithelial cells in vitro as far as their paracrine suppressive activities on breast cancer were concerned. The present in vitro study demonstrates that treatment of myoepithelial cells with tamoxifen but not 17beta-estradiol increases both maspin secretion and invasion-blocking ability. Furthermore tamoxifen but not 17beta-estradiol increases inducible nitric oxide synthase (iNOS) expression and nitric oxide (NO) production by myoepithelial cells when they are co-cultured with conditioned media from or breast carcinoma cells directly. This increased myoepithelial NO exerts both autocrine and paracrine antiproliferative effects which can be blocked by inhibition of iNOS. 17beta-Estradiol, however, competes with all of these suppressive effects of tamoxifen suggesting that the mechanism of tamoxifen action is estrogen receptor mediated. Myoepithelial cells lack ER-alpha but express ER-beta. Tamoxifen, but not 17beta-estradiol, increases AP-1 CAT but not ERE-CAT activity. Again, 17beta-estradiol competes with the transcription-activating effects of tamoxifen. These experiments collectively suggest that the actions of tamoxifen on the increased secretion of maspin and increased production of NO by myoepithelial cells are mediated through ER-beta and the transcription-activation of an ER-dependent AP-1 response element.

The human myoepithelial cell displays a multifaceted anti-angiogenic phenotype

Human myoepithelial cells which surround ducts and acini of certain organs such as the breast form a natural border separating epithelial cells from stromal angiogenesis. Myoepithelial cell lines (HMS-1-6), derived from diverse benign myoepithelial tumors, all constitutively express high levels of active angiogenic inhibitors which include TIMP-1, thrombospondin-1 and soluble bFGF receptors but very low levels of angiogenic factors. These myoepithelial cell lines inhibit endothelial cell chemotaxis and proliferation. These myoepithelial cell lines sense hypoxia, respond to low O2 tension by increased HIF-1 alpha but with only a minimal increase in VEGF and iNOS steady state mRNA levels. Their corresponding xenografts (HMS-X-6X) grow very slowly compared to their non-myoepithelial carcinomatous counterparts and accumulate an abundant extracellular matrix devoid of angiogenesis but containing bound angiogenic inhibitors. These myoepithelial xenografts exhibit only minimal hypoxia but extensive necrosis in comparison to their non-myoepithelial xenograft counterparts. These former xenografts inhibit local and systemic tumor-induced angiogenesis and metastasis presumably from their matrix-bound and released circulating angiogenic inhibitors. These observations collectively support the hypothesis that the human myoepithelial cell (even when transformed) is a natural suppressor of angiogenesis. Oncogene (2000) 19, 3449 - 3459

Matrix-producing carcinoma of the breast: case report with radiographical and cytopathological features

A case of matrix-producing carcinoma (MPC), which is a unique variant of breast cancer, is described with cytopathological and radiographical findings. A 75-year-old female presented with a palpable mass in her right breast. After confirmation of the cytopathological diagnosis, the patient received a right mastectomy with ipsilateral axillary lymph node dissection. The histopathological examination revealed MPC. The Union Internationalis Contra Cancrum postoperative clinical stage was IIB. The patient is well and without disease 18 months after the surgery.

The human myoepithelial cell exerts antiproliferative effects on breast carcinoma cells characterized by p21WAF1/CIP1 induction, G2/M arrest, and apoptosis

Ductal carcinoma in situ of the breast (DCIS) is surrounded by a layer of myoepithelial cells. Our previous studies have suggested that these myoepithelial cells exert paracrine tumor-suppressive effects on invasion of breast carcinoma cells. Conditioned medium (CM), concentrated 10-100x of HMS-1, HMS-3, and HMS-4, human myoepithelial cell lines, block Matrigel invasion of a series of carcinoma cell lines. Immunoprecipitation of maspin, a recently described serpin, from these CM abolishes this anti-invasive effect. Both CM and maspin-immunoprecipitated CM, however, exert equal antiproliferative effects on a series of ER+ and ER- cell lines including MCF-7, T47D, MDA-MB-231, and MDA-MB-468. These antiproliferative effects are characterized by induction of a G2/M arrest, a twofold increase in p21(WAF1/CIP1) transcription and expression, and a threefold increase in apoptosis in the breast carcinoma lines examined. The antiproliferative effects mediated by myoepithelial cell CM do not manifest themselves in an autocrine manner, are not mediated by TGF-beta1, nor involve ER- or p53-dependent pathways. Neither the antiproliferative nor the anti-invasive effects of myoepithelial cell CM is observed with nonmyoepithelial cell CM. The in vitro observations of our present study may have relevance in explaining the increased degree of apoptosis exhibited by DCIS cells in vivo. Our findings illustrate another way myoepithelial cells function as natural paracrine tumor suppressors.

Humatrix, a novel myoepithelial matrical gel with unique biochemical and biological properties

Myoepithelial cells in situ and in vitro exert important paracrine effects on carcinoma cells which are mediated by high expression of extracellular matrix molecules, proteinase inhibitors and angiogenic inhibitors. Myoepithelial xenografts (human matrix secreting (HMS)-X, HMS-3X and HMS-4X) established from benign human salivary gland and breast myoepithelial tumors accumulate an abundant extracellular matrix which can be extracted with 6 M urea and 2 M guanidinium hydrochloride to form a gel at 25-37 degrees C. This gel, termed Humatrix, exhibits different biochemical and biological properties than the conventional non-human matrical gels in existence, i.e. Matrigel and Vitrogen 100. Whereas Matrigel consists mainly of basement membrane molecules, e.g. laminin, type IV collagen and heparan sulfate proteoglycan, and Vitrogen 100 consists mainly of non-basement membrane molecules, e.g. type I and type III collagen, Humatrix contains significant amounts of both basement membrane and non-basement membrane molecules, including large amounts of chondroitin sulfate proteoglycan. Like Matrigel, Humatrix contains bound growth factors, including epidermal growth factor (EGF) and insulin-like growth factor-I (IGF-I); unlike Matrigel, which contains predominantly significant quantities of bound proteinases, including tissue-type plasminogen activator (tPA), matrix metalloproteinase (MMP)-2 and MMP-9, and angiogenic factors, including basic fibroblast growth factor (bFGF) and transforming growth factor (TGF)-beta, Humatrix contains predominantly bound proteinase inhibitors such as protease nexin II (PN-II) and alpha1-antitrypsin and angiogenic inhibitors such as thrombospondin-1. Humatrix selectively stimulates the growth and tumorigenicity of human myoepithelial cell lines but inhibits invasion, angiogenesis and metastasis of other non-myoepithelial malignant cell lines. Because of its myoepithelial origin Humatrix represents a more natural source of extracellular matrix molecules and bound factors that carcinoma cells encounter in vivo.

'Revertant' DCIS in human axillary breast carcinoma metastases

Recent experimental evidence obtained in Scid mice has suggested that the metastatic process is in large part epigenetically regulated and undergoes partial reversion once the metastatic process is completed: the metastatic colonies become more engaged in the process of growing in situ than actively metastasizing. Based on this experimental evidence, examples were sought of metastatic human cancers where similar reversion to an in situ growth state was occurring. Review of 200 cases of metastatic human breast cancer revealed a 21 per cent incidence of reversion to a ductal carcinoma in situ (DCIS) growth pattern within axillary nodal metastases. The revertant DCIS areas were characterized by an intact and circumferential basement membrane, as demonstrated by extracellular laminin and type IV collagen immunoreactivity. These revertant DCIS areas could be distinguished from primary DCIS, however, by the absence of surrounding myoepithelial cells in the former, identified in the latter by their positive maspin, S-100, and smooth muscle actin immunoreactivity. The pattern of revertant DCIS, poorly differentiated (comedo) (13 per cent), intermediate (non-comedo) (6 per cent), or well-differentiated (non-comedo) (2%), exhibited complete 100 per cent concordance with the primary DCIS pattern. The concordance of histological patterns held true for even the subtypes of DCIS determined by architectural pattern, such as the micropapillary or cribriform subtypes. Nuclear size by digital image analysis and Her-2/neu, p53, and Ki-67 status in the revertant DCIS also exhibited complete concordance with the primary DCIS counterparts. Cases exhibiting a revertant DCIS pattern tended to be ER-negative/EGFR-positive and exhibited significant nodal involvement (mean number, 9; mean area, 90 per cent) compared with cases lacking a revertant pattern (mean number, 4; mean area, 15 per cent) (P < 0.01) These findings suggest that reversion of the metastatic phenotype may also be occurring within autochthonous human metastasis.

The myoepithelial defense: a host defense against cancer

The behavior of human tumors depends not only on the nature of the tumor cells themselves but also on the modifying effects of various normal host cells such as fibroblasts and endothelial cells. One cell type, however--the myoepithelial cell--has not been studied scientifically. Myoepithelial cells normally surround ducts and acini of glandular organs such as the breast and salivary glands and contribute to the synthesis of a surrounding basement membrane. This relationship suggests that myoepithelial cells may exert paracrine effects on glandular epithelium and also regulate the progression of ductal carcinoma in situ (DCIS) to invasive carcinoma. Myoepithelial tumors, in turn, tend to be benign or low-grade neoplasms that exhibit the rare property of accumulating rather than degrading extracellular matrix material. To better understand the nature of myoepithelial tumors, as well as the possible role of normal myoepithelial host cells in cancer, we have established immortal cell lines and a number of transplantable xenografts from various human myoepithelial tumors of the salivary gland and breast. The cell lines exhibit a normal myoepithelial phenotype and the xenografts continue to accumulate an abundant extracellular matrix. Further ultrastructural, immunocytochemical, molecular, and biochemical studies reveal that myoepithelial cells secrete relatively low levels of matrix-degrading proteinases but relatively high levels of maspin and various other anti-invasive proteinase inhibitors, that some of these inhibitors accumulate within the myoepithelial matrix, and that myoepithelial cells can induce epithelial morphogenesis (spheroid formation) and inhibit tumor-cell invasion in vitro. Myoepithelial cells, which surround normal breast ducts and DCIS, have also been found to selectively express maspin and certain proteinase inhibitors in situ. These inherent myoepithelial properties are likely to contribute to the low-grade nature of myoepithelial neoplasms and advance our hypothesis that host myoepithelial cells regulate the progression of in situ to invasive carcinoma by providing an important host defense against cancer invasion.