I. Introduction

The epithelium of normal and pre-invasive human breast tumor tissues is physically separated from the stroma by both the myoepithelial (ME) cell and basement membrane (BM). ME cells are joined by intercellular junctions and adhesion molecules, constituting a largely continuous sheet that encircles the epithelium (Murad and von Haam, 1986; Tsubura et al, 1988; Guelstein et al, 1993). The BM is composed of type IV collagens, laminins, and other molecules, forming a continuous lining surrounding and attaching to the ME cell layer (Nerlich, 1995; Damiani et al, 1999; Miosge, 2001). Since the epithelium is normally devoid of blood vessels and lymphatic ducts, its metabolism needed oxygen, nutrients, and growth factors must first pass through the BM, then the ME layer, in order to reach the epithelium. In contrast, epithelium-derived tumor cells must first pass through the ME layer, then the BM, in order to invade or metastasize.

Breast cancer invasion has been attributed primarily, if not solely, to the over-production of proteolytic enzymes predominantly by tumor cells, which results in the degradation of the BM (Goldfarb and Liotta, 1986; Duffy et al, 2000). This theory alone, however, appears not to fully reflect the intrinsic mechanism of breast tumor invasion for three main reasons. First, the mechanism and process of the ME cell layer degradation are unknown. Second, results from all recent worldwide clinical trials with specific enzyme inhibitors have been very disappointing (Coussens et al, 2002; Matrisian et al, 2003). Third, our previous studies detected a significantly higher level of matrix metalloproteinase-26 (MMP-26), a key enzyme for BM degradation, and its mediated pro-gelatinase B in ductal carcinoma in situ (DCIS) than in invasive carcinoma using immunohistochemistry and an integrated morphometry analysis (Zhao et al, 2004). Our recent studies further revealed that the invasive front of early invasive lesions in over 50 cases examined completely lacked expression of MMP-26, whereas cells within the same tumor and in adjacent micro-invasive lesions were strongly positive for MMP-26 (Man et al, 2005a,b). Since over 90% of breast cancer related death is caused by invasion related illness (Parker et al, 1997), there is an urgent need to disclose the intrinsic mechanism of, and to develop more effective approaches to prevent, breast tumor invasion.

Promoted by the fact that several tumor suppressors are exclusively produced by the ME cells and that the absence of the ME cell layers is the most distinct sign of invasive or metastatic breast lesions (Murad et al, 1986; Tsubura et al, 1988; Guelstein et al, 1993; Zou et al, 1994; Barbareschi et al, 2001; Man et al, 2002a), our recent studies have attempted to identify the early signs of ME cell layer alterations and their impact on biological presentations of tumor cells. Using a double immunohistochemical method to simultaneously elucidate ME and tumor cells, our initial study detected 405 focal ME cell layer disruptions (the absence of ME cells resulting in a gap greater than the combined size of at least 3 ME cells) in 5,698 duct cross sections from 220 female patients with estrogen receptor (ER) positive, pre-invasive breast tumors (Man et al, 2003a). Of these disruptions, 367 (90.6%) occurred in malignant lesions and 38 (9.4%) in hyperplastic or normal appearing ducts. Focal disruptions in ME cell layers appeared to substantially impact the ER expression status in the overlying epithelial cells. A vast majority (86.4%) of the cell clusters overlying disruptions completely lacked ER expression, in sharp contrast to adjacent cells within the same duct but distant from disruptions, which were strongly and uniformly positive for ER. Compared to their adjacent ER positive counterparts, ER NCC displayed several unique features, including: [1] a significantly higher proliferation rate, [2] a significantly higher frequency of loss of heterozygosity at multiple chromosomal loci; [3] a significantly higher expression of cell cycle control and tumor invasion related genes; [4] physical signs of stromal and vascular invasion (Man et al, 2002b,c, 2003a,b, 2004a, 2005c; Man and Sang, 2004). Together, our findings suggest that focal ME cell layer disruptions substantially impact the biological presentations of the associated epithelial cells, and that ER NCC might represent the precursor of invasive lesions, which are in the process, or at the early stage of invasion.

Our findings are in total agreement with previous reports that breast tumor progression is paralleled by a progressive hormonal independence, and that ER negative tumors have a substantially more aggressive clinical behavior and worse prognosis than ER positive tumors (Clarke et al, 1989; Schmitt, 1995; Sheikh et al, 1994, 1995; Rochefort et al, 2003). Our findings are also consistent with previous findings that breast tumor progression from one stage to another is driven by sequential expression of stage-specific molecules and the selection of biologically more aggressive cell clones (Pitot, 1993; Beckmann, 1997; Lakhani, 1999). Our findings, however, are hard to reconcile with the fact that over 80% of invasive breast cancers are reported to be ER positive (Luna-More et al, 2000; Swain et al, 2004).

Our current study intended to identify potential factors accounting for this discrepancy. For the following three reasons: [1] it has been documented that most malignant tumors arise from a single cell (Nowell, 1976; Sell et al, 1994; Kordon and Smith et al, 1998; Middleton et al, 2000), [2] epithelial cell “budding” is a common event shared by both normal development and tumor invasion, which occurred exclusively at the site of focally degraded BM (Yang et al, 2003; Lu et al, 2005; Jourquin et al, 2006), and [3] our previous studies in both breast and prostate tumors have revealed that a vast majority of the proliferating cells are located at the site of focal ME or basal cell layer disruptions, and that tumor cells “budding” from these disruptions are often in direct continuity with invasive lesions (Man et al, 2002b,c, 2003a,b, 2004a, 2005d in press; Man and Sang, 2004; Yousefi et al, 2005), we have hypothesized that ER NCC and their “budding” derivatives are derived from a monoclonal proliferation of the same progenitor, whereas they are at different differentiation stages with a different ER expression status. In addition, as our previous studies in normal and surgically operated adult submandibular glands using the combination of immunohistochemistry and autoradiography had shown that newly formed cell clusters by stem cells completely lacked the expression of several proteins that were heavily expressed in their adjacent adult counterparts, while they gradually gained the expression of these molecules with time (Man et al, 1995, 2001a), we have further hypothesized that ER NCC may represent tumor progenitors that are not mature enough to manufacture ER and other molecules, or are with “resting” genes for these molecules. The derivatives of ER NCC, however, may gradually gain the expression of these molecules after “budding” from focally disrupted ME layers and invading deeper into the stroma, due to a programmed genetic sequence or direct interactions with stromal or immunoreactive cells.

Our current study intended to identify potential factors accounting for this discrepancy. Since our previous studies in normal and surgically operated adult submandibular glands with a combination of autoradiography and immunohistochemistry had revealed that newly formed cell clusters by stem cells completely lacked the expression of several proteins that were heavily expressed in their adjacent adult counterparts, whereas they gradually gained the expression of these molecules with time (Man, 1995, 2001a), we hypothesized that ER NCC might represent tumor progenitors that are not mature enough to manufacture ER and other molecules, or are with “resting” genes for these molecules. The derivatives of ER NCC, however, might gradually gain the expression of these molecules after “budding” from focal ME cell layer disruptions and invading deeper into the stroma, probably due to a programmed genetic sequence or direct interactions with stromal or immunoreactive cells.

II. Materials and methods

A total of 100 formalin-fixed, paraffin-embedded human mammary tumors with co-existing pre-invasive, invasive, or micro-invasive lesions were retrieved from the files of The Armed Forces Institute of Pathology and our own tissue bank. Consecutive sections at 4-5 mm thickness were made, placed on positively charged microscopic slides, and stained with H&E for morphological classification, based on our published criteria (Tavassoli and Man, 1995).

To identify ER NCC and their potential derivatives (defined as cell clusters that directly “bud” or “sprout” from, but maintain direct physical continuity with, ER NCC overlying focally disrupted ME cell layers), sections were subjected to double immunohistochemical staining for smooth muscle actin (SMA) and ER (Vector, Burlingame, CA), as previously described (Man and Tavassoli, 1996).

Immunostained sections were independently examined by the investigators to identify pure ER NCC and larger ER NCC (more than 15 cells/cluster) with “budding” derivatives. To compare the ER expression profiles between ER NCC and their potential derivatives, four technical approaches were used. First, the morphological and immunohistochemical profiles between pure ER NCC and ER NCC with “budding” derivatives were compared. Second, the derivatives of ER NCC were artificially divided into different parts based on their locations (see “Results” for detail), and the ER expression status in each part was compared. Third, sections were examined to identify microinvasive lesions that were immediately adjacent to ER NCC, and the number of ER positive cells and intensity of ER immunostaining in microinvasive lesions at different distances to focal ME cell layer disruptions were compared. Fourth, the morphological and cytological features of ER NCC and their derivatives were compared.

To assess the possibility that ER NCC and their derivatives may respond differently to hormonal and drug therapies, sections were double immunostained for SMA and topoisomerase II alpha (TOPOIIa), which is an essential nuclear enzyme involved in DNA replication and is the target for many anti-cancer drugs used for cancer therapy (Houlbrook, 1995). The loss or reduction has been reported to be the main mechanism for drug resistance (Houlbrook et al, 1995). The expression status of TOPOIIa in ER NCC and their derivatives was compared.

To assess the possibility that ER NCC and their derivatives may have different expression profiles of tumor progression- and invasion-related molecules, sections were double immunostained for SMA and MMP-26 (Zhao et al, 2004), or for SMA and beta protein 1 (BP1), a homobox gene product that was preferentially seen in over 80% of the invasive breast lesions (Man et al, 2005e).

As our previous studies have consistently shown that focally disrupted ME cell layers have a significantly higher infiltration of immunoreactive cells (Man et al, 2005f; Man, 2005g; Yousefi et al, 2005), which have been reported to directly impact the gene expression and proliferation of associated tumor cells (Freeman et al, 1995; Takahashi et al, 1996; Asano-kato, et al 2005; Nienartowicz et al, 2006; Qu, 2006), selected sections were double immunostained for SMA and leukocyte common antigen (LCA).

To assess the histological origin of ER NCC and their derivatives, sections were immunostained with three epithelial specific markers, cytokeratins (CK) AE1/AE3, epithelial specific antigen (ESA), and E-cadherin (Vector, Burlingame, CA). Sections were also immunostained with two stromal cell specific markers, SMA and vimentin (Vector, Burlingame, CA).

III. Results

Of 100 selected cases, 52 (52%) contained pure ER NCC, and 17 (17%) contained a total of 26 larger ER NCC with “budding” derivatives. All or nearly all the cells in pure ER NCC seen in our current study were completely devoid of ER expression, identical to these seen in our previous studies (Figure 1). Compared to pure ER NCC, ER NCC with derivatives showed three unique features: [1] the size was substantially larger; [2] the clusters were more deeply “puncturing” into the stroma; [3] a variable number of ER positive cells were also present.

The “budding” derivatives of ER NCC were generally arranged as tongue-like projections, “puncturing” into the stroma (Figure 2). These “buds” could be roughly divided into three parts: the base, tip, and shaft, which are in a direct continuity with the tumor core and disruption, within the stroma, and between the base and tip, respectively. All or nearly all the cells at the base were completely devoid of ER expression (Figure 2). A vast majority of the cells at the shaft were also completely devoid of ER expression, while ER positive cells were occasionally seen (Figure 2). The number of ER positive cells and intensity of ER immunostaining appeared to be linearly increasing as tumor cells invading deeper into the stroma (Figure 2). Over 75% of the budding ER NCC were seen in DCIS (Figures 2a-2f). About 20% of the budding ER NCC were seen in both normal or hyperplastic appearing ducts (Figure 2g-2h).

In addition to “budding” ER NCC, three cases contained micro-invasive breast lesions that were immediately adjacent to “budding” ER NCC (Figure 3). All or nearly all the cells overlying ME cell layer disruptions were completely devoid of ER expression. The number of ER positive cells and intensity of ER immunostaining in these micro-invasive lesions appeared to increase linearly as they moved away from the tumor core and focal ME cell layer disruptions (Figure 3).

Figure 5. MMP-26 and BP1 expression in ER NCC and their derivatives Paraffin-embedded breast tissue sections were immunostained for MMP-26 (a & b) and BP1 (c & d). Circles identify ER NCC, which are negative for MMP-26 and BP1. Arrows identify “budding” derivatives and micro-invasive lesions, which are positive for MMP-26 and BP1. a & c: 150X. b & d: a higher magnification (400X) of a & c, respectively.

Table 1. Comparison of expression of different molecules in ER NCC and their budding derivatives

Cell Type	Total number	ER (+)	TOPOIIa (+)	MMP-26 (+)	BP1 (+)
ERNCC	26	0	0	0	0
Derivatives	26	23	21	20	21

Figure 6. Increased leukocyte infiltration in ER NCC with “budding” derivatives Paraffin-embedded breast tissue sections were double immunostained for SMA (red) and leukocyte common antigen (LCA; brown). Thick arrows identify LCA positive cells. Thin arrows identify the remaining ME cell layers. a, c, and e: 150X. b, d, and f: a higher magnification (400X) of a, c, and e, respectively.

It is interesting to note that no distinct “budding” ER positive cell clusters were found in any of these selected cases. The lack of corresponding frozen tissues in these cases and the lack of previously published references on ER NCC and their potential derivatives, however, have hampered our efforts of further genetic assessments, including clonality analysis, on these cell clusters.

IV. Discussion

Our previous studies revealed that a subset of pre-invasive, ER positive mammary tumors contained focally disrupted ME cell layers (Man et al, 2003a). All or nearly all of the cells overlying these focal disruptions were uniformly devoid of ER expression, in a sharp contrast to their adjacent counterparts within the same duct but overlying the remaining non-disrupted ME cell layer, which were strongly positive for ER (Man et al, 2003a). The frequency and pattern of ME cell layer disruptions detected in our current study in tumors with co-existing in situ and invasive components were similar to those seen in our previous studies. The ER NCC seen in our current study, however, differed from those seen in our previous studies in three main expects: [1] the size was substantially larger; [2] the clusters were more deeply “puncturing” into the stroma; [3] a variable number of ER positive cells were also present.

Despite these differences, ER NCC seen in our previous and current studies are likely to be derived from a monoclonal proliferation of the same progenitor for several reasons. First, they are exclusively located at focally disrupted ME cell layers. Second, cells within each or among different clusters share the same morphological and immunohistochemical features at the site of, or near the focal disruptions. Third, they are morphologically distinct from adjacent cells within the same duct in the cell density, size, nuclear shape, and polarity. More importantly, our previous studies have revealed that ER NCC from different cases share a very similar genetic and immunohistochemical profile, but they show a significantly higher rate of cell proliferation, genetic instabilities, and expression of tumor invasion and progression related genes than their adjacent counterparts within the same duct (Man et al, 2002b,c, 2003a, 2004a, 2005c; Man and Sang, 2004). Our findings and hypothesis are in total agreement with a number of previous reports: [1] a vast majority of malignant tumors are derived from monoclonal proliferation of a single tumor stem cell (Nowell, 1976; Sell, 1994; Kordon, 1998; Middleton, 2000), [2] epithelial cell “budding” is a common event shared by both normal development and tumor invasion, which occurred exclusively at the site of focally degraded BM (Yang, 2003; Lu, 2005; Jourquin, 2006), and [3] our previous studies in both human breast and prostate tumors have consistently shown that a vast majority of the proliferating cells are located at the site of focal ME or basal cell layer disruptions, and that tumor cells “budding” from these disruptions are generally in direct continuity with the adjacent invasive lesions (Man et al, 2002b,c, 2003a,b, 2004a, 2005c, in press; Man and Sang, 2004; Yousefi et al, 2005; Man and Nieburgs, 2006;). The discrepancy seen in our current and previous studies in ER NCC and their derivatives is likely to reflect differences in the differentiation status. The evolution of breast cancer is believed to be a multistep process, sequentially or progressively undergoing normal, hyperplastic, in situ, invasive, and metastatic stages (Pitot, 1993; Beckmann et al, 1997; Lakhani, 1999). The progression from one stage to the other is believed to be driven by a sequential expression of stage-related signature genes and the selection of biologically more aggressive sub-clones, recapitulating the normal development process (Pitot, 1993; Beckmann et al, 1997; Lakhani, 1999). Our previous studies in both normal and surgically operated adult rat submandibular glands (which are structurally comparable to the human breast), using a combination of immunohistochemistry and autoradiograph, had shown that newly formed cell clusters by progenitor or stem cells not only differed morphologically from, but also completely lacked the expression of several proteins that were heavily expressed in their adjacent adult counterparts (Man et al, 1995, 2001a). The morphology and protein expression in these newly formed cell clusters, however, increasingly resembled their adult counterparts with time, and became indistinguishable from their normal adult counterparts two months after the operation (Man et al, 1995, 2001a). Similar changes had been observed in the human breast tissues (Nanba et al, 2001). Our findings are consistent with a number of previous reports in human breast tissues, which have well demonstrated the dissociation between ER expression and cell proliferation, and have suggested that some of the ER negative cells may represent normal or tumor stem cells of the human breast (Clarke, 1997, 2005; Rosso, 1999; Nanba, 2001). Alternatively, the discrepancy might represent the difference in the extent of interactions with the adjacent stromal and immunoreactive cells. Previous studies have well documented that the stroma not only exerts significant influences on epithelial cell proliferation, differentiation, and apoptosis, but also impacts the metastatic behavior of epithelial tumors (Fibach et al, 1972; Nakammura et al, 1997; Silberstein, 2001). On the other hand, under certain circumstances, normal stromal cells could convert malignant tumors, even highly aggressive acute leukemia into morphologically and functionally normal, or biologically less aggressive lesions (Fibach et al, 1972; Nakammura et al, 1997; Silberstein, 2001). A recent study revealed that microdissected cells from the periphery and center of the same DCIS had a substantially different frequency and pattern in the expression of 22 genes, assessed with Atlas human Cancer 1.2 Arrays containing 1176 known genes (Zhu et al, 2003). Our previous studies in human breast, cervical, and lung tumors had consistently shown that tumor and their adjacent stromal cells shared a high degree of concurrent immunohistochemical and genetic alterations (Man et al, 2000, 2001b, 2004c; Moinfar et al, 2000, 2005), although the mechanism is unknown. Our recent studies further revealed that both human breast and prostate tumors with focally disrupted ME or basal cell layers had a significantly higher (p<0.01) frequency of immunoreactive cell infiltration than their morphologically similar counterparts with an intact ME or basal cell layer (Yousefi et al, 2005; Man et al, 2004d, 2005f, in press; Man, 2005g,h). Tumor cells overlying focal ME cell layer disruptions or adjacent to immunoreactive cells generally showed distinct morphological and immunohistochemical changes, and a vast majority of the proliferating cells were located at or near focal ME or basal cell layer disruptions (Yousefi et al, 2005; Man et al, 2004d, 2005f, in press; Man, 2005g,h).

Additionally, the discrepancy might result from the combination of multiple factors. Since the epithelium is normally devoid of both blood vessels and lymphatic ducts, and several tumor suppressors are exclusively produced by ME cells (Zou et al, 1994; Barbareschi et al, 2001; Man et al, 2002a), a focal disruption in the ME cell layer is likely to have several consequences. These consequences include: [1] a localized loss of tumor suppressors and paracrine inhibitory function; [2] a localized increasing of permeability for oxygen, nutrients, or growth factors; [3] a localized increasing of leukocyte infiltration; [4] the direct tumor-stromal cell contact, which could promote and facilitate tumor invasion through different mechanisms. The loss of tumor suppressors confers tumor cell growth advantages to escape the programmed cell death (Oliveira et al, 2005; Brummer et al, 2006; Yanochko and Eckhart, 2006). The change of oxygen to the physiologic level and the increase of growth factors could selectively favors the proliferation of progenitor or stem cells (Chakravarthy et al, 2001; Csete et al, 2001; Studer et al, 2001), or activate MAP kinases and protein kinase C that trigger the exit of cells from quiescence (Boulikas, 1994, 1995). The infiltrated leukocytes directly export growth factors to tumor cells (Freeman et al, 1995; Takahashi et al, 1996; Asano-kato, 2005; Nienartowicz et al, 2006; Qu, 2006). Direct tumor-stromal cell contact augments the expression of stromal MMP and facilitates epithelial-mesenchymal transition (Kang et al, 2004; Sato et al, 2004; Strizzi et al, 2004). Direct tumor-stromal cell contact augments the expression of stromal MMP or represses the expression of E-cadherin and other epithelial cell specific markers, which facilitates epithelial-mesenchymal transition (Kang and Massague, 2004; Sato et al, 2004; Strizzi et al, 2004; DeCraene et al, 2005; Martinez-Estrada et al, 2006). The combined effects of these factors are likely to be able to alter the gene expression pattern in the derivatives of ER NCC.

In either case, the unique genetic and immunohistochemical presentation and “budding” capability suggest that these ER NCC might represent tumor progenitor or stem cells that are not mature enough to produce ER and other molecules, or were with “resting” genes for these molecules. These clusters, however, could gain the expression of ER and other molecules after invading into the stroma, probably resulting from interactions with stromal and immunoreactive cells. Thus, it is very likely that ER NCC and their derivatives may have substantially different biochemical properties. Consequently, the derivatives of ER NCC might respond tomaxifen and other therapies, whereas ER NCC might resist the same treatment, representing “seeds” for drug resistant and recurrent tumors.

Acknowledgements

This study was supported in part by grants DAMD17-01-1-0129, DAMD17-01-1-0130, PC051308 from Congressionally Directed Medical Research Programs, and 05AA from AFIP/ARP initiative fund to Yan-gao Man et al, MD., PhD.

References

Asano-kato N, Fukagawa K, Okada N (2005) Tryptase increase proliferative activity of human conjunctive fibroblasts through protease-activated receptor. Invest Ophthalmol Vis Sci 46,4622-4626.

Barbareschi M, Pecciarini L, Cangi MG (2001) p63, a p53 homologue, is a selective nuclear marker of myoepithelial cells of the human breast. Am J Surg Pathol 25,1954-1060.

Beckmann MW, Niederacher D, Schnurch HG, Gusterson BA, Bender HG (1997) Multistep carcinogenesisof breast cancer and tumour heterogeneity. J Mol Med 75, 429-439.

Boulikas T (1994) Control of DNA replication by protein phosphorylation. Anticancer Res 14, 2465-2472.

Boulikas T (1995) Phosphorylation of transcription factors and control of the cell cycle. Crit Rev Eukaryot Gene Expr 5, 1-77.

Brummer T, Schramek D, Hayes VM, Bennett HL, Callon CE, Musgrove EA, Daly RJ (2006) Increased proliferation and altered growth factor dependence of human mammary epithelial cells overexpressing the Gab2 docking protein. J Biol Chem 281, 626-631.

Chakravarthy MV, Spangenhurg EE, Booth FW (2001) Culture in low levels of oxygen enhances in vitro proliferation potential os satellite cells from old skeletal muscles. Cell Mol Life Sci 58,1150-1158.

Clarke R, Brunner N, Katzenellenbogen BS (1989) Progression of human breast cancer cells from hormone-dependent to hormone-independent growth both in vitro and in vivo. Proc Natl Acad Sci USA86, 3649-3653.

Coussens LM, Fingleton B, Matrisian LM (2002) Matrix metalloproteinase inhibitors and cancer: trial and tribulations. Science 295, 2387-2392.

Csete M, Walikonis J, Slawny N (2001) Oxygen-mediated regulation of skeletal muscle satellite cell proliferation and adipogrnesis in culture. J Cell Physical 189, 189-196.

Damiani S, Ludvikova M, Tomasic G (1999) Myoepithelial cells and basal lamina in poorly differentiated in situ duct carcinoma of the breast. An immunocytochemical study. Virchows Arch 434, 227-34.

Decraene D, Smaers K, Maes D, Matsui M, Declercq L, Garmyn M (2005) A low UVB dose, with the potential to trigger a protective p53-dependent gene program, increases the resilience of keratinocytes against future UVB insults. J Invest Dermatol 125, 1026-31.

Duffy MJ, Maguire TM, Hill A, McDermott E, O’Higgins N (2000) Metalloproteinases: role in breast carcinogenesis, invasion and metastasis. Breast Cancer Res 2, 252-257.

Fibach E, Landau T, Sachs L (1972) Normal differentiation of myeloid leukaemic cells induced by a differentiation-inducing protein. Nat New Biol 237, 276-282.

Freeman MR, Schneck FX, Gagnon ML, Corless C, Soker S, Niknejad K, Peoples GE, Klagsbrun M (1995) Peripheral blood T-lymphocytes and lymphocytes infiltrating human cancers express vascular endothelial growth factor: a potential role for T cells in angiogenesis. Cancer Res 55, 4140-4145.

Goldfarb RH, Liotta LA (1986) Proteolytic enzymes in cancer invasion and metastasis. Semin Thromb Hemost 12, 294-307.

Guelstein VI, Tchypysheva A, Ermilova VD, Ljubimov AV (1993) Myoepithelial and basement membrane antigens in benign and malignant human breast tumors. Int J Cancer 53, 269-277.

Houlbrook S, Addison CM, Davies SL (1995) Relationship between expression of topoisomerase II isoforms and intrinsic sensitivity to topoisomerase inhibitors in breast cancer cell lines. Br J Cancer 72, 454-461.

Jourquin J, Yang N, Kamy (2006) Dispersal of epithelial cancer cell colonies by lysophosphatidic acid (LPA). J Cell Physiol 206, 337-346.

Kang Y, Massague J (2004) Epithelial-mesenchymal transition: twist in development and metastasis. Cell 118,277-279.

Kordon EC, Smith GH (1998) An entire functional mammary gland may comprise the progeny from a single cell. Development 125, 1921-1930.

Lakhani SR (1999) The transition from hyperplasia to invasive carcinoma of the breast. J Pathol 187,272-278.

Lu J, Izyolsky KI, Qian J (2005) Identification of FGF10 targets in the embryonic lung epithelium during bud morphogenesis. J Biol Chem 280, 4834-4841.

Luna-More S, Casquero S, Perey-Mellado A, Rius F, Weill B, Goremann I (2000) Importance of estrogen receptors for the behavior of invasive micropapillary carcinoma of the breast. Review of 68 cases with follow-up of 54. Pathol Res Pract 196, 35-39.

Man YG (2005g) CD8 and mast cell tryptase positive cells are preferentially associated with focal myoepithelial cell layer disruptions: implications for breast tumor invasion. Proceedings of Department of Defense Breast Cancer Research Program Meeting 2, P38-13, 263.

Man YG (2005h) Focal degenerations in surrounding structures and infiltration of immunoreactive cells are a potential trigger for invasion of breast and other epithelium-derived tumors. Proceedings of Department of Defense Breast Cancer Research Program Meeting P10-7, 75-76.

Man YG, Ball WD, Culp AJ, Hand AR, Moreira JE (1995) Persistence of a perinatal cellular phenotype in the ducts of adult glands. J Histochem Cytochem 43, 1203-1215.

Man YG, Ball WD, Marchetti L, Hand AR (2001a) Contributions of intercalated duct cells to normalparenchyma of submandibular glands of adult rats. Anat Rec 263, 202-214.

Man YG, Berg PE, Sang QXA (2005b) Differential expression of tumor invasion related proteins in cells overlying focally disrupted myoepithelial cell layers and adjacent cells within the same duct. Proceedings of Department of Defense Breast Cancer Research Program Meeting P10-5, 75.

Man YG, Fu SW, Pinzone JJ, Schwartz AM, Simmens SJ, Berg PE (2005e) Expression of BP1, a homeobox gene, correlates with progression and invasion of mammary ductal Carcinoma. Breast Cancer Res Treatment 90, 241-247.

Man YG, Magrane GG, Lininger RA, Shen T, Kuhls E, Bratthauer GL (2004b) Morphologically similarepithelial and stromal cells in primary bilateral breast tumors display different genetic profiles: implications for treatment. Appl Immunohistochem Mol Morph 12, 305-314.

Man YG, Mannion C, Albores-Saaverdra J, Bratthauer GL, Kuhls E, Tavassoli FA (2001b) Allelic lossesat 3p and 11p are detected in both epithelial and stromal components of cervical small cell neuroendocrine carcinoma. Appl Immunohistochem Mol Morph 9, 340-345.

Man YG, Martinez A, Avis IM, Hong SH, Cuttitta F, Venzon DJ, Mulshine JL (2000) Phenotypically different respiratory epithelial cells with hnRNP A2/B1 over-expression display similar genetic alterations. Am J Respir Cell Mol Biol 23, 636-645.

Man YG, Sang QXA (2004) The significance of focal myoepitehlial cell layer disruptions in breast tumor invasion: a paradigm shift from the “protease-centered” hypothesis. Exp Cell Res 301,103-118.

Man YG, Sang QXA, Zhao CQ, Mannion C, Gardner WA. Focal basal cell degeneration induced lymphocyte infiltration is a potential trigger factor for prostate tumor invasion: implications for treatmentand prevention. Cancer Detect Prev, in press.

Man YG, Shekitka KM, Bratthauer GL, Tavassoli FA (2002c) Immunohistochemical and genetic alterations in mammary epithelial cells overlying focally disrupted myoepithelial cell layers. Breast Cancer Res Treat 76 (suppl1), S143, 569.

Man YG, Shen T, Weisz J, Berg PE, Schwartz AM, Mulshine JL, Sang QXA, Nieburgs HE (2005a) Asubset of in situ breast tumor cell clusters lacks expression of proliferation and progression related markers but shows signs of stromal and vascular invasion. Cancer Detect Prev 29, 323-331.

Man YG, Shen T, Zhao YG, Sang QX (2004c) Focal prostate basal cell layer disruptions and leukocyte infiltration are correlated events: Implications for basal call layer degradation and tumor invasion. Cancer Detection & Prevention, 2004 Symposium Issue S-51, 15.

Man YG, Shen T, Zhao YG, Sang QXA (2005d) Focal prostate basal cell layer disruptions and leukocytein filtration are correlated events: A potential mechanism for basal cell layer disruptions and tumor invasion. Cancer Detect Prev 29, 161-169.

Man YG, Tai L, Barner R, Vang R, Saenger JS, Shekitka KM, Bratthauer GL, Wheeler DT, Liang CL, Vinh TN, Strauss BL (2003a) Cell clusters overlying focally disrupted mammary myoepithelial cell layers and adjacent cells within the same duct display different immunohistochemical and genetic features: implications for tumor progression and invasion. Breast Cancer Res 5, R231-241.

Man YG, Tai L, Barner Ross, Liang CY, Vang RS, Saenger JS, Shekitka KM, Bratthauer GL, Chen PY, Tavassoli FA (2002b) Focal losses of ER expression in epithelial cells and disruptions of subjacent myoepithelial cell layers are correlated events in ER (+) ductal intraepithelial neoplasia. Proceedings of Department of Defense Breast Cancer Research Program Meeting 1, P9, 17.

Man YG, Tavassoli FA (1996) A simple epitope retrieval method without the use of microwave oven orenzyme digestion. Appl Immunohistochem 4, 139-141.

Man YG, Vang RS, Saenger JS (2002a) Co-expression of maspin and wilms’ tumor 1 proteins in mammarymyoepithelial cells-implication for tumor progression and invasion. Proceedings of Department of Defense Breast Cancer Research Program Meting 1, P9, 16.

Man YG, Vinh T, Zhao C, Walker A, Barner R (2005f) Potential roles of T-lymphocytes and natural killer cells in human myoepithelial cell layer disruptions and tumor invasion. Mod Pathol 18, 42A, 179.

Man YG, Zeng X, Shen T, Vang R, Barner R, Wheeler DT, Vihn T, Liang CY, Strauss BL (2004a) Cell clusters overlying focally disrupted myoepithelial cell layers and their adjacent counterparts within the same duct display a different pattern of mRNA expression. Mod Pathol 17 (suppl1), 40- 41a.

Man YG, Zhang R, Mattu R, Shen T, Sang QXA (2003b) A subset of mammary epithelial cells overlying focally disrupted myoepithelial cell layers shows an unusual immunostaining pattern for proliferation -related proteins. Breast Cancer Res Treat 82 (supplement 1), S163-164.

Man YG, Zhang Y, Shen T, Vinh TN, Zeng X, Tauler J, Mulshine JL, Strauss BL (2005c) cDNA expression profiling identifies elevated expressions of tumor progression and invasion related genes in cell clusters of in situ breast tumors. Breast Cancer Res Treat 89, 199-208.

Man YG. Focal degeneration of aged or injured myoepithelial cells and resultant auto-immunoreactionstrigger genetic instability and breast cancer invasion. Nova Science Publishes, Inc (an invited book chapter), in press.

Martinez-Estrada OM, Culleres A, Soriano FX, Peinado H, Bolos V, Martinez FO, Reina M, Cano A, Fabre M, Vilaro S (2006) The transcription factors Slug and Snail act as repressors of Claudin-1 expression in epithelial cells. Biochem J 394, 449-57.

Matrisian LM, Sledge GW, Mohla S (2003) Exacellular proteolysis and cancer: meeting summary andfuture directions. Cancer Res 63, 6105-6109.

Middleton LP, Vlastos G, Mirza NQ, Eva S, Sahin AA (2000) Multicentric mammary carcinoma: evidence of monoclonal proliferation. Cancer 94, 1910-1916.

Miosge N (2001) The ultrastructural composition of basement membrane in vivo. Histol Histopathol 16, 1239-1248.

Moinfar F, Gogg-Kamerer M, Sommersacher A, Regitnig P, Man YG, Zatloukal K, Denk H, Tavassoli FA (2005) Endometrial stromal sarcomas frequently express epidermal growth facor receptor (EGFR, HER-1): potential basis for a new therapeutic approach. Am J Surg Pathol 29, 485-489.

Moinfar F, Man YG, Arnould L, Bratthauer GL, Ratschek M, Tavassol FA (2000) Concurrent and independent genetic alterations in the stromal and epithelial cells of mammary carcinoma: Implications for tumorigenesis. Cancer Res 60, 2562-2566.

Murad TM, von Haam E (1986) Ultrastructure of myoepithelial cells in human mammary gland tumor. Cancer 21, 1137-1149.

Nakamura T, Matsumoto K, Kiritoshi, Tano Y, Nakamura T (1997) Induction of hepatocyte growth factor in fibroblasts by tumor-derived factors affects invasive growth of tumor cells: in vitro analysis of tumor-stromal interactions. Cancer Res 57, 3305-3313.

Nanba D, Nakanishi Y, Hieda Y (2001) Changes in adhesive properties of epithelial cells during early morphogenesis of the mammary gland. Dev Growth Differ 43, 535-544.

Nerlich A (1995) Morphology of basement membrane and associated matrix proteins in normal and pathological tissues. Veroff Pathol 145, 1-139.

Nienartowicz A, Sobaniee-Lotowska ME, Jarocka-Cyrta E, Lemancewicz D (2006) Mast cells in neoangiogenesis. Med Sci Monit 12, RA53-56.

Nowell PC (1976) The clonal evolution of tumor cell populations. Science 194, 23-28.

Oliveira AM, Ross JS, Fletcher JA (2005) Tumor suppressor genes in breast cancer: the gate keepers and the caretakers. Am J Clin Pathol 124(Suppl), S16-28.

Parker SL, Tong T, Bolders S, Wingo PA (1997) Cancer statistics. Cancer J Clin 47, 5-27.

Pitot HC (1993) Multistage carcinogenesis-genetic and epigenetic mechanisms in relation to cancer prevention. Cancer Detect Prev 17, 567-573.

Qu Z (2006) Immunohistological detection of growth factors and cytokines in tissue mast cells. Methods Mol Biol 315, 257-272.

Rochefort H, Glondu M, Sahla ME, Platat N, Garcia M (2003) How to target estrogen receptor-negative breast cancer. Endocr Relat Cancer 10, 261-266.

Sato T, Sakai T, Noguchi Y, Takita M, Hirakawa S, Ito A (2004) Tumor-stromal cell contact promotes invasion of human uterine cervical carcinoma cells by augmenting the expression and activation of stromal matrix metalloproteinases. Gynecol Oncol 92, 47-56.

Schmitt FC (1995) Multistep progression from an estrogen-dependent growth towards an autonomous growth in breast carcinogenesis. Eur J Cancer 31A, 2049-2052.

Sell S, Pierce GB (1994) Maturation arrest of stem cell differentiation is a common pathway for the cellular origin of teratocarcinomas and epithelial cancers. Lab Invest 70, 6-22.

Sheikh MS, Garcia M, Pujol P, Fontana JA, Rochefort H (1994-95) Why are estrogen receptor negative breast cancers more aggressive than the estrogen receptor positive breast cancers? Invasion Metastasis 14, 329-336.

Silberstein GB (2001) Tumour-stromal interactions. Role of the stroma in mammary development. Breast Cancer Res 3, 218-223.

Strizzi L, Bianco C, Normanno N (2004) Epithelial mesenchymal transition is a characteristic of hyperplasias and tumors in mammary gland from MMTV-Criptol-1 transgenic mice. J Cell Physiol 201, 266-276.

Studer L, Csete M, Lee SH (2000) Enhanced proliferation, survival, and dopaminergic differentiation of CNS precursors in lowed pxygen. J Neurosci 20, 7377-7383.

Swain SM, Wilson JW, Mamounas EP, Bryant J, Wickerham DL, Fisher B, Paik S, Wolmark N (2004) Estrogen receptor status of primary breast cancer is predictive of estrogen receptor status of contralateral breast cancer. J Natl Cancer Inst 96, 497-499.

Takahashi Y, Bucana CD, Liu W, Yoneda J, Kitadai Y, Cleary KR, Ellis LM (1996) Platelet-derived endothelial cell growth factor in human colon cancer angiogenesis: role of infiltrating cells. J Natl Cancer Inst 88, 1146-1151.

Tavassoli FA, Man YG (1995) Morphofunctional features of intraductal hyperplasia, atypical hyperplasia, and various grades of intraductal carcinoma. Breast J 1, 155-162.

Tsubura A, Shikata N, Inui T, Morii S, Hatano T, Oikawa T (1988) Immunohistochemical localization of myoepithelial cells and basement membrane in normal, benign and malignant human breast lesions. Virchows Arch 413,133-139.

Yang M, Moriya T, DeLa Cruz C, Endoh M, Ishida T, Hirakawa H, Orita Y, Ohuchi N, Sasano H (2003) Microinvasive ductal carcinoma (T1mic) of the breast. The clinicopathological profile and immunohistochemical features of 28 cases. Pathol Int 53, 422-428.

Yanochko GM, Eckhart W (2006) Type 1 insulin-like growth factor overexpression induce proliferation and anti-apoptotic signaling in a three-dimensional culture model of breast epithelial cells. Breast CancerRes 8, R18 [Epub ahead of print].

Yousefi M, Mattu R, Gao CL, Man YG (2005) Mammary ducts with and without focal myoepithelial cell layer disruptions show a different frequency of white blood cell infiltration and growth pattern: Implications for tumor progression and invasion. Appl Immunohistochem Mol Morph 13, 30-37.

Zhao YG, Xiao AZ, Park HI, Newcomer RG, Yan M, Man YG, Heffelfinger SC, Sang QXA (2004) Endometase/matrilysin-2 in human breast ductal carcinoma in situ and its inhibition by tissue inhibitors of metalloproteinase-2 and 4: A putative role in the initiation of breast cancer invasion. Cancer Res 64, 590-598.

Zhu G, Reynolds L, Crnogorac-Jurceic T, Gillett CE, Dublin EA, Mars Barnes D, D’Arrigo C, van Trappen PO, Lemoine NR, Hart IR (2003) Combination of microdissection and microarray analysis to identify gene expression changes between differentially located tumour cells of breast cancer. Oncogene 22, 3742-3748.

Zou Z, Anisowicz A, Hendrix MJ, Thor A, Neveu M, Sheng S, Rafidi K, Seftor E, Sager R (1994) Maspin, a serpin with tumor-suppressing activity in human mammary epithelial cells. Science 263, 526-529.

Yan-gao Man

Research Article