Advertisement

Apocrine Cysts of the Breast

Biomarkers, Origin, Enlargement, and Relation with Cancer Phenotype*
  • Julio E. Celis
    Correspondence
    To whom correspondence should be addressed. Tel.: 45-35257363; Fax: 45-35257375
    Affiliations
    Danish Centre for Translational Breast Cancer Research (DCTB), Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49

    Department of Proteomics in Cancer, Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49
    Search for articles by this author
  • Pavel Gromov
    Affiliations
    Danish Centre for Translational Breast Cancer Research (DCTB), Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49

    Department of Proteomics in Cancer, Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49
    Search for articles by this author
  • José M.A. Moreira
    Affiliations
    Danish Centre for Translational Breast Cancer Research (DCTB), Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49

    Department of Proteomics in Cancer, Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49
    Search for articles by this author
  • Teresa Cabezón
    Affiliations
    Danish Centre for Translational Breast Cancer Research (DCTB), Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49

    Department of Proteomics in Cancer, Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49
    Search for articles by this author
  • Esbern Friis
    Affiliations
    Danish Centre for Translational Breast Cancer Research (DCTB), Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49

    Department of Breast and Endocrine Surgery, the Centre of Diagnostic Investigations, Rigshospitalet, DK-2100 Copenhagen, Denmark
    Search for articles by this author
  • Ilse M.M. Vejborg
    Affiliations
    Danish Centre for Translational Breast Cancer Research (DCTB), Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49

    Radiology Department, Mammography Division, the Centre of Diagnostic Investigations, Rigshospitalet, DK-2100 Copenhagen, Denmark
    Search for articles by this author
  • Gottfried Proess
    Affiliations
    Danish Centre for Translational Breast Cancer Research (DCTB), Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49

    Eurogentec, Parc Scientifique du Sart Tilman, 4102 Seraing, Belgium
    Search for articles by this author
  • Fritz Rank
    Affiliations
    Danish Centre for Translational Breast Cancer Research (DCTB), Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49

    Department of Pathology, the Centre of Diagnostic Investigations, Rigshospitalet, DK-2100 Copenhagen, Denmark
    Search for articles by this author
  • Irina Gromova
    Affiliations
    Danish Centre for Translational Breast Cancer Research (DCTB), Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49

    Department of Proteomics in Cancer, Institute of Cancer Biology, Danish Cancer Society, Strandboulevarden 49
    Search for articles by this author
  • Author Footnotes
    * This work was supported by the Danish Cancer Society through the budget of the Institute of Cancer Biology; by grants from the Danish Medical Research Council, the Natural and Medical Sciences Committee of the Danish Cancer Society, Novo Nordisk, and the John and Birthe Meyer Foundation; and by the Marketing Department at the Danish Cancer Society.
      Up to one-third of women aged 30–50 years have cysts in their breasts and are presumed to be at increased risk of developing breast cancer. Here we present an extensive proteomic and immunohistochemistry (IHC) study of breast apocrine cystic lesions aimed at generating specific biomarkers and elucidating the relationship, if existent, of apocrine cysts with cancer phenotype. To this end we compared the expression profiles of apocrine macrocysts obtained from mastectomies from high risk cancer patients with those of cancerous and non-malignant mammary tissue biopsies collected from the same patients. We identified two biomarkers, 15-hydroxyprostaglandin dehydrogenase and 3-hydroxymethylglutaryl-CoA reductase, that were expressed specifically by apocrine type I cysts as well as by apocrine metaplastic cells in type II microcysts, terminal ducts, and intraductal papillary lesions. No expression of these markers was observed in non-malignant terminal ductal lobular units, type II flat cysts, stroma cells, or fat tissue as judged by IHC analysis of matched non-malignant tissue samples collected from 93 high risk patients enrolled in our cancer program. IHC analysis of the corresponding 93 primary tumors indicated that most apocrine changes have little intrinsic malignant potential, although some may progress to invasive apocrine cancer. None of the apocrine lesions examined, however, seemed to be a precursor of invasive ductal carcinomas, which accounted for 81% of the tumors analyzed. Our studies also provided some insight into the origin, development, and enlargement of apocrine cysts in mammary tissue. The successful identification of differentially expressed proteins that characterize specific steps in the progression from early benign lesions to apocrine cancer opens a window of opportunity for designing and testing new approaches for pharmacological intervention, not only in a therapeutic setting but also for chemoprevention, to inhibit cyst development as both 15-hydroxyprostaglandin dehydrogenase and 3-hydroxymethylglutaryl-CoA reductase are currently being targeted for chemoprevention strategies in various malignancies.
      The human breast presents many benign lesions that involve both the glandular and stromal tissues. These include fibrocystic changes, benign breast tumors, and breast inflammatory disease (
      • Guinebretière J.M.
      • Menet E.
      • Tardivon A.
      • Cherel P.
      • Vanel D.
      Normal and pathological breast, the histological basis.
      ). Fibrocystic changes affect more than 50% of women during their lifetime and are comprised of cystic dilation of ducts, apocrine metaplasia of ductal epithelium, fibrosis, adenosis, and intraductal epithelial proliferation (Ref.
      • Haagensen C.D.
      and references therein).
      Cysts are fluid-filled sacs that usually develop in the upper half of the breast and are most common in women between 30 and 50 years of age (
      • Wellings S.R.
      • Alpers C.E.
      Apocrine cystic metaplasia: subgross pathology and prevalence in cancer-associated versus random autopsy breasts.
      ). They are also found in menopausal women on hormone replacement therapy. Cysts start as a microscopic dilation of the ductules (microcyst) but can enlarge and reach sizes of a couple of centimeters in diameter (macrocysts) causing pain and discomfort due to the tension effect on the surrounding stroma. In addition, cysts may rupture, eliciting chronic inflammation. Draining of the cystic fluid by fine needle aspiration often alleviates the symptoms, but although cysts may disappear following aspiration, in about a third of the cases they recur (
      • Rosai J.
      ).
      Cysts are divided into apocrine (type I, acidophil cells), flattened (type II, basophil cells), and intermediate types based on the K+/Na+ concentration ratio (
      • Dixon J.M.
      • Scott W.N.
      • Miller W.R.
      Natural history of cystic disease: the importance of cyst type.
      • Enriori C.L.
      • Novelli J.E.
      • Cremona M.D.C.
      • Hirsig R.J.P.
      • Enriori P.J.
      Biochemical study of cyst fluid in human breast cystic disease a review.
      • Miller W.R.
      • Scott W.N.
      • Harris W.H.
      • Wang D.
      Using biological measurements, can patients with benign breast disease who are at high risk for breast cancer be identified?.
      • Bruzzi P.
      • Dogliotti L.
      • Naldoni C.
      • Bucchi L.
      • Costantini M.
      • Cicognani A.
      • Torta M.
      • Buzzi G.F.
      • Angeli A.
      Cohort study of association of risk of breast cancer with cyst type in women with gross cystic disease of the breast.
      • Tsung J.S.
      • Wang T.Y.
      • Wang S.M.
      • Yang P.S.
      Cytological and biochemical studies of breast cyst fluid.
      ). Type I cysts have a higher K+/Na+ ratio (≥1.5) (
      • Dixon J.M.
      • Scott W.N.
      • Miller W.R.
      Natural history of cystic disease: the importance of cyst type.
      ), are lined by metaplastic apocrine secretory epithelium that resembles sweat glands, and exhibit tight cell-cell junctions. Apocrine changes can be complex and usually show apical snouts that are shed off to the lumen of the duct (
      • Trenkic S.
      • Katic V.
      • Pashalina M.
      • Zivkovic V.
      • Milentijevic M.
      • Kostov M.
      The histologic spectrum of apocrine lesions of the breast.
      ). In some cases, cells within type I cysts present transitions from cuboidal to flat apocrine epithelia, most likely representing differences in their stage of differentiation and/or metabolic activity. Breast lesions undergoing apocrine differentiation in most cases do not express estrogen receptor (ER)
      The abbreviations used are: ER, estrogen receptor; PR, progesterone receptor; IHC, immunohistochemistry; 2D, two-dimensional; HMG, 3-hydroxymethylglutaryl; PG, prostaglandin, 15-PGDH, 15-hydroxyprostaglandin dehydrogenase; TNF, tumor necrosis factor; IL, interleukin; apo, apolipoprotein; GCDFP, gross cystic disease fluid protein; DCTB, Danish Centre for Breast Cancer Research; COX, cyclooxygenase; AR, androgen receptor; TDLU, terminal ductal lobular units; CIS, carcinoma in situ.
      1The abbreviations used are: ER, estrogen receptor; PR, progesterone receptor; IHC, immunohistochemistry; 2D, two-dimensional; HMG, 3-hydroxymethylglutaryl; PG, prostaglandin, 15-PGDH, 15-hydroxyprostaglandin dehydrogenase; TNF, tumor necrosis factor; IL, interleukin; apo, apolipoprotein; GCDFP, gross cystic disease fluid protein; DCTB, Danish Centre for Breast Cancer Research; COX, cyclooxygenase; AR, androgen receptor; TDLU, terminal ductal lobular units; CIS, carcinoma in situ.
      -α or progesterone receptor (PR) but are often positive for the androgen receptor (
      • Tavassoli F.A.
      • Purcell C.A.
      • Man Y.G.
      Down regulation of bcl2, ER, PR associated with AR expression is triggered by apocrine differentiation of mammary epithelium.
      • Gatalica Z.
      Immunohistochemical analysis of apocrine breast lesions. Consistent over-expression of androgen receptor accompanied by the loss of estrogen and progesterone receptors in apocrine metaplasia and apocrine carcinoma in situ.
      • Selim A.G.
      • Wells C.A.
      Immunohistochemical localisation of androgen receptor in apocrine metaplasia and apocrine adenosis of the breast: relation to oestrogen and progesterone receptors.
      • Scawn R.
      • Shousha S.
      Morphologic spectrum of estrogen receptor-negative breast carcinoma.
      ). At present, little is known about the molecular mechanisms underlying apocrine differentiation of mammary lesions.
      Type II cysts on the other hand have an electrolyte composition more similar to serum (K+/Na+ ratio <1.5) and are lined by flat epithelial cells showing open cell-cell junctions (
      • Angeli A.
      • Bradlow H.L.
      • Bodian C.A.
      • Chasalow F.I.
      • Dogliotti L.
      • Haagensen Jr., D.E.
      Criteria for classifying breast cyst fluids.
      ). Contrary to the epithelia of normal breast ductules that show heterogeneous and lower levels of ER-α expression, flat type II microcysts are composed of contiguous cells that stain strongly for ER-α and PR (
      • Shoker B.S.
      • Jarvis C.
      • Sibson D.R.
      • Walker C.
      • Sloane J.P.
      Oestrogen receptor expression in the normal and pre-cancerous breast.
      ,
      • Tran D.D.
      • Lawson J.S.
      Microcysts and breast cancer: a study of biological markers in archival biopsy material.
      ). Flat type II microcysts are believed to be derived from a cystic transformation of the glandular cells that become columnar followed by dilation of the ductules due to secretion (
      • Oyama T.
      • Maluf H.
      • Koerner F.
      Atypical cystic lobules: an early stage in the formation of low-grade ductal carcinoma in situ.
      ). To date little is known about the mechanism(s) that leads to cyst development and enlargement, although there is some information indicating that their development is regulated by hormones.
      The protein, hormone, and electrolyte content of breast cyst fluid has been the subject of numerous studies that have shown that this fluid contains a variety of biologically active substances including steroids (
      • Bradlow H.L.
      • Rosenfeld R.S.
      • Kream J.
      • Fleisher M.
      • O’Connor J.
      • Schwartz M.K.
      Steroid hormone accumulation in human breast cyst fluid.
      • Vanluchene E.
      • Vandekerckhove D.
      • De Boever J.
      • Sandra P.
      Structures and concentrations of fifteen different steroid sulfates in human breast cyst fluids.
      • Belanger A.
      • Caron S.
      • Labrie F.
      • Naldoni C.
      • Dogliotti L.
      • Angeli A.
      Levels of eighteen non-conjugated and conjugated steroids in human breast cyst fluid: relationships with cyst type.
      • Nahoul K.
      • Kottler M.L.
      Relationships between androgen and estrogen sulfates in breast cyst fluid.
      • Angeli A.
      • Dogliotti L.
      • Naldoni C.
      • Orlandi F.
      • Puligheddu B.
      • Caraci P.
      • Bucchi L.
      • Torta M.
      • Bruzzi P.
      Steroid biochemistry and categorization of breast cyst fluid: relation to breast cancer risk.
      ) and cytokines and growth factors (insulin-like growth factor-binding protein-3, tumor necrosis factor (TNF)-α, transforming growth factor-α and -β, interleukin (IL)-1, and IL-6)) (
      • Tapper D.
      • Gajdusek C.
      • Moe R.
      • Ness J.
      Identification of a unique biological tumor marker in human breast cyst fluid and breast cancer tissue.
      • Reed M.J.
      • Coldham N.G.
      • Patel S.R.
      • Ghilchik M.W.
      • James V.H.
      Interleukin-1 and interleukin-6 in breast cyst fluid: their role in regulating aromatase activity in breast cancer cells.
      • Ness J.C.
      • Sedghinasab M.
      • Moe R.E.
      • Tapper D.
      Identification of multiple proliferative growth factors in breast cyst fluid.
      • Lai L.C.
      • Kadory S.
      • Siraj A.K.
      • Lennard T.W.
      Oncostatin M, interleukin 2, interleukin 6 and interleukin 8 in breast cyst fluid.
      • Lai L.C.
      • Siraj A.K.
      • Erbas H.
      • Lennard T.W.
      Transforming growth factor α, β1 and β2 in breast cyst fluid.
      ) as well as proteins such as prostate-specific antigen (
      • Tsung J.S.
      • Wang T.Y.
      • Wang S.M.
      • Yang P.S.
      Cytological and biochemical studies of breast cyst fluid.
      ,
      • Mannello F.
      • Bocchiotti G.
      • Bianchi G.
      • Marcheggiani F.
      • Gazzanelli G.
      Quantification of prostate-specific antigen immunoreactivity in human breast cyst fluids.
      • Diamandis E.P.
      • Yu H.
      • Lopez-Otin C.
      Prostate specific antigen—a new constituent of breast cyst fluid.
      • Lai L.C.
      • Erbas H.
      • Lennard T.W.
      • Peaston R.T.
      Prostate-specific antigen in breast cyst fluid: possible role of prostate-specific antigen in hormone-dependent breast cancer.
      ), proteases (kallikrein 6 and pepsinogen C) (
      • Diamandis E.P.
      • Yousef G.M.
      • Soosaipillai A.R.
      • Grass L.
      • Porter A.
      • Little S.
      • Sotiropoulou G.
      Immunofluorometric assay of human kallikrein 6 (zyme/protease M/neurosin) and preliminary clinical applications.
      ,
      • Borchert G.H.
      • Melegos D.N.
      • Yu H.
      • Giai M.
      • Roagna R.
      • Ponzone R.
      • Sgro L.
      • Diamandis E.P.
      Quantification of pepsinogen C and prostaglandin D synthase in breast cyst fluid and their potential utility for cyst type classification.
      ), c-ErbB-2 oncoprotein (ErbB-2 protein) (
      • Inaji H.
      • Koyama H.
      • Motomura K.
      • Noguchi S.
      • Tsuji N.
      • Kimura Y.
      • Sato H.
      • Sugano K.
      • Ohkura H.
      Simultaneous assay of ErbB-2 protein and carcinoembryonic antigen in cyst fluid as an aid in diagnosing cystic lesions of the breast.
      ), and carcinoembryonic antigen (
      • Inaji H.
      • Koyama H.
      • Motomura K.
      • Noguchi S.
      • Tsuji N.
      • Kimura Y.
      • Sato H.
      • Sugano K.
      • Ohkura H.
      Simultaneous assay of ErbB-2 protein and carcinoembryonic antigen in cyst fluid as an aid in diagnosing cystic lesions of the breast.
      ). Zinc-α2-glycoprotein (
      • Bundred N.J.
      • Miller W.R.
      • Walker R.A.
      An immunohistochemical study of the tissue distribution of the breast cyst fluid protein, zinc α2 glycoprotein.
      ,
      • Bundred N.J.
      • Scott W.N.
      • Davies S.J.
      • Miller W.R.
      • Mansel R.E.
      Zinc α-2 glycoprotein levels in serum and breast fluids: a potential marker of apocrine activity.
      ), apolipoprotein D (apoD) (
      • Balbin M.
      • Freije J.M.
      • Fueyo A.
      • Sanchez L.M.
      • Lopez-Otin C.
      Apolipoprotein D is the major protein component in cyst fluid from women with human breast gross cystic disease.
      ,
      • Sanchez L.M.
      • Diez-Itza I.
      • Vizoso F.
      • Lopez-Otin C.
      Cholesterol and apolipoprotein D in gross cystic disease of the breast.
      ), also known as gross cystic disease fluid protein (GCDFP)-24, and GCDFP-15 (
      • Mazoujian G.
      • Pinkus G.S.
      • Davis S.
      • Haagensen Jr., D.E.
      Immunohistochemistry of a gross cystic disease fluid protein (GCDFP-15) of the breast. A marker of apocrine epithelium and breast carcinomas with apocrine features.
      ,
      • Sapino A.
      • Caisson P.
      • Bussolati G.
      Gross cystic disease fluid protein (GCDFP-15) in the breast: past and present.
      ) are the three most abundant proteins present in breast gross cystic disease fluid. Whether breast cysts are active endocrine glands or inactive reservoirs of hormones, growth factors, and other compounds is at present unknown.
      For some time, microcysts were considered as a normal stage in the development and involution of the breast (
      • Shoker B.S.
      • Jarvis C.
      • Sibson D.R.
      • Walker C.
      • Sloane J.P.
      Oestrogen receptor expression in the normal and pre-cancerous breast.
      ,
      • Oyama T.
      • Maluf H.
      • Koerner F.
      Atypical cystic lobules: an early stage in the formation of low-grade ductal carcinoma in situ.
      ,
      • Hayward J.L.
      • Parks A.G.
      Alterations in the microanatomy of the breast as a result of changes in the hormonal environment.
      ), but lately several publications have associated them with the transition of mammary epithelium from a benign to malignant phenotype (
      • Shoker B.S.
      • Jarvis C.
      • Sibson D.R.
      • Walker C.
      • Sloane J.P.
      Oestrogen receptor expression in the normal and pre-cancerous breast.
      ,
      • Tran D.D.
      • Lawson J.S.
      Microcysts and breast cancer: a study of biological markers in archival biopsy material.
      ,
      • Tsuchiya S.
      Atypical ductal hyperplasia, atypical lobular hyperplasia and interpretation of a new borderline lesion.
      ,
      • Kusama R.
      • Fujimori M.
      • Matsuyama I.
      • Fu L.
      • Ishi K.
      • Hama Y.
      • Asanuma K.
      • Shingu K.
      • Kobayashi S.
      • Tsuchiya S.I.
      Clinicopathological characteristics of atypical cystic duct (ACD) of the breast: assessment of ACD as a precancerous lesion.
      ). In addition, clinical follow-up studies have indicated that the presence of cysts increases the risk for subsequent development of breast cancer (
      • Bruzzi P.
      • Dogliotti L.
      • Naldoni C.
      • Bucchi L.
      • Costantini M.
      • Cicognani A.
      • Torta M.
      • Buzzi G.F.
      • Angeli A.
      Cohort study of association of risk of breast cancer with cyst type in women with gross cystic disease of the breast.
      ,
      • Haagensen C.D.
      • Bodian C.
      • Haagensen Jr., D.E.
      • Bundred N.J.
      • West R.R.
      • Dowd J.O.
      • Mansel R.E.
      • Hughes L.E.
      Is there an increased risk of breast cancer in women who have had a breast cyst aspirated?.
      • Devitt J.E.
      • To T.
      • Miller A.B.
      Risk of breast cancer in women with breast cysts.
      • Dixon J.M.
      • McDonald C.
      • Elton R.A.
      • Miller W.R.
      Risk of breast cancer in women with palpable breast cysts: a prospective study. Edinburgh Breast Group.
      ). In particular, women with type I apocrine cysts have been reported to have a 2–4 times higher risk for breast cancer development as compared with women with type II cysts (
      • Angeli A.
      • Dogliotti L.
      • Naldoni C.
      • Orlandi F.
      • Puligheddu B.
      • Caraci P.
      • Bucchi L.
      • Torta M.
      • Bruzzi P.
      Steroid biochemistry and categorization of breast cyst fluid: relation to breast cancer risk.
      ,
      • Dupont W.D.
      • Parl F.F.
      • Hartmann W.H.
      • Brinton L.A.
      • Winfield A.C.
      • Worrell J.A.
      • Schuyler P.A.
      • Plummer W.D.
      Breast cancer risk associated with proliferative breast disease and atypical hyperplasia.
      ). The latter observations have been contested by Dixon et al. (
      • Dixon J.M.
      • McDonald C.
      • Elton R.A.
      • Miller W.R.
      Risk of breast cancer in women with palpable breast cysts: a prospective study. Edinburgh Breast Group.
      ), who reported that all women with breast cysts are at increased risk of breast cancer irrespective of the cyst type. Clearly the results are ambiguous at this point as different authors have described apocrine cysts as either representing a risk factor, a non-obligate precursor lesion of apocrine carcinoma, or benign lesions with no correlation with malignancy (
      • Wellings S.R.
      • Alpers C.E.
      Apocrine cystic metaplasia: subgross pathology and prevalence in cancer-associated versus random autopsy breasts.
      ,
      • Ahmed A.
      Calcification in human breast carcinomas: ultrastructural observations.
      • Yates A.J.
      • Ahmed A.
      Apocrine carcinoma and apocrine metaplasia.
      • Page D.L.
      • Dupont W.D.
      • Jensen R.A.
      Papillary apocrine change of the breast: associations with atypical hyperplasia and risk of breast cancer.
      • Viacava P.
      • Naccarato A.G.
      • Bevilacqua G.
      Apocrine epithelium of the breast: does it result from metaplasia?.
      • Jones C.
      • Damiani S.
      • Wells D.
      • Chaggar R.
      • Lakhani S.R.
      • Eusebi V.
      Molecular cytogenetic comparison of apocrine hyperplasia and apocrine carcinoma of the breast.
      • Durham J.R.
      • Fechner R.E.
      The histologic spectrum of apocrine lesions of the breast.
      ).
      We are part of a large multidisciplinary long term research initiative, the Danish Centre for Breast Cancer Research (DCTB), aimed at the comprehensive analysis of breast cancer lesions using multiple experimental paradigms from genomics, proteomics, and transcriptomics. Our strategy is based on the study of mammary tumors and matched benign fresh tissues collected from mastectomies of high risk cancer patients along with the integration of the “omic” and clinical datasets (
      • Celis J.E.
      • Gromov P.
      • Gromova I.
      • Moreira J. M
      • Cabezon T.
      • Ambartsumian N.
      • Grigorian M.
      • Lukanidin E.
      • Thor Straten P.
      • Guldberg P.
      • Bartkova J.
      • Bartek J.
      • Lukas J.
      • Lukas C.
      • Lykkesfeldt A.
      • Jaattela M.
      • Roepstorff P.
      • Bolund L.
      • Orntoft T.
      • Brunner N.
      • Overgaard J.
      • Sandelin K.
      • Blichert-Toft M.
      • Mouridsen H.
      • Rank F.E.
      Integrating proteomic and functional genomic technologies in discovery-driven translational breast cancer research.
      • Celis J.E.
      • Moreira J.M.
      • Cabezon T.
      • Gromov P.
      • Friis E.
      • Rank F.
      • Gromova I.
      Identification of extracellular and intracellular signaling components of the mammary adipose tissue and its interstitial fluid in high risk breast cancer patients: toward dissecting the molecular circuitry of epithelial-adipocyte stromal cell interactions.
      • Celis J.E.
      • Gromov P.
      • Cabezon T.
      • Moreira J.M.
      • Ambartsumian N.
      • Sandelin K.
      • Rank F.
      • Gromova I.
      Proteomic characterization of the interstitial fluid perfusing the breast tumor microenvironment: a novel resource for biomarker and therapeutic target discovery.
      • Celis J.E.
      • Moreira J.M.
      • Gromova I.
      • Cabezon T.
      • Ralfkiaer U.
      • Guldberg P.
      • Straten P.T.
      • Mouridsen H.
      • Friis E.
      • Holm D.
      • Rank F.
      • Gromov P.
      Towards discovery-driven translational research in breast cancer.
      ). Gross pathological inspection of the mastectomy specimens often revealed the presence of macrocysts of various sizes (Fig. 1) that were localized at varying distances to the tumor. Considering that apocrine metaplasia occurs in a spectrum of breast epithelial lesions (
      • Gatalica Z.
      Immunohistochemical analysis of apocrine breast lesions. Consistent over-expression of androgen receptor accompanied by the loss of estrogen and progesterone receptors in apocrine metaplasia and apocrine carcinoma in situ.
      ,
      • O’Malley F.P.
      • Bane A.L.
      The spectrum of apocrine lesions of the breast.
      ) and taking advantage of the unique possibility of having access to matched tissue samples collected from the same patient as well as our experience in biomarker discovery (
      • Celis J.E.
      • Bravo R.
      • Larsen P.M.
      • Fey S.J.
      Cyclin (PCNA): a nuclear protein whose level correlates directly with the proliferative state of normal as well as transformed cells.
      • Madsen P.
      • Rasmussen H.H.
      • Leffers H.
      • Honore B.
      • Dejgaard K.
      • Olsen E.
      • Kiil J.
      • Walbum E.
      • Andersen A.H.
      • Basse B.
      • Celis J.E.
      Molecular cloning, occurrence, and expression of a novel partially secreted protein “psoriasin” that is highly up-regulated in psoriatic skin.
      • Leffers H.
      • Madsen P.
      • Rasmussen H.H.
      • Honore B.
      • Andersen A.H.
      • Walbum E.
      • Vandekerckhove J.
      • Celis J.E.
      Molecular cloning and expression of the transformation sensitive epithelial marker stratifin. A member of a protein family that has been involved in the protein kinase C signalling pathway.
      • Celis J.E.
      • Ostergaard M.
      • Basse B.
      • Celis A.
      • Lauridsen J.B.
      • Ratz G.P.
      • Andersen I.
      • Hein B.
      • Wolf H.
      • Orntoft T.F.
      • Rasmussen H.H.
      Loss of adipocyte-type fatty acid binding protein and other protein biomarkers is associated with progression of human bladder transitional cell carcinomas.
      ), we carried out a comprehensive proteomic study of apocrine cysts with the aim of generating specific biomarkers for breast apocrine metaplasia that could provide some insight into their origin and mechanisms underlying their development and enlargement as well as to enlighten their relationship, if existent, with cancer phenotype.
      Figure thumbnail gr1
      Fig. 1Overview of the various steps involved in the analysis of cyst tissue and fluid. A similar procedure was used to analyze non-malignant and tumor specimens. As seen from the keratin 19 staining of macrocyst 81, the cyst is composed of several microcysts.

      EXPERIMENTAL PROCEDURES

       Patient Selection

      Thirty-six women with primary, operable, high risk
      The criteria for high risk cancer applied by the Danish Breast Cancer Cooperative Group (DBCG) are age below 35 years old, and/or tumor diameter of more than 20 mm, and/or histological malignancy grade 2 or 3, and/or negative estrogen and progesterone receptor status, and/or positive axillary status.
      invasive breast cancer were selected for this prospective study (DCTB patient numbers 59–94 in Table I). Patients had no previous surgery to the breast and did not receive preoperative treatment. They underwent mastectomy, including axillary dissection. Data concerning age of the patient, size of the tumor, histological malignancy grade, axillary nodal status, and ER and PR status are given in Table I. The Her-2neu status was also recorded but is not included in Table I. Paraffin-embedded tissue blocks from additional 57 high risk invasive breast cancer patients enrolled in the DCTB program were used for validation purposes. These included 50 invasive ductal carcinomas, six lobular carcinomas, and one mucinous lesion. Archival paraffin-embedded tissue samples from six patients diagnosed with pure apocrine carcinoma were obtained from the Rigshospitalet tissue bank. The project was approved by the Scientific and Ethical Committee of the Copenhagen and Frederiksberg Municipalities (KF 01-069/03).
      Table IClinicopathological characteristics of high risk breast cancer patients
      DCTB patient numberAgeTumor type
      D, ductal; L, lobular.
      Tumor sizeGrade
      The histological malignancy grade was determined according to Elston and Ellis (101).
      Axillary nodal metastasisHormone receptor status
      ER and PR status was determined by IHC analysis according to DBCG guidelines. Tumors were regarded as negative when both receptors were expressed in less than 10% of tumor cell nuclei.
      Presence of macrocysts
      As determined by visual inspection of the mastectomies.
      yrmm
      5938D251NegativeER-α+ PR+
      6072D302PositiveER-α+ PR−
      6199D402PositiveER-α+ PR+
      6285L252PositiveER-α+ PR+
      6392D162NegativeER-α+ PR+
      6452Tubular231PositiveER-α+ PR+
      6581L501PositiveER-α+ PR−Yes
      6660L702NegativeER-α+ PR+
      6778L222NegativeER-α+ PR+
      6851L352PositiveER-α+ PR+Yes
      6986D252PositiveER-α+ PR+
      7083D333PositiveER-α− PR−
      7140D503NegativeER-α− PR−
      7252D253PositiveER-α+ PR+
      7375L501PositiveER-α+ PR−
      7449D213PositiveER-α− PR−Yes
      7562D203PositiveER-α− PR−
      7676D302NegativeER-α+ PR−
      7762L302PositiveER-α+ PR−
      7831D323PositiveER-α− PR−
      7956Apocrine351NegativeER-α− PR−Yes
      8041D403PositiveER-α+ PR+Yes
      8139L502PositiveER-α+ PR+Yes
      8272D252PositiveER-α+ PR+
      8357L452PositiveER-α+ PR+
      8438D182PositiveER-α+ PR+
      8580D302PositiveER-α+ PR−
      8677D1102PositiveER-α+ PR+Yes
      8766D253NegativeER-α− PR−
      8883D352PositiveER-α+ PR+
      8943D163PositiveER-α+ PR−
      9049D403PositiveER-α− PR−Yes
      9159D302PositiveER-α+ PR−
      9249D273NegativeER-α+ PR−
      9360D352PositiveER-α+ PR+
      9457D213PositiveER-α− PR−Yes
      a D, ductal; L, lobular.
      b The histological malignancy grade was determined according to Elston and Ellis (
      • Elston C.W.
      • Ellis I.O.
      Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up.
      ).
      c ER and PR status was determined by IHC analysis according to DBCG guidelines. Tumors were regarded as negative when both receptors were expressed in less than 10% of tumor cell nuclei.
      d As determined by visual inspection of the mastectomies.

       Sample Collection and Handling

      Tissue specimens were collected from the Pathology Department at Rigshospitalet within 20–30 min after surgery and were rapidly transported on ice (average transport time was 15–20 min) to the Institute of Cancer Biology for further processing. Nine of the patients (patient numbers 65, 68, 74, 79, 80, 81, 86, 90, and 94; Table I) presented macrocysts ranging in size from a few millimeters to about 0.3 cm (Fig. 1). All nine patients contained more than one macrocyst. Non-malignant and tumor tissue was collected from all 36 patients, and a fraction of each sample was stored at −80 °C for gel analysis. A similar tissue block was paraffin-embedded for immunohistochemistry (IHC) analysis.

       Preparation of Tissue and Cyst Fluid Samples for Two-dimensional Gel Electrophoresis

      Cysts were cleaned of surrounding tissue, and the cyst fluid was aspirated using an elongated Pasteur pipette (Fig. 1). Because the volume collected was very small (5–50 μl), the samples were not centrifuged. Empty macrocysts were washed with PBS to minimize traces of cyst fluid and were stored at −80 °C. Twenty to 30, 6-μm cryostat sections of frozen macrocyst tissue (Fig. 1) were resuspended in 0.1 ml of CBL-1 lysis solution (Zeptosens AG, Witterswil, Switzerland) and were kept at −20 °C until used (
      • Celis J.E.
      • Moreira J.M.
      • Cabezon T.
      • Gromov P.
      • Friis E.
      • Rank F.
      • Gromova I.
      Identification of extracellular and intracellular signaling components of the mammary adipose tissue and its interstitial fluid in high risk breast cancer patients: toward dissecting the molecular circuitry of epithelial-adipocyte stromal cell interactions.
      ). The first and last sections of each sample were used for immunofluorescence analysis using keratin 19 antibodies as this epithelial marker is ubiquitously expressed by mammary epithelial cells (Fig. 1) (
      • Moll R.
      Cytokeratins as markers of differentiation in the diagnosis of epithelial tumors.
      ). The availability of these pictures greatly facilitated the interpretation of the two-dimensional (2D) PAGE studies as it gave a rough estimate of the ratio of glands/cyst/tumor to stromal tissue. The same procedure was applied to both non-malignant and tumor tissue. 2D gels were analyzed using the PDQUEST software from Bio-Rad.
      Cyst fluid samples were freeze-dried and resuspended in lysis solution prior to electrophoresis. A few milliliters of breast gross cystic disease fluid were also collected from four patients bearing palpable cysts but deemed free of breast cancer. In the latter set of cases, the samples were centrifuged at 5000 rpm for 20 min at 4 °C, and the supernatant was kept at −80 °C until used.

       Two-dimensional Gel Electrophoresis

      Tissue samples and freeze-dried fluids resuspended in lysis solution were subjected to IEF 2D PAGE as described previously (
      • Celis J.E.
      • Trentem⊘lle S.
      • Gromov P.
      Gel-based proteomics: high-resolution two-dimensional gel electrophoresis of proteins. Isoelectric focusing (IEF) and nonequilibrium pH gradient electrophoresis (NEPHGE).
      ,
      • O’Farrell P.H.
      High resolution two-dimensional electrophoresis of proteins.
      ). Between 5 and 40 μl of sample were applied to the first dimension, and at least three IEF gels were run for each sample. Proteins were visualized using a silver staining procedure compatible with mass spectrometry analysis (
      • Gromova I.
      • Celis J. E
      Protein detection in gels by silver staining: a procedure compatible with mass spectrometry.
      ).

       Protein Identification by Mass Spectrometry

       In-gel Digestion Protocol—

      Protein spots were excised from the dry gels followed by rehydration of gel plugs in water for 30 min at room temperature. The gel pieces were detached from the cellophane film, rinsed twice with water, and cut into about 1-mm2 pieces with subsequent additional washes. Proteins were “in-gel” digested with bovine trypsin (unmodified, sequencing grade; Roche Diagnostics) for 8 h as described by Shevchenko et al. (
      • Shevchenko A.
      • Wilm M.
      • Vorm O.
      • Mann M.
      Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels.
      ). The reaction was stopped by adding TFA (up to 0.4%) followed by shaking for 20 min at room temperature to increase peptide recovery. In most cases, peptides were analyzed using the supernatant. In the few cases where the amount of peptides was too low or when no conclusive identification was achieved by peptide fingerprinting, the remaining amount of supernatant (approximately 10 μl) as well as the peptides additionally extracted from the gel pieces with 1% TFA and 50% ACN were concentrated on micro-ZipTipμ-C18 columns in accordance with the manufacturer’s instructions (Millipore).

       Probe Preparation and Acquisition of the MALDI-TOF Spectra—

      Samples were prepared for analysis by applying 2 μl of digested supernatant or microcolumn-eluted material on the surface of a 400/384 AnchorChip target (Bruker Daltonics, GmbH) followed by co-crystallization with 0.3 μl of α-cyano matrix. After drying, the droplets were washed twice with 0.5% TFA to remove contamination from the samples.
      Mass spectrometry was performed using a Reflex IV MALDI-TOF mass spectrometer equipped with a Scout 384 ion source. All spectra were obtained in positive reflector mode with delayed extraction using an accelerating voltage of 28 kV. Each spectrum represented an average of 100–200 laser shots, depending on the signal-to-noise ratio. The resulting mass spectra were internally calibrated using the autodigested tryptic mass values (805.417/906.505/1153.574/1433.721/2163.057/2273.160) visible in all spectra. Calibrated spectra were processed by the Xmass 5.1.1 and BioTools 2.1 software packages (Bruker Daltonics, GmbH). All spectra were analyzed manually as described previously (
      • Celis J.E.
      • Gromov P.
      • Cabezon T.
      • Moreira J.M.
      • Ambartsumian N.
      • Sandelin K.
      • Rank F.
      • Gromova I.
      Proteomic characterization of the interstitial fluid perfusing the breast tumor microenvironment: a novel resource for biomarker and therapeutic target discovery.
      ). No restriction on the protein molecular mass and taxonomy has been applied. A number of fixed modifications (acrylamide-modified cysteine, i.e. propionamide/carbamidomethylation) as well as variable ones (methionine oxidation and protein N terminus acetylation) were included in the search parameters. The peptide tolerance did not exceed 50 ppm, and as a maximum only one trypsin missed cleavage was allowed. The protein identifications were considered to be confident when the protein score of the hit exceeded the threshold significance score of 70 (p < 0.05) and no fewer than seven peptides were recognized. The database was checked for redundancy, and whenever possible Swiss-Prot accession numbers were assigned. The second search was performed for all identifications as follows. 1) The predicted peptide digest was compared with the experimental one to reveal additional peptides present within the spectra. 2) The unmatched molecular weight values from the initial search were applied for extra search with the same reproducibility requirements for identification of the second and the third component in the spot. Whenever the protein score hit was close to the threshold significance score of 70, the PSD was performed as an additional means to confirm the identity of the proteins identified by post-translational modification. The following PSD search parameters were used: peptide tolerance of 50 ppm and MS/MS tolerance of 1 Da without any restriction on the protein molecular mass and taxonomy. Because the amount of peptides extracted from the silver-stained gels did not yield overall peak intensities high enough to allow multiple peptide sequencing (prerequisite for conclusive PSD analysis), the identification of proteins was never made solely based on PSD analysis. The molecular weight and pI of the identified proteins were evaluated by analysis of mobility of the corresponding protein band in the 2D gel images. Positive protein identification was achieved in 80% of the cases with average sequence coverage in the range of 8–80% depending on the protein size and spot intensity.

       Antibody Arrays

      Detection of multiple cytokines present in the cyst fluid of one breast cancer patient and breast gross cystic disease fluid from four patients, free of cancer, attending the mammography clinic was done using array-based technology. For this purpose RayBio™ Cytokine Antibody Arrays 6.1 and 7.1 were purchased from RayBiotech, Inc. Arrays were incubated with either 25 μl (breast cancer patient) or 250 μl (mammography clinic patients) of cyst fluid at 4 °C overnight, and bound cytokines were detected according to the manufacturer’s instructions. The sensitivity of the cytokine antibody array ranges from 1–2000 pg/ml.

       Antibodies

      Anti-peptide antibodies against cathepsin D and apoJ were prepared by Eurogentec (Seraing, Belgium). Polyclonal rabbit antibodies against human 15-hydroxyprostaglandin dehydrogenase (15-PGDH) recovered from 2D gels were prepared as described previously (
      • Huet C.
      Production of polyclonal antibodies in rabbits.
      ). In addition, a commercially available rabbit polyclonal antibody against 15-PGDH (Cayman Chemicals, Ann Arbor, MI) was used. Monoclonal antibodies specific for psoriasin were described previously (
      • Ostergaard M.
      • Wolf H.
      • Orntoft T.F.
      • Celis J.E.
      Psoriasin (S100A7): a putative urinary marker for the follow-up of patients with bladder squamous cell carcinomas.
      ). Antibodies against 3-hydroxymethylglutaryl (HMG)-CoA reductase (CRL 18811) isolated from the A9 hybridoma cell line (ATCC, Manassas, VA) and the nuclear antigen RN3 were kindly provided by Richard J. Cenedella (Kirksville, MO) and Frans Ramaekers (Nijmegen, The Netherlands), respectively. Antibodies against cyclooxygenase 2 (COX-2), Ki67, ER-α, PR, ErbB-2, androgen receptor (AR), and p53 were purchased from Dako Cytomation (Glostrup, Denmark). Antibodies against keratin 19 were from NeoMarker (Fremont, CA). The antibodies against apoD, GCDFP-15, and active caspase-3 were from Zymed Laboratories Inc. (San Francisco, CA), Oncogene Research Products (San Diego, CA), and Promega (Madison, WI), respectively. The specificity of the 15-PGDH, cathepsin D, apoJ, psoriasin, and keratin 19 antibodies was confirmed by 2D gel immunoblotting (
      • Gromov P.
      • Celis J. E
      High-resolution two-dimensional gel electrophoresis and protein identification using western blotting and ECL detection.
      ).

       Immunohistochemistry

      Following surgery, fresh tumor blocks were immediately placed in formalin fixative and paraffin-embedded for archival use. Four-micrometer sections were cut from the tissue blocks and mounted on Super Frost Plus slides (Menzel-Gläser, Braunschweig, Germany), baked at 60 °C for 60 min, deparaffinized, and rehydrated through graded alcohol rinses. Nonspecific staining of slides was blocked (10% fetal calf serum in PBS buffer) for 15 min, and endogenous peroxidase activity was quenched using 1% H2O2 in 99% ethanol for 30 min. Heat-induced antigen retrieval was performed by immersing slides in Tris-EDTA buffer (pH 9.0) and microwaving in a 750-watt microwave oven for 10 min. The slides were then cooled at room temperature for 20 min and rinsed abundantly in tap water. Antigen was detected with a relevant primary antibody followed by a species-matched suitable secondary antibody conjugated to a peroxidase complex (EnVision+, Dako Cytomation). Finally color development was done with enhanced 3,3′-diaminobenzidine (DAB+, Dako Cytomation) as a chromogen to detect bound antibody complex. Slides were counterstained with hematoxylin. Standardization of the dilution, incubation, and development times appropriate for each antibody allowed an accurate comparison of expression levels in all cases. Sections were imaged using either a standard bright field microscope (Leica DMRB) equipped with a high resolution digital camera (Leica DC500) or a motorized digital microscope (Leica DM6000B) controlled by Objective Imaging’s Surveyor software (Objective Imaging Ltd.) for automated scanning and imaging that enables tiled mosaic image acquisition (see Fig. 11). Original magnification for all images is 200×.

      RESULTS

       Characterization of the Proteome Profile of Apocrine Macrocysts—

      Macrocysts, devoid of cyst fluid, were frozen in liquid nitrogen and sectioned serially into 30 6-μm-thick slices as described under “Experimental Procedures” and illustrated in Fig. 1. The first and last of the sections were kept for immunofluorescence analysis using keratin 19 antibodies to assess the type of cyst as well as the ratio of epithelial cells to stroma in the specimen. The remaining slices were dissolved in lysis solution for 2D PAGE analysis (
      • Celis J.E.
      • Moreira J.M.
      • Cabezon T.
      • Gromov P.
      • Friis E.
      • Rank F.
      • Gromova I.
      Identification of extracellular and intracellular signaling components of the mammary adipose tissue and its interstitial fluid in high risk breast cancer patients: toward dissecting the molecular circuitry of epithelial-adipocyte stromal cell interactions.
      ,
      • Celis J.E.
      • Gromov P.
      • Cabezon T.
      • Moreira J.M.
      • Ambartsumian N.
      • Sandelin K.
      • Rank F.
      • Gromova I.
      Proteomic characterization of the interstitial fluid perfusing the breast tumor microenvironment: a novel resource for biomarker and therapeutic target discovery.
      ).
      All macrocysts analyzed were of apocrine nature (type I), but only five specimens (two from patient 81, one from patient 86, and two from patient 94) exhibited abundant apocrine epithelium relative to stroma, yielding distinct protein patterns as judged by 2D PAGE analysis. The remaining cyst specimens contained too much connective tissue, and consequently their protein expression patterns were not informative. The expression profiles of the cysts from patients 81 and 94 were nearly identical as determined by PDQUEST analysis of their 2D gel patterns, whereas that of patient 86 exhibited a slightly different ratio of some proteins due to relatively higher levels of surrounding stroma. Fig. 2 shows representative silver-stained 2D gel of acidic (IEF) proteins of apocrine macrocysts from patients 81 (Fig. 2A) and 94 (Fig. 2B), both of which were located distant to the tumor in the mastectomized breast. Because most of the cells in apocrine macrocyst 81 corresponded to apocrine epithelium (see immunofluorescence picture in Fig. 1) and given the abundance of available tissue, which allowed us to perform both 2D PAGE and IHC analysis, we selected its protein expression profile as a reference map to single out markers of breast apocrine epithelia as well as for creating a database of apocrine cyst proteins identified by mass spectrometry (Table II). A total of 75 primary translation products, several of which were highly expressed in the apocrine cysts as compared with the corresponding non-malignant and tumor samples, are listed in alphabetical order in Table II together with the accession number, apparent molecular weight and pI, and gene map locus as well as molecular functions and biological processes according to the Human Protein Resource Database (www.hprd.org). Fifty-one of the identified proteins are known to be regulated by hormones, particularly androgens (36 proteins), and eight of these are known to interact directly with AR (Table II). Interestingly more than 50% of the identified proteins play a role in energy metabolism, particularly lipid biosynthesis, and about 45% are secreted or participate in transport processes. An annotated 2D gel image of all the proteins identified will be made available through our website (proteomics.cancer.dk).
      Figure thumbnail gr2
      Fig. 2IEF 2D PAGE of whole protein extracts from breast apocrine macrocysts.A, apocrine macrocyst 81. B, apocrine macrocyst 94. Arrows indicate some proteins that are preferentially expressed by the apocrine cysts as compared with non-malignant and tumor tissues (see also ). The position of α-enolase is indicated in red as a reference as this protein and its modified forms migrate very close to HMG-CoA synthase.
      Table IIProteins identified in apocrine macrocysts
      Protein nameAccession number
      Whenever possible, the Swiss-Prot accession number was used.
      Mr/pI
      The Mr and pI were calculated directly from the sequences. In some cases these values do not fit with the apparent molecular weights observed in the 2D gels shown in Fig. 2A.
      Hormone-regulated proteins
      Proteins whose expression is known to be regulated by hormones are indicated.
      Gene map locusBiological process
      Molecular function as well as biological process was assigned in accordance with the Human Protein Reference Database (www.hprd.org).
      ScorePeptides and PSDCoverage
      %
      15-Hydroxyprostaglandin dehydrogenase (NAD+)P1542829/5.6Yes, androgen4q34-q35Metabolism: energy pathway1541447
      Acylamino-acid-releasing enzyme, oxidized protein hydrolaseP1379882.3/5.33p21Metabolism: energy pathway1851522
      Acyl-CoA dehydrogenase, medium chain-specificP1131047/8.6Yes1p31Metabolism: energy pathway908 (PSD: 1598.79)16
      Acyl-CoA-binding protein (Diazepam binding inhibitor)P0710810/6.1Yes, androgen2q12-q21Metabolism: energy pathway1147 (PSD: 1535.72)44
      Alcohol dehydrogenase (NADP+)P1455037/6.3Yes1p33-p32Metabolism: energy pathway1401137
      Aldo-keto reductase family 1 member C2P5289537/7.1Yes10p15-p14Metabolism: energy pathway1008 (PSD: 1130.59)22
      Aldose reductase-like 1gi‖5149297439.7/6.07q33Metabolism: Energy pathway2601757
      α-EnolaseP0673347/7.0Yes, androgen1pter-p36.13Metabolism: energy pathway1141027
      α1-AntitrypsinP0100946.9/5.3Yes, androgen14q32.1Protein metabolism1851437
      α1-B-glycoproteinP0421754/5.519q13.4Biological process unknown1107 (PSD: 1372.69)20
      α-Actinin 4O43707105/5.219q13Cell growth and/or maintenance: cytoskeletal protein3812124
      α-Tubulin 1 chainP0520950/4.9Yes, androgen2q36.2Cell growth and/or maintenance: cytoskeletal protein96918
      Aminoacylase-1Q0315446/5.73p21Metabolism: energy pathway1351024
      Antithrombin-IIIP0100853/6.3Yes, androgen1q23-q25Protein metabolism: protease inhibitor1511428
      Apolipoprotein A-IVP0672745/5.2Yes11q23Transport/cargo protein1521022
      Apolipoprotein DP0509021.6/5.06Yes, androgen3q26.2-qterTransport/cargo protein1007 (PSD: 1671.77)30
      Apolipoprotein J (clusterin, complement-associated protein SP-40)P1090953/6.0Yes, androgen8p21-p12Immune response90921
      ATP-dependent DNA helicase II, Ku autoantigen protein p86P1301083.2/5.52q35DNA repair protein1201916
      β-TubulinP0743750/4.7, C-terminal endYes, androgen6p25Cell growth and/or maintenance: cytoskeletal protein807 (PSD: 1042.56)8
      β-GlucuronidaseP0823675/6.5, without C-terminalYes, androgen7q21.11Metabolism: energy pathway1471813
      Bile salt-activated lipase (lysophospholipase)P1983577.7/5.1Yes9q34.3Metabolism: energy pathway1641626
      Calreticulin precursorP2779748/4.3Yes, androgen19p13.2Protein metabolism: chaperone
      CalumeninO4385237.2/4.57q32Signal transduction: cell communication1631235
      Carbonic anhydrase IP0091528.7/6.6Yes, androgen8q22Metabolism: energy pathway1301051
      Cathepsin D (heavy chain only)P0733945/6.1Yes11p15.5Protein metabolism: protease1701324
      Chaperonin-containing TCP 1, subunit ϵ (CCT-ϵ)P4864360/5.45p15.2Protein metabolism: chaperone1891827
      Chaperonin-containing TCP 1, subunit θ (CCT-θ)P5099060/5.421q22.11Protein metabolism: chaperone1791426
      Complement C4 precursor, Complement C4 γ chainP01028192/6.6, C-terminal end6p21.3Immune response1401960
      Cyclophilin AP6293718/7.8Yes, androgen7p13Protein metabolism: chaperone1401050
      Cytokeratin 18P0578348/5.3Yes, androgen12q13Cell growth and/or maintenance: cytoskeletal protein2061834
      Cytokeratin 19P0872744/5.0417q21-q22Cell growth and/or maintenance: cytoskeletal protein3162654
      Cytokeratin 7P0872951/5.512q12-q14Cell growth and/or maintenance: cytoskeletal protein2862452
      Cytokeratin 8P0578753/5.5Yes, androgen12q13Cell growth and/or maintenance: cytoskeletal protein2332540
      Cytosolic nonspecific dipeptidase (metallopeptidase M20 family)Q96KP453/5.618q22.3Protein metabolism: protease1401420
      Elongation factor 2 (EF-2)P1363996/6.4Yes, androgen19pter-q12Protein metabolism: translation regulatory protein1311211
      %
      Endoplasmin precursor (94-kDa glucose-regulated protein, tumor rejection antigen gp96)P1462592/4.7Yes12q24.2-q24.3Signal transduction: cell communication1501520
      Fetuin-A (α2-HS-glycoprotein)P0276540.3/5.4Yes3q27-q29Cell growth and/or maintenance1651531
      Filamin 1P21333280/5.7, C-terminal endYes, androgenXq28Cell growth and/or maintenance: cytoskeletal protein17018 (PSD: 1533.75)27
      Gelsolin isoform bgi‖3804428886/5.6Yes, androgen9q33Metabolism: energy pathway2832635
      GCDFP-15 (prolactin-inducible protein)P1227316.9/8.2Yes, androgen7q34Protein metabolism1209 (PSD: 1384.57, 1254.69)58
      Glucose-6-phosphate 1-dehydrogenaseP1141359.6/6.6YesXq28Metabolism: energy pathway1171235
      Glutathione S-transferase Mu 1P0948825/6.3Yes, androgen1p13.3Metabolism: energy pathway1871767
      Glutathione S-transferase Mu 5P4643925/7.3Yes, androgen1p13.3Metabolism: energy pathway1871451
      Glutathione S-transferase Theta 2P3071227.5/6.0122q11.2Metabolism: energy pathway1501035
      Glycerol-3-phosphate dehydrogenase (NAD+), cytoplasmic, GPDH-CP2169538/5.8Yes, androgen12q12-q13Immune response: protein cargo30421 (PSD: 1556.79)53
      Haptoglobin-2P0073846/6.1Yes, androgen16q22.1Protein metabolism: chaperone1081120
      Heat shock 27-kDa proteinP0479222.8/5.9Yes, androgen7q11.23Protein metabolism: chaperone1401041
      Heat shock 70-kDa protein 9B (glucose-regulated protein 75, grp75)P3864673/5.85q31.1Protein metabolism: chaperone1331216
      Heat shock 70-kDa protein 1 (hsp70.1)P0810770/5.5Yes6p21.3Protein metabolism: chaperone2201726
      Heat shock 70-kDa protein 5 (78-kDa glucose-regulated protein precursor, grp78)P1102172/5.0Yes, androgen9q34Protein metabolism: heat shock protein1301425
      Heat shock protein 60 (hsp60)P1080961/5.7Yes2q33.1Protein metabolism: chaperone1441435
      Heat shock protein 90 (hsp90)P0790085/5.0Yes, androgen14q32Transport/cargo protein1992130
      Hemoglobin β chainP6887116/7.011p15.5Metabolism: energy pathway2121484
      Hydroxymethylglutaryl-CoA synthaseP5486857/8.4Yes, androgen1p13-p12Metabolism: energy pathway1251316
      Isocitrate dehydrogenase (NADP)O7587446.9/6.5Yes, androgen2q33.3Metabolism: energy pathway1711530
      l-Lactate dehydrogenase B chainP0719536.8/5.712p12.2-p12.1Signal transduction: cell communication17212 (PSD: 1262.63)31
      L-plastin (lymphocyte cytosolic protein 1)P1379671/5.2Yes, androgen13q14.3Metabolism: energy pathway3793858
      Methylcrotonoyl-CoA carboxylase α chainQ96RQ381/7.63q25-q27Metabolism: energy pathway1401414
      Methylcrotonoyl-CoA carboxylase β chain, mitochondrialQ9HCC062/7.55q12-q13Metabolism: energy pathway1601526
      NADP-dependent malic enzymeP4816365/5.8Yes6q12Metabolism: energy pathway1401523
      Peroxiredoxin 2P3211922/5.4Yes, androgen19p13.2Metabolism: energy pathway2331345
      Peroxiredoxin 6P3004125/6.021q24.2Biological process unknown11010 (PSD: 1409.66)37
      Protein-disulfide isomeraseP0723755/4.7Yes17q25Protein metabolism2252031
      Protein-disulfide isomerase A3P3010157/5.915q15Membrane transport protein2302134
      Sterol carrier protein 2 (SCP-2)P2230759.8/6.4Yes, androgen1p32Protein metabolism1401420
      Stress-induced phosphoprotein 1 (STI1)P3194863.4/6.411q13Protein metabolism: chaperone1011528
      T-complex protein 1, α subunit (TCP-1α)P1798761/5.8Yes6q25-q27Metabolism: energy pathway1521321
      TransketolaseP2940168/7.6Yes3p14.3Transport/cargo protein1401121
      TransthyretinP0276616/5.5Yes, androgen18q11.2-q12.1Cell growth and/or maintenance: cytoskeletal protein1407 (PSD: 1366.70)57
      Tropomyosin α 4 chainP6793628.6/4.619p13.1Protein metabolism: heat shock protein2402046
      %
      Vacuolar ATP synthase, catalytic subunit AP3860668.7/5.35Yes, androgen3q13.31Metabolism: energy pathway1001016
      Vitamin D-binding proteinP0277455/5.4Yes, androgen4q11-q13Transport/cargo protein2511844
      WD repeat protein 1O7508366/6.1Yes, androgen4p16.1Cell growth and/or maintenance: cytoskeletal protein10010 (PSD: 1540.75)10
      Zinc-α2-glycoproteinP2531134/5.5Yes, androgen7q22.1Immune response: adhesion molecule21417 (PSD: 1127.59, 1464.68)44
      AAA ATPase p97 (valosin-containing protein)P5507290/5.19p13-p12Metabolism: energy pathway3814054
      a Whenever possible, the Swiss-Prot accession number was used.
      b The Mr and pI were calculated directly from the sequences. In some cases these values do not fit with the apparent molecular weights observed in the 2D gels shown in Fig. 2A.
      c Proteins whose expression is known to be regulated by hormones are indicated.
      d Molecular function as well as biological process was assigned in accordance with the Human Protein Reference Database (www.hprd.org).
      As expected, the apocrine cysts expressed major cyst fluid proteins like GCFDP-15, apoD (
      • Balbin M.
      • Freije J.M.
      • Fueyo A.
      • Sanchez L.M.
      • Lopez-Otin C.
      Apolipoprotein D is the major protein component in cyst fluid from women with human breast gross cystic disease.
      • Sanchez L.M.
      • Diez-Itza I.
      • Vizoso F.
      • Lopez-Otin C.
      Cholesterol and apolipoprotein D in gross cystic disease of the breast.
      • Mazoujian G.
      • Pinkus G.S.
      • Davis S.
      • Haagensen Jr., D.E.
      Immunohistochemistry of a gross cystic disease fluid protein (GCDFP-15) of the breast. A marker of apocrine epithelium and breast carcinomas with apocrine features.
      • Sapino A.
      • Caisson P.
      • Bussolati G.
      Gross cystic disease fluid protein (GCDFP-15) in the breast: past and present.
      ), and zinc-α2-glycoprotein (
      • Bundred N.J.
      • Miller W.R.
      • Walker R.A.
      An immunohistochemical study of the tissue distribution of the breast cyst fluid protein, zinc α2 glycoprotein.
      ,
      • Bundred N.J.
      • Scott W.N.
      • Davies S.J.
      • Miller W.R.
      • Mansel R.E.
      Zinc α-2 glycoprotein levels in serum and breast fluids: a potential marker of apocrine activity.
      ), a fact that was confirmed by 2D PAGE analysis of cyst fluid 81 (Figs. 3A) as well as of the nine other cyst fluids investigated (a few examples are presented in Fig. 3, B and C). Loading of the gels with higher amounts of cyst fluid proteins allowed the additional identification of lower abundance proteins such as 15-PGDH, HMG-CoA synthase, cathepsin D, vitamin D-binding protein, hsp60, and a few others (Fig. 3D).
      Figure thumbnail gr3
      Fig. 3IEF 2D PAGE of apocrine cyst fluid proteins.A, cyst fluid 81. B, cyst fluid 80. C, cyst fluid 65. D, higher load of proteins from cyst fluid 81. Major proteins are indicated for reference.
      A number of studies have indicated that cytokines as well as growth factors are produced by cysts cells (
      • Tapper D.
      • Gajdusek C.
      • Moe R.
      • Ness J.
      Identification of a unique biological tumor marker in human breast cyst fluid and breast cancer tissue.
      • Reed M.J.
      • Coldham N.G.
      • Patel S.R.
      • Ghilchik M.W.
      • James V.H.
      Interleukin-1 and interleukin-6 in breast cyst fluid: their role in regulating aromatase activity in breast cancer cells.
      • Ness J.C.
      • Sedghinasab M.
      • Moe R.E.
      • Tapper D.
      Identification of multiple proliferative growth factors in breast cyst fluid.
      • Lai L.C.
      • Kadory S.
      • Siraj A.K.
      • Lennard T.W.
      Oncostatin M, interleukin 2, interleukin 6 and interleukin 8 in breast cyst fluid.
      • Lai L.C.
      • Siraj A.K.
      • Erbas H.
      • Lennard T.W.
      Transforming growth factor α, β1 and β2 in breast cyst fluid.
      ). These observations as well as compelling evidence suggesting a role in mammary epithelia development (
      • Gouon-Evans V.
      • Lin E.Y.
      • Pollard J.W.
      Requirement of macrophages and eosinophils and their cytokines/chemokines for mammary gland development.
      ) prompted us to perform a cytokine and growth factor analysis of the cyst fluid. Although the amount of fluid collected from the apocrine macrocysts was generally too small to warrant reliable analysis using antibody arrays (
      • Celis J.E.
      • Moreira J.M.
      • Cabezon T.
      • Gromov P.
      • Friis E.
      • Rank F.
      • Gromova I.
      Identification of extracellular and intracellular signaling components of the mammary adipose tissue and its interstitial fluid in high risk breast cancer patients: toward dissecting the molecular circuitry of epithelial-adipocyte stromal cell interactions.
      ,
      • Celis J.E.
      • Gromov P.
      • Cabezon T.
      • Moreira J.M.
      • Ambartsumian N.
      • Sandelin K.
      • Rank F.
      • Gromova I.
      Proteomic characterization of the interstitial fluid perfusing the breast tumor microenvironment: a novel resource for biomarker and therapeutic target discovery.
      ), we were able to obtain a single cytokine pattern from a small sample collected from patient 94, revealing the presence of a number of cytokines and growth factors in the cyst fluid, including IL-1α, IL-10, IL-12 p40, NAP-2, macrophage colony-stimulating factor, epidermal growth factor, basic fibroblast growth factor, RANTES (regulated on activation normal T-cell expressed and secreted), angiogenin, TNF-α, eotaxin-3, angiopoietin, betacellulin, Fas, growth-related oncogene, IGFBP-3, macrophage migration inhibitory factor, MIP-1α, TRAIL-R4, TIMP-1, TIMP-2, vascular endothelial growth factor, and vascular endothelial growth factor-D (Fig. 4). To substantiate this result we analyzed four additional fluid samples collected from large apocrine cysts of women free of breast cancer obtaining nearly identical cytokine patterns, thus corroborating the presence of many cytokines and growth factors in the cyst fluid of which only a few have been reported previously in the literature (
      • Enriori C.L.
      • Novelli J.E.
      • Cremona M.D.C.
      • Hirsig R.J.P.
      • Enriori P.J.
      Biochemical study of cyst fluid in human breast cystic disease a review.
      ,
      • Lai L.C.
      • Kadory S.
      • Siraj A.K.
      • Lennard T.W.
      Oncostatin M, interleukin 2, interleukin 6 and interleukin 8 in breast cyst fluid.
      ,
      • Lai L.C.
      • Siraj A.K.
      • Erbas H.
      • Lennard T.W.
      Transforming growth factor α, β1 and β2 in breast cyst fluid.
      ,
      • Lai L.C.
      • Kadory S.
      • Cornell C.
      • Lennard T.W.
      Possible regulation of soluble ICAM-1 levels by interleukin-1 in a sub-set of breast cysts.
      • Lai L.C.
      • Siraj A.K.
      • Erbas H.
      • Lennard T.W.
      Relationship between basic fibroblast growth factor and transforming growth factor-β2 in breast cyst fluid.
      • Lai L.C.
      • Erbas H.
      • Meadows K.A.
      • Lennard T.W.
      • Holly J.M.
      Insulin-like growth factor binding protein-3 in breast cyst fluid: relationships with insulin-like growth factors I and II and transforming growth factor-β 1 and 2.
      ). In addition, the 2D PAGE profiles of the four fluids were indistinguishable from those obtained with the cyst fluids of breast cancer patients (results not shown).
      Figure thumbnail gr4
      Fig. 4Cytokine profiling of apocrine cyst fluids. Cyst fluid was collected from a cyst of one breast cancer patient and from large apocrine cysts of four patients, free of cancer, attending the mammography clinic. Cytokine-specific antibody arrays (RayBio Human Cytokine Array Series 1000, RayBiotech, Inc.) were incubated with cyst fluid, and detection of bound cytokines was performed according to the manufacturer’s instructions. Shown are the results obtained with the cyst fluid of the breast cancer patient (cyst 1) and one of the four cyst fluids collected from mammography clinic patients (cyst 2). The apocrine nature of the cells composing the cysts was confirmed by cytology (exemplified by hematoxylin staining of cyst 2, shown in rightmost panel). POS, positive; NEG, negative; PDGF, platelet-derived growth factor; RANTES, regulated on activation normal T-cell expressed and secreted; MCP, monocyte chemotactic protein; CNTF, ciliary neurotrophic factor; BMP, bone morphogenic protein; FGF, fibroblast growth factor; SDF, stromal cell-derived factor; M-CSF, macrophage colony-stimulating factor; GM-CSF, granulocyte-macrophage colony-stimulating factor; BDNF, brain-derived neurotrophic factor; BLC, B lymphocyte chemoattractant; SCF, stem cell factor; MDC, macrophage-derived chemokine; MIG, γ interferon-induced monokine; MIP, macrophage inflammatory protein; IGFBP, insulin-like growth factor-binding protein; NT, neurotrophin; TARC, thymus and activation-regulated chemokine; PARC, pulmonary and activation-regulated chemokine; IFN, interferon; CTACK, cutaneous T-cell-attracting chemokine; ICAM, intercellular adhesion molecule; I-TAC, interferon-γ-inducible T-cell α chemoattractant; TECK, thymus-expressed chemokine; EGF-R, epidermal growth factor receptor; TRAIL, tumor necrosis factor-related apoptosis-inducing ligand; VEGF, vascular endothelial growth factor; GITR, glucocorticoid-induced TNF receptor; HCC, human CC chemokine; PIGF, phosphatidylinositol-glycan biosynthesis, class F protein; bFGF, basic fibroblast growth factor; uPAR, urokinase plasminogen activator surface receptor precursor; R, receptor; sTNF, soluble TNF; HGF, hepatocyte growth factor; MSP, macrophage-stimulating protein; GRO, growth-related oncogene; ENA-78, epithelial neutrophil-activating protein 78.
      As a whole, our results point toward the apocrine cyst as being a gland capable of secreting a wide variety of macromolecules that may play key roles in the mechanisms underlying cyst evolution. Also the presence of inflammatory cytokines in the fluid may explain why rupture of the cysts can elicit inflammation. As expected from published data, apocrine cells in all type I macrocysts analyzed were ER-α-negative (Fig. 5A), PR-negative (not shown), AR-positive (Fig. 5B) (
      • Tavassoli F.A.
      • Purcell C.A.
      • Man Y.G.
      Down regulation of bcl2, ER, PR associated with AR expression is triggered by apocrine differentiation of mammary epithelium.
      • Gatalica Z.
      Immunohistochemical analysis of apocrine breast lesions. Consistent over-expression of androgen receptor accompanied by the loss of estrogen and progesterone receptors in apocrine metaplasia and apocrine carcinoma in situ.
      • Selim A.G.
      • Wells C.A.
      Immunohistochemical localisation of androgen receptor in apocrine metaplasia and apocrine adenosis of the breast: relation to oestrogen and progesterone receptors.
      ), and ErbB-2-positive (Fig. 5C) (
      • Tran D.D.
      • Lawson J.S.
      Microcysts and breast cancer: a study of biological markers in archival biopsy material.
      ) and exhibited a low proliferation index as judged by staining with the Ki67 antibody (Fig. 5D) (
      • Moriya T.
      • Sakamoto K.
      • Sasano H.
      • Kawanaka M.
      • Sonoo H.
      • Manabe T.
      • Ito J.
      Immunohistochemical analysis of Ki-67, p53, p21, and p27 in benign and malignant apocrine lesions of the breast: its correlation to histologic findings in 43 cases.
      ).
      Figure thumbnail gr5
      Fig. 5IHC pictures of apocrine microcyst 65 immunostained with antibodies against ER-α (A), AR (B), ErbB-2 (C), and Ki67 (D), respectively.

       Identification of Markers of Apocrine Metaplasia—

      Comparison of the protein profiles of apocrine macrocysts (Fig. 2) with those of tumor and non-malignant tissues, exemplified in Fig. 6, A–F, revealed a few major cyst proteins that were either absent or present at much lower levels (factor of 4 or more) in non-malignant (Fig. 6, A, C, and E) and tumor (Fig. 6, B, D, and F) tissues. In addition, our studies identified several proteins that were preferentially expressed by malignant cells in some of the tumors. One of these, psoriasin (
      • Madsen P.
      • Rasmussen H.H.
      • Leffers H.
      • Honore B.
      • Dejgaard K.
      • Olsen E.
      • Kiil J.
      • Walbum E.
      • Andersen A.H.
      • Basse B.
      • Celis J.E.
      Molecular cloning, occurrence, and expression of a novel partially secreted protein “psoriasin” that is highly up-regulated in psoriatic skin.
      ), is described in more detail below in the context of apocrine tumors.
      Figure thumbnail gr6
      Fig. 6Silver-stained IEF 2D gels of whole protein extracts from non-malignant and tumor tissues.A, non-malignant tissue 81. B, tumor 81. C, non-malignant tissue 64. D, tumor 64. E, non-malignant tissue 67. F, tumor 67. Arrows indicate some proteins, or their positions, that are preferentially expressed by the apocrine macrocysts.
      Putative apocrine cell biomarkers included 15-PGDH, HMG-CoA synthase, cathepsin D, β-glucuronidase, GCFDP-15, apoJ, and apoD (Table II). Antibodies against 15-PGDH, cathepsin D, GCFDP-15, apoJ, and apoD were used to confirm the 2D gel results by IHC using paraffin sections from cyst 81 as well as non-malignant and tumor tissue obtained from the same patient. Fig. 7 shows representative IHC pictures of tissue sections from cyst, non-malignant tissue, and tumor 81 immunostained with 15-PGDH (Fig. 7, A–C), GCDFP-15 (Fig. 7, D–F), cathepsin D (Fig. 7, G–I), apoD (Fig. 7, J–L), and apoJ (Fig. 7, M–O) antibodies, respectively. The results showed a clear correlation between the gel-derived data and the expression levels observed by IHC. GCDFP-15, cathepsin D, apoD, and apoJ were expressed predominantly, but not solely, by the apocrine epithelia as variable levels of staining were also observed in some cases in the non-malignant glands (Fig. 7H), the surrounding stroma (Fig. 7E), and the tumor (Fig. 7, F and I). IHC analysis of all available macrocysts indicated that only 15-PGDH was expressed exclusively by the apocrine epithelial cells, a fact that was also corroborated by 2D PAGE analysis, underscoring its potential value as a marker for these lesions. A 2D gel immunoblot (non-equilibrium pH gradient electrophoresis and IEF) showing the antigen specificity of the 15-PGDH antibody is presented in Fig. 8. In addition, matching of IHC results for a given tissue specimen with the corresponding 2D PAGE gel pattern provided cross-validation of antigen specificity for IHC.
      Table IIIImmunohistochemical detection of markers in pure apocrine carcinomas
      Tumor markerRH-34697RH-21990RH-9777RH-10515RH-13319RH-24678
      Carcinoma in situInvasive areaCarcinoma in situInvasive areaCarcinoma in situInvasive areaCarcinoma in situInvasive areaCarcinoma in situInvasive areaCarcinoma in situInvasive area
      ER-α, PRNegNegNegNegNegNegNegNegNegNegNegNeg
      p53NegNegNegPosNegNegWeak PosWeak PosNegNegPosPos
      15-PGDHPosPosPosPosPosPosPosPosPosPosNegNeg
      Lack of expression of 15-PGDH by tumor RH-24678 was confirmed by 2D PAGE (see Fig. 15D).
      HMG-CoA reductasePosPosPosNegNegNegNegNegNegNegNegNeg
      COX-2PosPosNegNegPosNegPosPosNegNegPosPos
      PsoriasinPosPosPos and NegPosNegNegPosPosPosPosPosPos
      a Lack of expression of 15-PGDH by tumor RH-24678 was confirmed by 2D PAGE (see Fig. 15D).
      Figure thumbnail gr7
      Fig. 7IHC pictures of apocrine macrocyst (A, D, G, J, and M), non-malignant tissue (B, E, H, K, and N), and tumor (C, F, I, L, and O) from patient 81 immunostained with antibodies against proteins preferentially expressed by apocrine macrocysts.A–C, 15-PGDH; D–F, GCDFP-15; G–I, cathepsin D; J–L, apoD; and M–O, apoJ.
      Figure thumbnail gr8
      Fig. 8Specificity of the rabbit anti-15-PGDH polyclonal antibody as determined by 2D gel immunoblotting (non-equilibrium pH gradient electrophoresis and IEF) of whole protein extracts from a transitional cell carcinoma.

       Validation of the 15-PGDH Biomarker in Apocrine Microcysts Present in a Large Sample Set of Non-malignant Breast Tissue—

      To further validate 15-PGDH as a marker of apocrine cysts and apocrine metaplasia, we analyzed by IHC non-malignant tissue from all the patients enrolled in the study (Table I) as well as retrospective samples from 57 additional high risk breast cancer patients (see “Experimental Procedures”). As shown in Fig. 9A, which is representative of all the sections analyzed, the 15-PGDH antibody specifically stained apocrine microcysts. These were found in 11 of the prospective samples (patients 65, 68, 71, 74, 79, 80, 81, 82, 86, 90, and 94), and in nine of the retrospective ones. The antibody detected not only obvious apocrine microcysts, easily identified by plain histology, but also pinpointed areas within some microcysts where only a few cells showed apocrine differentiation (see Fig. 12B). No reactivity was observed with terminal ductal lobular units (TDLU, Fig. 9B), type II flat microcysts (Fig. 9C), stroma cells (Fig. 9, A–C), or fat tissue (Fig. 9D), validating 15-PGDH as a specific marker for detection of apocrine metaplasia in the breast. Essentially the same results were obtained using a commercially available 15-PGDH polyclonal antibody.
      Figure thumbnail gr9
      Fig. 9IHC staining of apocrine lesions immunostained with the 15-PGDH antibody.A, apocrine macrocyst in patient 81. B, TDLU. C, type II flat microcyst. D, fat tissue.
      Figure thumbnail gr12
      Fig. 12Apocrine metaplastic cells immunostained with antibodies against 15-PGDH.A, duct apocrine metaplasia in patient 81. B and C, intraductal apocrine papillary lesions in patient 81.

       Identification of an Additional Biomarker: HMG-CoA Reductase—

      Although we did not have access to appropriate antibodies against the putative apocrine marker HMG-CoA synthase (Fig. 2) (Ref.
      • Olivier L.M.
      • Krisans S.K.
      Peroxisomal protein targeting and identification of peroxisomal targeting signals in cholesterol biosynthetic enzymes.
      and references therein), it was apparent from data in the literature that this enzyme is co-regulated with HMG-CoA reductase (
      • Morikawa S.
      • Murakami T.
      • Yamazaki H.
      • Izumi A.
      • Saito Y.
      • Hamakubo T.
      • Kodama T.
      Analysis of the global RNA expression profiles of skeletal muscle cells treated with statins.
      ,
      • de Nigris F.
      • Lerman L.O.
      • Napoli C.
      New insights in the transcriptional activity and coregulator molecules in the arterial wall.
      ), a key enzyme in the biosynthesis of mevalonate. Additionally Farmer et al. (
      • Farmer P.
      • Bonnefoi H.
      • Becette V.
      • Tubiana-Hulin M.
      • Fumoleau P.
      • Larsimont D.
      • Macgrogan G.
      • Bergh J.
      • Cameron D.
      • Goldstein D.
      • Duss S.
      • Nicoulaz A.L.
      • Brisken C.
      • Fiche M.
      • Delorenz I.M.
      • Iggo R.
      Identification of molecular apocrine breast tumours by microarray analysis.
      ) using microarrays recently identified a new group of breast tumors, “molecular apocrine,” that expressed genes associated with androgen signaling and lipid metabolism, including HMG-CoA reductase. Encouraged by these data we obtained antibodies against HMG-CoA reductase and performed a comprehensive IHC analysis on all the tissue specimens available to us. Indeed the antibody specifically detected apocrine breast metaplasias (Fig. 10A) just like the 15-PGDH probe. The staining showed a punctuated pattern in line with the peroxisomal localization of the enzyme where mevalonate is further converted to farnesyl diphosphate (
      • Olivier L.M.
      • Krisans S.K.
      Peroxisomal protein targeting and identification of peroxisomal targeting signals in cholesterol biosynthetic enzymes.
      ). No reactivity was observed with terminal ductal lobular units (Fig. 10B), type II microcysts (Fig. 10A), stroma cells (Fig. 10, A and B), or fat tissue (results not shown), suggesting that HMG-CoA reductase is also a bona fide biomarker of breast apocrine metaplasia. From the above results it is likely that HMG-CoA synthase may also represent a candidate marker of apocrine metaplasia. Further studies, however, are needed to confirm this possibility. Having successfully identified specific apocrine biomarkers we were then able to address some of the more unclear and contentious points in the biology of apocrine cysts, namely their origin and mechanisms underlying their development as well as their relationship, if existent, with the cancer phenotype.
      Figure thumbnail gr10
      Fig. 10IHC staining of apocrine lesions immunostained with the HMG-CoA reductase antibody.A, apocrine macrocyst in patient 81. B, TDLU.

       Origin and Enlargement of Apocrine Microcysts—

      It has been suggested that apocrine metaplasia of the epithelium lining the ductules may involve type II microcystic structures containing contiguous cells that stain strongly with ER-α and PR antibodies (
      • Tsung J.S.
      • Wang T.Y.
      • Wang S.M.
      • Yang P.S.
      Cytological and biochemical studies of breast cyst fluid.
      ,
      • Jones C.
      • Damiani S.
      • Wells D.
      • Chaggar R.
      • Lakhani S.R.
      • Eusebi V.
      Molecular cytogenetic comparison of apocrine hyperplasia and apocrine carcinoma of the breast.
      ) (Fig. 11A). Because apocrine cells are ER-α- and PR-negative and given the fact that we observed cystlike structures in which only a few cells expressed 15-PDGH (results not shown), we surmised that some type I cysts may be derived from type II lesions by mechanisms that could involve loss of receptor expression. Accordingly we screened the non-malignant tissue from 20 prospective and retrospective patients harboring apocrine microcysts for the presence of type II microcysts that have partially lost receptor expression as depicted in Fig. 11B. Subsequently we used the immunowalking approach (
      • Celis J.E.
      • Celis P.
      • Palsdottir H.
      • Ostergaard M.
      • Gromov P.
      • Primdahl H.
      • Orntoft T.F.
      • Wolf H.
      • Celis A.
      • Gromova I.
      Proteomic strategies to reveal tumor heterogeneity among urothelial papillomas.
      ), i.e. staining of consecutive sections with antibodies against 15-PDGH, PR, and ER-α, in an effort to determine their relative expression in various areas of the same cyst. Although these structures were rare, we did identify a few, and a representative example from patient 45 (retrospective samples) is shown in Fig. 11, C–E. The relatively large type II microcyst depicted in Fig. 11C exhibits both PR-positive (lower half of the microcyst, open arrow) and -negative areas (upper half of the microcyst, closed arrow) that stained differentially with the 15-PGDH antibody (Fig. 11D). As expected, areas negative for PR stain positively for 15-PGDH (Fig. 11D, upper half of the microcyst), implying that this area of the type II lesion is undergoing apocrine metaplasia. Staining of the next section with the ER-α antibody showed a good correlation with the PR staining (not shown). Interestingly, however, the small 15-PGDH-positive apocrine microcyst observed in Fig. 11, C and D, upper right corners (enclosed by a rectangle), showed groups of cells that expressed ER-α (Fig. 11E, dotted boxes) but that were negative for PR (Fig. 11C, dotted boxes). From these data it seems likely that loss of PR expression precedes that of ER-α receptor during apocrine microcyst development.
      Figure thumbnail gr11
      Fig. 11IHC staining of various cystic structures.A, type II microcyst from patient 45 stained with PR antibodies. B, type II microcyst from patient 45 stained with PR antibodies showing positive and negative cells. C–E, immunowalking of serial paraffin sections from non-malignant tissue of patient 45 immunostained with PR (C), 15-PDGH (D), and ER-α (E) antibodies, respectively. Arrows in C and D show areas that display differential staining with the PR and 15-PGDH antibodies. Dotted boxes in C and E show cells in the apocrine microcysts that stain with the ER-α antibody (E) but that do not express PR (C).
      Data in the literature indicate that apocrine metaplasias may also arise from terminal ducts and intraductal papillary lesions. Indeed both the 15-PGDH and HMG-CoA reductase biomarkers identified these lesions (illustrated in Fig. 12, A–C, with the 15-PGDH antibody). In fact, some of the apocrine intraductal papillary lesions could not be recognized without the aid of the biomarker antibody probes (Fig. 12C).
      Visual analysis of sections from macrocysts often showed apocrine intraductal papillary structures that stained positively with the 15-PGDH antibody and that protruded into the cyst cavity from opposite sides (Fig. 13A), suggesting that microcysts may combine or merge to generate larger cysts. Indeed careful IHC analysis of several areas of macrocyst 81, which was composed of several microcysts (Fig. 13A), revealed regions where apocrine cells from two adjacent microcysts had become very close to each other, separated only by a thin layer of stroma (Fig. 13B). Cells in the middle of the bridges showed loss of nuclear content as judged by hematoxylin counterstaining (Fig. 13B, arrow), a fact that was corroborated in different preparations by using a variety of antibodies against nuclear proteins (Fig. 13C). Loss of nuclear content is most likely not due to apoptosis because these cells did not stain with an antibody specific for active caspase-3 (data not shown) or show any of the characteristic alterations in cell morphology associated with programmed cell death. The data support a model for growth of apocrine cysts where coalescence of two microcysts with subsequent rupture of the connecting bridge(s), either by fluid pressure alone or in combination with peptidases present in the fluid (
      • Borchert G.H.
      • Melegos D.N.
      • Yu H.
      • Giai M.
      • Roagna R.
      • Ponzone R.
      • Sgro L.
      • Diamandis E.P.
      Quantification of pepsinogen C and prostaglandin D synthase in breast cyst fluid and their potential utility for cyst type classification.
      ), leads to cyst enlargement. Apocrine cells devoid of nuclear staining were also observed in apocrine cysts lined by a single layer of cuboidal cells (not shown), but the mechanism underlying this phenomenon remains unknown.
      Figure thumbnail gr13
      Fig. 13Cyst structures that may be involved in microcyst enlargement.A, apocrine macrocyst 81 sections immunostained with antibodies against 15-PGDH. B and C, as above but stained with antibodies against COX-2 (B) and a nuclear antigen (RN3) (C), respectively. Apocrine cells devoid of nuclear content are indicated with arrows in B and C.

       Apocrine Biomarkers and the Cancer Phenotype—

      An important question that has been debated for some time concerns the relationship between apocrine changes and breast carcinoma (
      • Trenkic S.
      • Katic V.
      • Pashalina M.
      • Zivkovic V.
      • Milentijevic M.
      • Kostov M.
      The histologic spectrum of apocrine lesions of the breast.
      ,
      • Tran D.D.
      • Lawson J.S.
      Microcysts and breast cancer: a study of biological markers in archival biopsy material.
      ,
      • Viacava P.
      • Naccarato A.G.
      • Bevilacqua G.
      Apocrine epithelium of the breast: does it result from metaplasia?.
      ,
      • Jones C.
      • Damiani S.
      • Wells D.
      • Chaggar R.
      • Lakhani S.R.
      • Eusebi V.
      Molecular cytogenetic comparison of apocrine hyperplasia and apocrine carcinoma of the breast.
      ). Published studies have shown contradictory results as different authors have described apocrine lesions as either representing a risk factor, non-obligate precursor lesions of apocrine carcinoma, or benign lesions with no correlation with malignancy (
      • Trenkic S.
      • Katic V.
      • Pashalina M.
      • Zivkovic V.
      • Milentijevic M.
      • Kostov M.
      The histologic spectrum of apocrine lesions of the breast.
      ,
      • Ahmed A.
      Calcification in human breast carcinomas: ultrastructural observations.
      • Yates A.J.
      • Ahmed A.
      Apocrine carcinoma and apocrine metaplasia.
      • Page D.L.
      • Dupont W.D.
      • Jensen R.A.
      Papillary apocrine change of the breast: associations with atypical hyperplasia and risk of breast cancer.
      • Viacava P.
      • Naccarato A.G.
      • Bevilacqua G.
      Apocrine epithelium of the breast: does it result from metaplasia?.
      • Jones C.
      • Damiani S.
      • Wells D.
      • Chaggar R.
      • Lakhani S.R.
      • Eusebi V.
      Molecular cytogenetic comparison of apocrine hyperplasia and apocrine carcinoma of the breast.
      • Durham J.R.
      • Fechner R.E.
      The histologic spectrum of apocrine lesions of the breast.
      ,
      • O’Malley F.P.
      • Bane A.L.
      The spectrum of apocrine lesions of the breast.
      ).
      To shed some light into the problem and taking advantage of the fact that we had collected non-malignant and tumor tissue from the same patients, we analyzed by IHC the 93 sets of matched samples available to us using the 15-PGDH and HMG-CoA reductase antibodies with the aim of assessing the relationship, if any, between benign apocrine changes and breast carcinoma. Of all the lesions analyzed, including the retrospective samples, which consisted of 75 ductal carcinomas, 15 lobular carcinomas, one apocrine carcinoma, one mucinous carcinoma, and one tubular lesion, only tumor cells in patient 79, which presented an invasive apocrine carcinoma (Table I), expressed one of the apocrine markers, namely 15-PGDH. Expression of both markers, however, was detected in intraductal papillomas showing apocrine atypia that were observed in or very close to the invasive area of the ductal carcinoma in patient 71 (Table I). These results are described in detail below.

       Tumor 79—

      Tumor 79 expressed 15-PGDH (Fig. 14A), albeit at much lower levels as compared with the reference apocrine microcysts (Fig. 7A; see also Figs. 2 and 15A), but did not express detectable levels of HMG-CoA reductase (Fig. 14B). The lesion was p53-, ER-, and PR-negative (Table I), and about 50% of the tumor cells stained with the AR antibody (Fig. 14C) as expected for an apocrine carcinoma (
      • Farmer P.
      • Bonnefoi H.
      • Becette V.
      • Tubiana-Hulin M.
      • Fumoleau P.
      • Larsimont D.
      • Macgrogan G.
      • Bergh J.
      • Cameron D.
      • Goldstein D.
      • Duss S.
      • Nicoulaz A.L.
      • Brisken C.
      • Fiche M.
      • Delorenz I.M.
      • Iggo R.
      Identification of molecular apocrine breast tumours by microarray analysis.
      ,
      • Bratthauer G.L.
      • Lininger R.A.
      • Man Y.G.
      • Tavassoli F.A.
      Androgen and estrogen receptor mRNA status in apocrine carcinomas.
      ,
      • Honma N.
      • Takubo K.
      • Akiyama F.
      • Sawabe M.
      • Arai T.
      • Younes M.
      • Kasumi F.
      • Sakamoto G.
      Expression of GCDFP-15 and AR decreases in larger or node-positive apocrine carcinomas of the breast.
      ). Tumor 79 expressed a number of proteins not present in apocrine cysts or normal gland cells as judged by 2D PAGE (data not shown), and of these, psoriasin (
      • Madsen P.
      • Rasmussen H.H.
      • Leffers H.
      • Honore B.
      • Dejgaard K.
      • Olsen E.
      • Kiil J.
      • Walbum E.
      • Andersen A.H.
      • Basse B.
      • Celis J.E.
      Molecular cloning, occurrence, and expression of a novel partially secreted protein “psoriasin” that is highly up-regulated in psoriatic skin.
      ) was the most striking because this protein has been shown to be highly up-regulated in ductal carcinoma in situ (CIS) and associated with a worse prognosis in estrogen receptor-negative invasive ductal carcinomas (
      • Emberley E.D.
      • Alowami S.
      • Snell L.
      • Murphy L.C.
      • Watson P.H.
      S100A7 (psoriasin) expression is associated with aggressive features and alteration of Jab1 in ductal carcinoma in situ of the breast.
      ,
      • Carlsson H.
      • Yhr M.
      • Petersson S.
      • Collins N.
      • Polyak K.
      • Enerback C.
      Psoriasin (S100A7) and Calgranulin-B (S100A9) induction is dependent on reactive oxygen species and is downregulated by Bcl-2 and antioxidants.
      ). Expression of psoriasin by tumor 79 was confirmed by IHC using a monoclonal antibody specific for this protein (Fig. 14D) (
      • Lai L.C.
      • Siraj A.K.
      • Erbas H.
      • Lennard T.W.
      Relationship between basic fibroblast growth factor and transforming growth factor-β2 in breast cyst fluid.
      ).
      Figure thumbnail gr14
      Fig. 14IHC pictures of invasive apocrine breast tumors.A–D, apocrine cancer 79 immunostained with antibodies against 15-PGDH (A), HMG-CoA reductase (B), AR (C), and psoriasin (D), respectively. E and G, apocrine carcinoma RH-21990 immunostained with 15-PGDH (E) and HMG-CoA reductase antibodies (G), respectively. F and H, CIS from carcinoma RH-21990 immunostained with 15-PGDH (F) and HMG-CoA reductase antibodies (H), respectively.
      Figure thumbnail gr15
      Fig. 15Expression of 15-PGDH by apocrine tumors. Silver-stained IEF 2D gels of whole protein extracts from tumor 79 (A), tumor RH-9777 (B), tumor RH-21990 (C), and tumor RH-24678 (D) are shown.
      Because the expression of 15-PGDH by tumor 79 suggested a link between apocrine metaplasia and the tumor phenotype, we analyzed six pure apocrine carcinomas by IHC for expression of the two apocrine-specific biomarkers, 15-PGDH and HMG-CoA reductase. Of these, five stained positively with the 15-PGDH antibody, which decorated both invasive and apocrine CIS cells in each one of the tumors (RH-34697, RH-21990, RH-9777, RH-10515, and RH-13319; Table III). These results were confirmed by 2D PAGE as shown in Fig. 15, B and C, which presents selected areas of silver-stained gels of whole protein extracts of apocrine tumors RH-9777 (Fig. 15B), RH-21990 (Fig. 15C), and RH-24678 (Fig. 15D), a lesion that did not stain with the 15-PGDH antibody (Table III). All five apocrine cancers were ER-α- and PR-negative (Table III). Interestingly only one of the 15-PGDH-positive tumors (RH-34697) expressed HMG-CoA reductase both in the invasive and CIS apocrine cells, whereas another (RH-21990) expressed the enzyme solely in the CIS cells (Table III). Fig. 14 shows IHC pictures of apocrine tumor RH-21990, which stained positively with the 15-PGDH antibody (Fig. 14E) but was devoid of HMG-CoA reductase (Fig. 14G). Areas within the tumor exhibiting apocrine CIS cells that are thought to be a precursor of invasive disease stained positively with both antibodies (Fig. 14, F and H), suggesting that the expression of HMG-CoA reductase may be lost prior to 15-PGDH during cancer progression (see below). With one exception (RH-9777), all tumors expressing 15-PGDH also stained positively with the psoriasin antibody, including apocrine CIS cells (Table III), suggesting that this protein may be a valuable marker for late events in apocrine cancer progression, a process that may encompass benign metaplasia, hyperplasia, atypia, CIS, and invasive disease (
      • Durham J.R.
      • Fechner R.E.
      The histologic spectrum of apocrine lesions of the breast.
      ,
      • O’Malley F.P.
      • Page D.L.
      • Nelson E.H.
      • Dupont W.D.
      Ductal carcinoma in situ of the breast with apocrine cytology: definition of a borderline category.
      ,
      • Leal C.
      • Henrique R.
      • Monteiro P.
      • Lopes C.
      • Bento M.J.
      • De Sousa C.P.
      • Lopes P.
      • Olson S.
      • Silva M.D.
      • Page D.L.
      Apocrine ductal carcinoma in situ of the breast: histologic classification and expression of biologic markers.
      ).

       Tumor 71—

      Immunohistochemical analysis of the tumor of patient 71, which was diagnosed as an invasive ductal carcinoma (Table I), and areas surrounding it revealed two types of intraductal papillary lesions with apocrine atypia that showed similar and yet distinct morphologies and phenotypes. Atypia a stained positively with the 15-PGDH (Fig. 16A) and HMG-CoA reductase (Fig. 16B) antibodies, exhibited some p53-positive cells (Fig. 16C), and expressed psoriasin (Fig. 16D). Atypia b, on the other hand, expressed similar levels of 15-PGDH (Fig. 16E), p53 (Fig. 16G), and psoriasin (Fig. 16H) but exhibited greatly reduced levels of HMG-CoA reductase (Fig. 16F). Because the analysis of the pure apocrine cancers suggested that the expression of HMG-CoA reductase may be lost prior to 15-PDGH reductase, we believe that there is a sequential relationship between the two atypias. The invasive component of the tumor was p53- and psoriasin-positive but did not show immunoreactivity with the 15-PGDH or the HMG-CoA reductase antibodies (results not shown). Whether tumor 71 was misclassified and corresponds to an apocrine lesion that has lost the expression of the two apocrine markers or corresponds to a mixed type is at present unknown, emphasizing the need to generate additional markers to cover all the intermediate stages involved in apocrine cancer progression.
      Figure thumbnail gr16
      Fig. 16Apocrine atypias in patient 71.A–D, apocrine atypia a in patient 71 immunostained with antibodies against 15-PGDH (A), HMG-CoA reductase (B), p53 (C), and psoriasin (D). E–H, apocrine atypia b immunostained with antibodies against 15-PGDH (E), HMG-CoA reductase (F), p53 (G), and psoriasin (H), respectively.

      DISCUSSION

      Up to one-third of the women aged 30–50 years have cysts in their breasts (
      • Wellings S.R.
      • Alpers C.E.
      Apocrine cystic metaplasia: subgross pathology and prevalence in cancer-associated versus random autopsy breasts.
      ), and both their frequency and their proposed association with an increased risk of breast cancer have underpinned the importance of investigating these lesions at the molecular level (
      • Jones C.
      • Damiani S.
      • Wells D.
      • Chaggar R.
      • Lakhani S.R.
      • Eusebi V.
      Molecular cytogenetic comparison of apocrine hyperplasia and apocrine carcinoma of the breast.
      ,
      • Selim A.G.
      • Ryan A.
      • El-Ayat G.
      • Wells C.A.
      Loss of heterozygosity and allelic imbalance in apocrine metaplasia of the breast: microdissection microsatellite analysis.
      ). In particular, apocrine changes, which span from benign metaplasias to atypical apocrine hyperplasia (
      • Gatalica Z.
      Immunohistochemical analysis of apocrine breast lesions. Consistent over-expression of androgen receptor accompanied by the loss of estrogen and progesterone receptors in apocrine metaplasia and apocrine carcinoma in situ.
      ,
      • Jones C.
      • Damiani S.
      • Wells D.
      • Chaggar R.
      • Lakhani S.R.
      • Eusebi V.
      Molecular cytogenetic comparison of apocrine hyperplasia and apocrine carcinoma of the breast.
      ), have attracted much attention because apocrine epithelia may be a precursor in malignant transformation. Their relationship, however, with invasive breast cancer remains contentious despite much research effort (
      • Trenkic S.
      • Katic V.
      • Pashalina M.
      • Zivkovic V.
      • Milentijevic M.
      • Kostov M.
      The histologic spectrum of apocrine lesions of the breast.
      ,
      • Ahmed A.
      Calcification in human breast carcinomas: ultrastructural observations.
      ,
      • Yates A.J.
      • Ahmed A.
      Apocrine carcinoma and apocrine metaplasia.
      ,
      • Viacava P.
      • Naccarato A.G.
      • Bevilacqua G.
      Apocrine epithelium of the breast: does it result from metaplasia?.
      • Jones C.
      • Damiani S.
      • Wells D.
      • Chaggar R.
      • Lakhani S.R.
      • Eusebi V.
      Molecular cytogenetic comparison of apocrine hyperplasia and apocrine carcinoma of the breast.
      • Durham J.R.
      • Fechner R.E.
      The histologic spectrum of apocrine lesions of the breast.
      ).
      The available data indicate that most breast cysts arise from apocrine metaplastic lobules (
      • Wellings S.R.
      • Alpers C.E.
      Apocrine cystic metaplasia: subgross pathology and prevalence in cancer-associated versus random autopsy breasts.
      ) and that type I apocrine cysts are more likely to progress to large lesions than type II flat microcysts (
      • Dixon J.M.
      • Scott W.N.
      • Miller W.R.
      Natural history of cystic disease: the importance of cyst type.
      ). Currently, however, very little is known as to the molecular mechanisms underlying apocrine metaplasia, a process that entails the conversion of breast epithelial cells into sweat gland type cells (
      • O’Malley F.P.
      • Bane A.L.
      The spectrum of apocrine lesions of the breast.
      ). Breast apocrine cells are cytologically identical to cells in apocrine glands (
      • O’Malley F.P.
      • Bane A.L.
      The spectrum of apocrine lesions of the breast.
      ,
      • Losi L.
      • Lorenzini R.
      • Eusebi V.
      • Bussolati G.
      Apocrine differentiation in invasive carcinoma of the breast.
      ) and exhibit rather large vesicular nuclei with prominent nucleoli and abundant eosinophilic cytoplasm that occasionally present apical snouts that slough off into the lumen. So far, no specific biomarker has been available to characterize breast apocrine lesions.
      The study presented here was undertaken in an effort to dissect and gain a better understanding of the steps leading to breast apocrine metaplasia and ultimately intended to generate specific biomarkers that may enlighten its relationship with cancer phenotype. The strategy we applied was based on the comparison of the protein expression profiles of breast apocrine macrocysts with those of matched cancerous and non-malignant tissues from the same patient followed by biomarker identification, antibody preparation, and validation using a large sample size. Among all of the markers identified for which high quality antibodies could be obtained, only 15-PGDH and HMG-CoA reductase were expressed specifically by breast benign apocrine metaplasias arising from dilated type II microcysts, terminal ducts, and intraductal papillary lesions (Fig. 17).
      Figure thumbnail gr17
      Fig. 17Expression of apocrine markers at various stages of apocrine carcinoma progression. Apocrine lesions span from apocrine metaplasia, apocrine atypia, CIS, and invasive apocrine carcinoma. The data concerning the apocrine CIS and the invasive apocrine carcinomas were derived from the IHC analysis of the six pure apocrine cancers. All other data were derived from the prospective and retrospective samples from the 93 high risk breast cancer patients. Markers in yellow are expressed already in benign metaplasias, whereas those indicated with red seem to be expressed at a later stage.
      15-PGDH is a prostaglandin (PG)-degrading enzyme whose levels have been shown to be down-regulated in bladder (
      • Celis J.E.
      • Ostergaard M.
      • Basse B.
      • Celis A.
      • Lauridsen J.B.
      • Ratz G.P.
      • Andersen I.
      • Hein B.
      • Wolf H.
      • Orntoft T.F.
      • Rasmussen H.H.
      Loss of adipocyte-type fatty acid binding protein and other protein biomarkers is associated with progression of human bladder transitional cell carcinomas.
      ), colon (
      • Backlund M.G.
      • Mann J.R.
      • Holla V.R.
      • Buchanan F.G.
      • Tai H.H.
      • Musiek E.S.
      • Milne G.L.
      • Katkuri S.
      • DuBois R.N.
      15-Hydroxyprostaglandin dehydrogenase is down-regulated in colorectal cancer.
      ), and gastrointestinal (
      • Yan M.
      • Rerko R.M.
      • Platzer P.
      • Dawson D.
      • Willis J.
      • Tong M.
      • Lawrence E.
      • Lutterbaugh J.
      • Lu S.
      • Willson J.K.
      • Luo G.
      • Hensold J.
      • Tai H.H.
      • Wilson K.
      • Markowitz S.D.
      15-Hydroxyprostaglandin dehydrogenase, a COX-2 oncogene antagonist, is a TGF-β-induced suppressor of human gastrointestinal cancers.
      ,
      • Backlund M.G.
      • Mann J.R.
      • Dubois R.N.
      Mechanisms for the prevention of gastrointestinal cancer: the role of prostaglandin E2.
      ) cancers. Decreased expression of this protein, which has tumor suppressor activity in colorectal cancer (
      • Yan M.
      • Rerko R.M.
      • Platzer P.
      • Dawson D.
      • Willis J.
      • Tong M.
      • Lawrence E.
      • Lutterbaugh J.
      • Lu S.
      • Willson J.K.
      • Luo G.
      • Hensold J.
      • Tai H.H.
      • Wilson K.
      • Markowitz S.D.
      15-Hydroxyprostaglandin dehydrogenase, a COX-2 oncogene antagonist, is a TGF-β-induced suppressor of human gastrointestinal cancers.
      ,
      • Backlund M.G.
      • Mann J.R.
      • Dubois R.N.
      Mechanisms for the prevention of gastrointestinal cancer: the role of prostaglandin E2.
      ), is accompanied by increased expression of COX-2, a PG-synthesizing enzyme and a 15-PGDH physiological antagonist that has been implicated in tumor progression in various malignancies (
      • Kutchera W.
      • Jones D.A.
      • Matsunami N.
      • Groden J.
      • McIntyre T.M.
      • Zimmerman G.A.
      • White R.L.
      • Prescott S.M.
      Prostaglandin H synthase 2 is expressed abnormally in human colon cancer: evidence for a transcriptional effect.
      • Kirschenbaum A.
      • Liu X.
      • Yao S.
      • Levine A.C.
      The role of cyclooxygenase-2 in prostate cancer.
      • Huang M.
      • Stolina M.
      • Sharma S.
      • Mao J.T.
      • Zhu L.
      • Miller P.W.
      • Wollman J.
      • Herschman H.
      • Dubinett S.M.
      Non-small cell lung cancer cyclooxygenase-2-dependent regulation of cytokine balance in lymphocytes and macrophages: up-regulation of interleukin 10 and down-regulation of interleukin 12 production.
      • Wolff H.
      • Saukkonen K.
      • Anttila S.
      • Karjalainen A.
      • Vainio H.
      • Ristimaki A.
      Expression of cyclooxygenase-2 in human lung carcinoma.
      • Soslow R.A.
      • Dannenberg A.J.
      • Rush D.
      • Woerner B.M.
      • Khan K.N.
      • Masferrer J.
      • Koki A.T.
      COX-2 is expressed in human pulmonary, colonic, and mammary tumors.
      • Subbaramaiah K.
      • Norton L.
      • Gerald W.
      • Dannenberg A.J.
      Cyclooxygenase-2 is overexpressed in HER-2/neu-positive breast cancer: evidence for involvement of AP-1 and PEA3.
      ). Recently Muller-Decker et al. (
      • Muller-Decker K.
      • Berger I.
      • Ackermann K.
      • Ehemann V.
      • Zoubova S.
      • Aulmann S.
      • Pyerin W.
      • Furstenberger G.
      Cystic duct dilatations and proliferative epithelial lesions in mouse mammary glands upon keratin 5 promoter-driven overexpression of cyclooxygenase-2.
      ) reported that overexpression of COX-2 in keratin 5-positive cells of the mouse mammary gland caused an increase in PGE2, cystic duct dilations, adenosis, and fibrosis. Because stronger expression of COX-2 was observed in fibrocystic changes as compared with normal tissue, we used specific anti-COX-2 antibodies to validate the results in our collection of human breast tissue samples. Indeed IHC analysis of numerous macro- and microcysts showed that apocrine lesions (Fig. 18A; see also Fig. 17) but not type II flat microcysts (Fig. 18B) expressed this protein. The antibody decorated apocrine cell-cell contacts as well as the areas where these cells anchor to the basement membrane (Fig. 18A, arrow). IHC analysis of all the available samples showed that COX-2 was not expressed by stroma cells (Fig. 18B) and with very few exceptions by normal glands. Only in some cases was COX-2 expressed by non-apocrine tumors (results not shown), an observation that is in accordance with previously published data (
      • Soslow R.A.
      • Dannenberg A.J.
      • Rush D.
      • Woerner B.M.
      • Khan K.N.
      • Masferrer J.
      • Koki A.T.
      COX-2 is expressed in human pulmonary, colonic, and mammary tumors.
      ,
      • Spizzo G.
      • Gastl G.
      • Wolf D.
      • Gunsilius E.
      • Steurer M.
      • Fong D.
      • Amberger A.
      • Margreiter R.
      • Obrist P.
      Correlation of COX-2 and Ep-CAM overexpression in human invasive breast cancer and its impact on survival.
      ). Hence breast apocrine metaplastic epithelia co-express two key antagonistic activities that modulate the expression of PGE2, a compound that has been involved in control of cell proliferation, migration, apoptosis, angiogenesis, and inflammation (
      • Wang D.
      • Dubois R.N.
      Prostaglandins and cancer.
      ). The differential compartmentalization of COX-2 and 15-PGDH (compare Figs. 7A and 18A), however, suggests that their interplay with regard to PGE2 may be more complex than previously thought from the study of cell lines. In addition, our IHC studies of apocrine tumors failed to provide a clear correlation between the expression of the two proteins (Table III), suggesting that their interplay may be different in diverse pathological conditions.
      Figure thumbnail gr18
      Fig. 18IHC staining of apocrine (A) and flat type II microcysts (B) with COX-2 antibodies.
      HMG-CoA reductase is a key enzyme in cholesterol biosynthesis, and its product mevalonate has been shown to stimulate cell proliferation in breast cell lines (
      • Duncan R.E.
      • El-Sohemy A.
      • Archer M.C.
      Statins and cancer development.
      ,
      • Kusama T.
      • Mukai M.
      • Tatsuta M.
      • Matsumoto Y.
      • Nakamura H.
      • Inoue M.
      Selective inhibition of cancer cell invasion by a geranylgeranyltransferase-I inhibitor.
      ). Also there is evidence from in vivo and in vitro studies indicating that statins (
      • Kusama T.
      • Mukai M.
      • Tatsuta M.
      • Matsumoto Y.
      • Nakamura H.
      • Inoue M.
      Selective inhibition of cancer cell invasion by a geranylgeranyltransferase-I inhibitor.
      ,
      • Katz M.S.
      • Schapira L.
      • Harisinghani M.G.
      • Hughes K.S.
      Palpable right breast mass in a pregnant woman.
      ), which are inhibitors of HMG-CoA reductase, may be effective in the treatment of malignancies by decreasing farnesylation and geranylgeranylation of key proteins such as the Ras and Rho family of GTPases, which are essential for cellular proliferation (
      • Jakobisiak M.
      • Golab J.
      Potential antitumor effects of statins.
      ,
      • Fritz G.
      HMG-CoA reductase inhibitors (statins) as anticancer drugs.
      ). Our studies, however, showed abundant expression of both HMG-CoA reductase (Fig. 10A) and synthase (Fig. 2) in apocrine cyst cells that exhibit a very low proliferation index (Fig. 5D), implying that increased levels of these enzymes and perhaps increased levels of mevalonate may not be sufficient to stimulate cell growth. In fact, most of the invasive apocrine tumors analyzed were devoid of or showed very low levels of HMG-CoA reductase (Table III), a characteristic that may be distinctive to the progression of apocrine tumors.
      IHC analysis of the 93 breast tumor samples using antibodies against the two specific apocrine biomarkers revealed that only one of the lesions, an invasive apocrine carcinoma in patient 79 (Table I), expressed 15-PGDH in the tumor cells. These results harmonize well with the fact that our set of tumors, which included 75 ductal carcinomas, 15 lobular carcinomas, one mucinous carcinoma, one tubular lesion, and one apocrine carcinoma, reflects the expected incidence of mammary tumor types prevalent in the general population (
      • Mossler J.A.
      • Barton T.K.
      • Brinkhous A.D.
      • McCarty K.S.
      • Moylan J.A.
      • McCarty Jr., K.S.
      Apocrine differentiation in human mammary carcinoma.
      ,
      • Eusebi V.
      • Betts C.
      • Haagensen Jr., D.E.
      • Gugliotta P.
      • Bussolati G.
      • Azzopardi J.G.
      Apocrine differentiation in lobular carcinoma of the breast: a morphologic, immunologic, and ultrastructural study.
      ).
      Expression of 15-PGDH by invasive apocrine cancer was further confirmed by IHC using archival paraffin-embedded samples from six patients diagnosed with pure apocrine carcinoma (Table III) as well as by 2D PAGE (Fig. 15). Five of these lesions expressed 15-PGDH both in the invasive and apocrine CIS cells in each tumor, providing for the first time a direct association between the phenotype of benign metaplasias and the apocrine cancer phenotype (Fig. 17). Only one of the 15-PGDH-positive pure carcinomas expressed HMG-CoA reductase both in the tumor and CIS cells, whereas another expressed the protein only in the apocrine CIS cells (Table III). These results imply that the expression of HMG-CoA reductase is lost earlier than 15-PGDH during apocrine cancer progression. This contention is further supported by the fact that HMG-CoA reductase-positive tumors constitute a subset of the 15-PGDH apocrine tumors (Table III) and by the analysis of two apocrine atypias, a and b, detected in patient 71 (Fig. 16), whose phenotypes were remarkably similar, although one of them, most likely the most advanced lesion (atypia b), expressed significantly lower levels of HMG-CoA reductase.
      The IHC analysis of the pure apocrine carcinomas indicated that individual cancers may evolve through diverse molecular pathways because these lesions, CIS apocrine cells included, were either p53-positive or -negative and showed various combinations of phenotypes as far as 15-PGDH, HMG-CoA reductase, and COX-2 (Table III and Fig. 17) expression was concerned. Fig. 17 summarizes the protein expression phenotype of various lesions observed in this study that ranged from benign apocrine metaplasia to atypia, apocrine CIS, and invasive apocrine cancer. As a whole, our results indicate that most apocrine changes have little intrinsic malignant potential, although some lesions may progress to apocrine carcinoma. None of these lesions, however, seems to be a precursor of invasive ductal carcinomas, which accounted for 81% of the tumors analyzed in this study.
      So far, there is no molecular marker that can be used to detect a precancerous state or identify which premalignant lesion(s) will develop into invasive breast cancer, but further proteomic analysis of apocrine cancers and benign metaplasias are expected to reveal additional biomarkers that may dissect all the intermediate stages involved in apocrine cancer progression. These studies, which require prospective samples, are currently underway in our laboratory. A corollary of this work is that the successful identification of specific apocrine biomarkers, which allowed the subsequent tackling of more unclear and contentious points in the biology of apocrine cysts, provides proof-of-concept for the validity of the strategic approach we took. In addition, our sample preparation methodology based on cryosectioning enabled us to overcome some of the limitations imposed by the availability of biological material inherent to certain lesions. Encouraged by these facts we are now applying the same experimental approach to the study of various biological problems relevant to the biology of breast cancer, most notably to investigate the progression of ductal carcinoma in situ, a type of premalignant mammary lesions, to invasive cancer.
      Finally the identification of differentially expressed proteins that characterize specific steps in the progression from early benign lesions to apocrine cancer opens a window of opportunity for designing and testing new approaches for pharmacological intervention, not only in a therapeutic setting but also for chemoprevention, to inhibit cyst development. Enzymatic activities such as 15-PGDH, HMG-CoA reductase, and COX-2 are well known therapeutic targets (
      • Lam L.
      • Cao Y.
      Statins and their roles in cancer.
      ,
      • Wang D.
      • Dubois R.N.
      Prostaglandins and cancer.
      ,
      • Katz M.S.
      • Schapira L.
      • Harisinghani M.G.
      • Hughes K.S.
      Palpable right breast mass in a pregnant woman.
      ,
      • Hirsch F.R.
      • Lippman S.M.
      Advances in the biology of lung cancer chemoprevention.
      ,
      • Waters D.D.
      What do the statin trials tell us?.
      ) with pharmacological agents already available, and at least in the case of COX-2 one has reason to suspect a causal relationship to the development of fibrocystic changes (
      • Muller-Decker K.
      • Berger I.
      • Ackermann K.
      • Ehemann V.
      • Zoubova S.
      • Aulmann S.
      • Pyerin W.
      • Furstenberger G.
      Cystic duct dilatations and proliferative epithelial lesions in mouse mammary glands upon keratin 5 promoter-driven overexpression of cyclooxygenase-2.
      ). Statins, which are HMG-CoA reductase inhibitors, have well characterized efficacy and safety profiles (Ref.
      • Katz M.S.
      • Schapira L.
      • Harisinghani M.G.
      • Hughes K.S.
      Palpable right breast mass in a pregnant woman.
      and references therein), and it is therefore tempting to speculate that these drugs might be useful for preventing the development of breast cysts. Further studies into the functional role that the various biomarkers identified in this study play in cyst formation are warranted as they may identify additional therapeutic targets for fibrocystic changes and premalignant lesions.

      Acknowledgments

      We thank Kitt Christensen, Gitte Lindberg Stort, Dorrit Lützhøft, Hanne Nors, Michael Radich Johansen, Britt Olesen, Signe Trentemøller, and Dorthe Holm for expert technical assistance.

      REFERENCES

        • Guinebretière J.M.
        • Menet E.
        • Tardivon A.
        • Cherel P.
        • Vanel D.
        Normal and pathological breast, the histological basis.
        Eur. J. Radiol. 2005; 54: 6-14
        • Haagensen C.D.
        Diseases of the Breast. 3rd Ed. W. B. Saunders, Philadelphia1986: 250-266
        • Wellings S.R.
        • Alpers C.E.
        Apocrine cystic metaplasia: subgross pathology and prevalence in cancer-associated versus random autopsy breasts.
        Hum. Pathol. 1987; 18: 381-386
        • Rosai J.
        Ackerman’s Surgical Pathology. Mosby-Year Book, New York1996: 1565-1659
        • Dixon J.M.
        • Scott W.N.
        • Miller W.R.
        Natural history of cystic disease: the importance of cyst type.
        Br. J. Surg. 1985; 72: 190-192
        • Enriori C.L.
        • Novelli J.E.
        • Cremona M.D.C.
        • Hirsig R.J.P.
        • Enriori P.J.
        Biochemical study of cyst fluid in human breast cystic disease a review.
        Breast Cancer Res. Treat. 1992; 24: 1-9
        • Miller W.R.
        • Scott W.N.
        • Harris W.H.
        • Wang D.
        Using biological measurements, can patients with benign breast disease who are at high risk for breast cancer be identified?.
        Cancer Detect. Prev. 1992; 16: 99-106
        • Bruzzi P.
        • Dogliotti L.
        • Naldoni C.
        • Bucchi L.
        • Costantini M.
        • Cicognani A.
        • Torta M.
        • Buzzi G.F.
        • Angeli A.
        Cohort study of association of risk of breast cancer with cyst type in women with gross cystic disease of the breast.
        BMJ. 1997; 314: 925-928
        • Tsung J.S.
        • Wang T.Y.
        • Wang S.M.
        • Yang P.S.
        Cytological and biochemical studies of breast cyst fluid.
        Breast. 2005; 14: 37-41
        • Trenkic S.
        • Katic V.
        • Pashalina M.
        • Zivkovic V.
        • Milentijevic M.
        • Kostov M.
        The histologic spectrum of apocrine lesions of the breast.
        Arch. Oncol. 2004; 12: 61-65
        • Tavassoli F.A.
        • Purcell C.A.
        • Man Y.G.
        Down regulation of bcl2, ER, PR associated with AR expression is triggered by apocrine differentiation of mammary epithelium.
        Mod. Pathol. 1996; 9: 25A
        • Gatalica Z.
        Immunohistochemical analysis of apocrine breast lesions. Consistent over-expression of androgen receptor accompanied by the loss of estrogen and progesterone receptors in apocrine metaplasia and apocrine carcinoma in situ.
        Pathol. Res. Pract. 1997; 193: 753-758
        • Selim A.G.
        • Wells C.A.
        Immunohistochemical localisation of androgen receptor in apocrine metaplasia and apocrine adenosis of the breast: relation to oestrogen and progesterone receptors.
        J. Clin. Pathol. 1999; 52: 838-841
        • Scawn R.
        • Shousha S.
        Morphologic spectrum of estrogen receptor-negative breast carcinoma.
        Arch. Pathol. Lab. Med. 2002; 126: 325-330
        • Angeli A.
        • Bradlow H.L.
        • Bodian C.A.
        • Chasalow F.I.
        • Dogliotti L.
        • Haagensen Jr., D.E.
        Criteria for classifying breast cyst fluids.
        Ann. N. Y. Acad. Sci. 1990; 586: 49-52
        • Shoker B.S.
        • Jarvis C.
        • Sibson D.R.
        • Walker C.
        • Sloane J.P.
        Oestrogen receptor expression in the normal and pre-cancerous breast.
        J. Pathol. 1999; 188: 237-244
        • Tran D.D.
        • Lawson J.S.
        Microcysts and breast cancer: a study of biological markers in archival biopsy material.
        Breast Cancer Res. Treat. 2002; 75: 213-220
        • Oyama T.
        • Maluf H.
        • Koerner F.
        Atypical cystic lobules: an early stage in the formation of low-grade ductal carcinoma in situ.
        Virchows Arch. 1999; 435: 413-421
        • Bradlow H.L.
        • Rosenfeld R.S.
        • Kream J.
        • Fleisher M.
        • O’Connor J.
        • Schwartz M.K.
        Steroid hormone accumulation in human breast cyst fluid.
        Cancer Res. 1981; 41: 105-107
        • Vanluchene E.
        • Vandekerckhove D.
        • De Boever J.
        • Sandra P.
        Structures and concentrations of fifteen different steroid sulfates in human breast cyst fluids.
        J. Steroid Biochem. 1984; 21: 367-371
        • Belanger A.
        • Caron S.
        • Labrie F.
        • Naldoni C.
        • Dogliotti L.
        • Angeli A.
        Levels of eighteen non-conjugated and conjugated steroids in human breast cyst fluid: relationships with cyst type.
        Eur. J. Cancer. 1990; 26: 277-281
        • Nahoul K.
        • Kottler M.L.
        Relationships between androgen and estrogen sulfates in breast cyst fluid.
        Clin. Chim. Acta. 1992; 209: 179-187
        • Angeli A.
        • Dogliotti L.
        • Naldoni C.
        • Orlandi F.
        • Puligheddu B.
        • Caraci P.
        • Bucchi L.
        • Torta M.
        • Bruzzi P.
        Steroid biochemistry and categorization of breast cyst fluid: relation to breast cancer risk.
        J. Steroid Biochem. Mol. Biol. 1994; 49: 333-339
        • Tapper D.
        • Gajdusek C.
        • Moe R.
        • Ness J.
        Identification of a unique biological tumor marker in human breast cyst fluid and breast cancer tissue.
        Am. J. Surg. 1990; 159: 473-478
        • Reed M.J.
        • Coldham N.G.
        • Patel S.R.
        • Ghilchik M.W.
        • James V.H.
        Interleukin-1 and interleukin-6 in breast cyst fluid: their role in regulating aromatase activity in breast cancer cells.
        J. Endocrinol. 1992; 132: R5-R8
        • Ness J.C.
        • Sedghinasab M.
        • Moe R.E.
        • Tapper D.
        Identification of multiple proliferative growth factors in breast cyst fluid.
        Am. J. Surg. 1993; 166: 237-243
        • Lai L.C.
        • Kadory S.
        • Siraj A.K.
        • Lennard T.W.
        Oncostatin M, interleukin 2, interleukin 6 and interleukin 8 in breast cyst fluid.
        Int. J. Cancer. 1994; 59: 369-372
        • Lai L.C.
        • Siraj A.K.
        • Erbas H.
        • Lennard T.W.
        Transforming growth factor α, β1 and β2 in breast cyst fluid.
        Anticancer Res. 1994; 14: 2805-2809
        • Mannello F.
        • Bocchiotti G.
        • Bianchi G.
        • Marcheggiani F.
        • Gazzanelli G.
        Quantification of prostate-specific antigen immunoreactivity in human breast cyst fluids.
        Breast Cancer Res. Treat. 1996; 38: 247-252
        • Diamandis E.P.
        • Yu H.
        • Lopez-Otin C.
        Prostate specific antigen—a new constituent of breast cyst fluid.
        Breast Cancer Res. Treat. 1996; 38: 259-264
        • Lai L.C.
        • Erbas H.
        • Lennard T.W.
        • Peaston R.T.
        Prostate-specific antigen in breast cyst fluid: possible role of prostate-specific antigen in hormone-dependent breast cancer.
        Int. J. Cancer. 1996; 66: 743-746
        • Diamandis E.P.
        • Yousef G.M.
        • Soosaipillai A.R.
        • Grass L.
        • Porter A.
        • Little S.
        • Sotiropoulou G.
        Immunofluorometric assay of human kallikrein 6 (zyme/protease M/neurosin) and preliminary clinical applications.
        Clin. Biochem. 2000; 33: 369-375
        • Borchert G.H.
        • Melegos D.N.
        • Yu H.
        • Giai M.
        • Roagna R.
        • Ponzone R.
        • Sgro L.
        • Diamandis E.P.
        Quantification of pepsinogen C and prostaglandin D synthase in breast cyst fluid and their potential utility for cyst type classification.
        Clin. Biochem. 1999; 32: 39-44
        • Inaji H.
        • Koyama H.
        • Motomura K.
        • Noguchi S.
        • Tsuji N.
        • Kimura Y.
        • Sato H.
        • Sugano K.
        • Ohkura H.
        Simultaneous assay of ErbB-2 protein and carcinoembryonic antigen in cyst fluid as an aid in diagnosing cystic lesions of the breast.
        Breast Cancer. 1994; 1: 25-30
        • Bundred N.J.
        • Miller W.R.
        • Walker R.A.
        An immunohistochemical study of the tissue distribution of the breast cyst fluid protein, zinc α2 glycoprotein.
        Histopathology. 1987; 11: 603-610
        • Bundred N.J.
        • Scott W.N.
        • Davies S.J.
        • Miller W.R.
        • Mansel R.E.
        Zinc α-2 glycoprotein levels in serum and breast fluids: a potential marker of apocrine activity.
        Eur. J. Cancer. 1991; 27: 549-552
        • Balbin M.
        • Freije J.M.
        • Fueyo A.
        • Sanchez L.M.
        • Lopez-Otin C.
        Apolipoprotein D is the major protein component in cyst fluid from women with human breast gross cystic disease.
        Biochem. J. 1990; 271: 803-807
        • Sanchez L.M.
        • Diez-Itza I.
        • Vizoso F.
        • Lopez-Otin C.
        Cholesterol and apolipoprotein D in gross cystic disease of the breast.
        Clin. Chem. 1992; 38: 695-698
        • Mazoujian G.
        • Pinkus G.S.
        • Davis S.
        • Haagensen Jr., D.E.
        Immunohistochemistry of a gross cystic disease fluid protein (GCDFP-15) of the breast. A marker of apocrine epithelium and breast carcinomas with apocrine features.
        Am. J. Pathol. 1983; 110: 105-112
        • Sapino A.
        • Caisson P.
        • Bussolati G.
        Gross cystic disease fluid protein (GCDFP-15) in the breast: past and present.
        J. Biol. Regul. Homeost. Agents. 2000; 14: 259-262
        • Hayward J.L.
        • Parks A.G.
        Alterations in the microanatomy of the breast as a result of changes in the hormonal environment.
        in: Currie A.R. Endocrine Aspects of Breast Cancer. Livingstone, London1958
        • Tsuchiya S.
        Atypical ductal hyperplasia, atypical lobular hyperplasia and interpretation of a new borderline lesion.
        Jpn. J. Cancer Clin. 1998; 44: 548-555
        • Kusama R.
        • Fujimori M.
        • Matsuyama I.
        • Fu L.
        • Ishi K.
        • Hama Y.
        • Asanuma K.
        • Shingu K.
        • Kobayashi S.
        • Tsuchiya S.I.
        Clinicopathological characteristics of atypical cystic duct (ACD) of the breast: assessment of ACD as a precancerous lesion.
        Pathol. Int. 2000; 50: 793-800
        • Haagensen C.D.
        • Bodian C.
        • Haagensen Jr., D.E.
        Breast Carcinoma. Risk and Detection. W. B. Saunders, London1981
        • Bundred N.J.
        • West R.R.
        • Dowd J.O.
        • Mansel R.E.
        • Hughes L.E.
        Is there an increased risk of breast cancer in women who have had a breast cyst aspirated?.
        Br. J. Cancer. 1991; 64: 935-955
        • Devitt J.E.
        • To T.
        • Miller A.B.
        Risk of breast cancer in women with breast cysts.
        Can. Med. Assoc. J. 1992; 147: 45-49
        • Dixon J.M.
        • McDonald C.
        • Elton R.A.
        • Miller W.R.
        Risk of breast cancer in women with palpable breast cysts: a prospective study. Edinburgh Breast Group.
        Lancet. 1999; 353: 1742-1745
        • Dupont W.D.
        • Parl F.F.
        • Hartmann W.H.
        • Brinton L.A.
        • Winfield A.C.
        • Worrell J.A.
        • Schuyler P.A.
        • Plummer W.D.
        Breast cancer risk associated with proliferative breast disease and atypical hyperplasia.
        Cancer. 1993; 71: 1258-1265
        • Ahmed A.
        Calcification in human breast carcinomas: ultrastructural observations.
        J. Pathol. 1975; 117: 247-251
        • Yates A.J.
        • Ahmed A.
        Apocrine carcinoma and apocrine metaplasia.
        Histopathology. 1988; 13: 228-231
        • Page D.L.
        • Dupont W.D.
        • Jensen R.A.
        Papillary apocrine change of the breast: associations with atypical hyperplasia and risk of breast cancer.
        Cancer Epidemiol. Biomark. Prev. 1996; 5: 29-32
        • Viacava P.
        • Naccarato A.G.
        • Bevilacqua G.
        Apocrine epithelium of the breast: does it result from metaplasia?.
        Virchows Arch. 1997; 431: 205-209
        • Jones C.
        • Damiani S.
        • Wells D.
        • Chaggar R.
        • Lakhani S.R.
        • Eusebi V.
        Molecular cytogenetic comparison of apocrine hyperplasia and apocrine carcinoma of the breast.
        Am. J. Pathol. 2001; 158: 207-214
        • Durham J.R.
        • Fechner R.E.
        The histologic spectrum of apocrine lesions of the breast.
        Am. J. Clin. Pathol. 2000; 113: S3-S18
        • Celis J.E.
        • Gromov P.
        • Gromova I.
        • Moreira J. M
        • Cabezon T.
        • Ambartsumian N.
        • Grigorian M.
        • Lukanidin E.
        • Thor Straten P.
        • Guldberg P.
        • Bartkova J.
        • Bartek J.
        • Lukas J.
        • Lukas C.
        • Lykkesfeldt A.
        • Jaattela M.
        • Roepstorff P.
        • Bolund L.
        • Orntoft T.
        • Brunner N.
        • Overgaard J.
        • Sandelin K.
        • Blichert-Toft M.
        • Mouridsen H.
        • Rank F.E.
        Integrating proteomic and functional genomic technologies in discovery-driven translational breast cancer research.
        Mol. Cell. Proteomics. 2003; 2: 369-377
        • Celis J.E.
        • Moreira J.M.
        • Cabezon T.
        • Gromov P.
        • Friis E.
        • Rank F.
        • Gromova I.
        Identification of extracellular and intracellular signaling components of the mammary adipose tissue and its interstitial fluid in high risk breast cancer patients: toward dissecting the molecular circuitry of epithelial-adipocyte stromal cell interactions.
        Mol. Cell. Proteomics. 2005; 4: 4492-4522
        • Celis J.E.
        • Gromov P.
        • Cabezon T.
        • Moreira J.M.
        • Ambartsumian N.
        • Sandelin K.
        • Rank F.
        • Gromova I.
        Proteomic characterization of the interstitial fluid perfusing the breast tumor microenvironment: a novel resource for biomarker and therapeutic target discovery.
        Mol. Cell. Proteomics. 2004; 4: 327-344
        • Celis J.E.
        • Moreira J.M.
        • Gromova I.
        • Cabezon T.
        • Ralfkiaer U.
        • Guldberg P.
        • Straten P.T.
        • Mouridsen H.
        • Friis E.
        • Holm D.
        • Rank F.
        • Gromov P.
        Towards discovery-driven translational research in breast cancer.
        FEBS J. 2005; 272: 2-15
        • Lam L.
        • Cao Y.
        Statins and their roles in cancer.
        Drugs Today. 2005; 41: 329-334
        • O’Malley F.P.
        • Bane A.L.
        The spectrum of apocrine lesions of the breast.
        Adv. Anat. Pathol. 2004; 1: 1-9
        • Celis J.E.
        • Bravo R.
        • Larsen P.M.
        • Fey S.J.
        Cyclin (PCNA): a nuclear protein whose level correlates directly with the proliferative state of normal as well as transformed cells.
        Leuk. Res. 1984; 8: 143-157
        • Madsen P.
        • Rasmussen H.H.
        • Leffers H.
        • Honore B.
        • Dejgaard K.
        • Olsen E.
        • Kiil J.
        • Walbum E.
        • Andersen A.H.
        • Basse B.
        • Celis J.E.
        Molecular cloning, occurrence, and expression of a novel partially secreted protein “psoriasin” that is highly up-regulated in psoriatic skin.
        J. Investig. Dermatol. 1991; 97: 701-712
        • Leffers H.
        • Madsen P.
        • Rasmussen H.H.
        • Honore B.
        • Andersen A.H.
        • Walbum E.
        • Vandekerckhove J.
        • Celis J.E.
        Molecular cloning and expression of the transformation sensitive epithelial marker stratifin. A member of a protein family that has been involved in the protein kinase C signalling pathway.
        J. Mol. Biol. 1993; 231: 982-998
        • Celis J.E.
        • Ostergaard M.
        • Basse B.
        • Celis A.
        • Lauridsen J.B.
        • Ratz G.P.
        • Andersen I.
        • Hein B.
        • Wolf H.
        • Orntoft T.F.
        • Rasmussen H.H.
        Loss of adipocyte-type fatty acid binding protein and other protein biomarkers is associated with progression of human bladder transitional cell carcinomas.
        Cancer Res. 1996; 56: 4782-4790
        • Moll R.
        Cytokeratins as markers of differentiation in the diagnosis of epithelial tumors.
        Subcell. Biochem. 1998; 31: 205-262
        • Celis J.E.
        • Trentem⊘lle S.
        • Gromov P.
        Gel-based proteomics: high-resolution two-dimensional gel electrophoresis of proteins. Isoelectric focusing (IEF) and nonequilibrium pH gradient electrophoresis (NEPHGE).
        in: Celis J.E. Carter N. Hunter T. Shotton D. Simons K. Small J.V. Cell Biology. A Laboratory Handbook. 4. Elsevier, San Diego, CA2005: 165-177
        • O’Farrell P.H.
        High resolution two-dimensional electrophoresis of proteins.
        J. Biol. Chem. 1975; 250: 4007-4021
        • Gromova I.
        • Celis J. E
        Protein detection in gels by silver staining: a procedure compatible with mass spectrometry.
        in: Celis J.E. Carter N. Hunter T. Shotton D. Simons K. Small J.V. Cell Biology. A Laboratory Handbook. 4. Elsevier, San Diego, CA2005: 219-223
        • Shevchenko A.
        • Wilm M.
        • Vorm O.
        • Mann M.
        Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels.
        Anal. Chem. 1996; 68: 850-858
        • Huet C.
        Production of polyclonal antibodies in rabbits.
        in: Celis J.E. Cell Biology. A Laboratory Handbook. Elsevier, San Diego, CA1994: 257-268
        • Ostergaard M.
        • Wolf H.
        • Orntoft T.F.
        • Celis J.E.
        Psoriasin (S100A7): a putative urinary marker for the follow-up of patients with bladder squamous cell carcinomas.
        Electrophoresis. 1999; 20: 349-354
        • Gromov P.
        • Celis J. E
        High-resolution two-dimensional gel electrophoresis and protein identification using western blotting and ECL detection.
        EXS (Basel). 2000; 88: 55-67
        • Gouon-Evans V.
        • Lin E.Y.
        • Pollard J.W.
        Requirement of macrophages and eosinophils and their cytokines/chemokines for mammary gland development.
        Breast Cancer Res. 2002; 4: 155-164
        • Lai L.C.
        • Kadory S.
        • Cornell C.
        • Lennard T.W.
        Possible regulation of soluble ICAM-1 levels by interleukin-1 in a sub-set of breast cysts.
        Int. J. Cancer. 1993; 55: 586-589
        • Lai L.C.
        • Siraj A.K.
        • Erbas H.
        • Lennard T.W.
        Relationship between basic fibroblast growth factor and transforming growth factor-β2 in breast cyst fluid.
        J. Clin. Endocrinol. Metab. 1995; 80: 711-714
        • Lai L.C.
        • Erbas H.
        • Meadows K.A.
        • Lennard T.W.
        • Holly J.M.
        Insulin-like growth factor binding protein-3 in breast cyst fluid: relationships with insulin-like growth factors I and II and transforming growth factor-β 1 and 2.
        Cancer Lett. 1996; 110: 207-212
        • Moriya T.
        • Sakamoto K.
        • Sasano H.
        • Kawanaka M.
        • Sonoo H.
        • Manabe T.
        • Ito J.
        Immunohistochemical analysis of Ki-67, p53, p21, and p27 in benign and malignant apocrine lesions of the breast: its correlation to histologic findings in 43 cases.
        Mod. Pathol. 2000; 13: 13-18
        • Olivier L.M.
        • Krisans S.K.
        Peroxisomal protein targeting and identification of peroxisomal targeting signals in cholesterol biosynthetic enzymes.
        Biochim. Biophys. Acta. 2000; 1529: 89-102
        • Morikawa S.
        • Murakami T.
        • Yamazaki H.
        • Izumi A.
        • Saito Y.
        • Hamakubo T.
        • Kodama T.
        Analysis of the global RNA expression profiles of skeletal muscle cells treated with statins.
        J. Atheroscler. Thromb. 2005; 12: 121-131
        • de Nigris F.
        • Lerman L.O.
        • Napoli C.
        New insights in the transcriptional activity and coregulator molecules in the arterial wall.
        Int. J. Cardiol. 2002; 86: 153-168
        • Farmer P.
        • Bonnefoi H.
        • Becette V.
        • Tubiana-Hulin M.
        • Fumoleau P.
        • Larsimont D.
        • Macgrogan G.
        • Bergh J.
        • Cameron D.
        • Goldstein D.
        • Duss S.
        • Nicoulaz A.L.
        • Brisken C.
        • Fiche M.
        • Delorenz I.M.
        • Iggo R.
        Identification of molecular apocrine breast tumours by microarray analysis.
        Oncogene. 2005; 24: 4660-4671
        • Celis J.E.
        • Celis P.
        • Palsdottir H.
        • Ostergaard M.
        • Gromov P.
        • Primdahl H.
        • Orntoft T.F.
        • Wolf H.
        • Celis A.
        • Gromova I.
        Proteomic strategies to reveal tumor heterogeneity among urothelial papillomas.
        Mol. Cell. Proteomics. 2002; 1: 269-279
        • Bratthauer G.L.
        • Lininger R.A.
        • Man Y.G.
        • Tavassoli F.A.
        Androgen and estrogen receptor mRNA status in apocrine carcinomas.
        Diagn. Mol. Pathol. 2002; 11: 113-118
        • Honma N.
        • Takubo K.
        • Akiyama F.
        • Sawabe M.
        • Arai T.
        • Younes M.
        • Kasumi F.
        • Sakamoto G.
        Expression of GCDFP-15 and AR decreases in larger or node-positive apocrine carcinomas of the breast.
        Histopathology. 2005; 47: 195-201
        • Emberley E.D.
        • Alowami S.
        • Snell L.
        • Murphy L.C.
        • Watson P.H.
        S100A7 (psoriasin) expression is associated with aggressive features and alteration of Jab1 in ductal carcinoma in situ of the breast.
        Breast Cancer Res. 2004; 6: 308-315
        • Carlsson H.
        • Yhr M.
        • Petersson S.
        • Collins N.
        • Polyak K.
        • Enerback C.
        Psoriasin (S100A7) and Calgranulin-B (S100A9) induction is dependent on reactive oxygen species and is downregulated by Bcl-2 and antioxidants.
        Cancer Biol. Ther. 2005; 4: 998-1005
        • O’Malley F.P.
        • Page D.L.
        • Nelson E.H.
        • Dupont W.D.
        Ductal carcinoma in situ of the breast with apocrine cytology: definition of a borderline category.
        Hum. Pathol. 1994; 25: 164-168
        • Leal C.
        • Henrique R.
        • Monteiro P.
        • Lopes C.
        • Bento M.J.
        • De Sousa C.P.
        • Lopes P.
        • Olson S.
        • Silva M.D.
        • Page D.L.
        Apocrine ductal carcinoma in situ of the breast: histologic classification and expression of biologic markers.
        Hum. Pathol. 2001; 32: 487-493
        • Selim A.G.
        • Ryan A.
        • El-Ayat G.
        • Wells C.A.
        Loss of heterozygosity and allelic imbalance in apocrine metaplasia of the breast: microdissection microsatellite analysis.
        J. Pathol. 2002; 196: 287-291
        • Losi L.
        • Lorenzini R.
        • Eusebi V.
        • Bussolati G.
        Apocrine differentiation in invasive carcinoma of the breast.
        Appl. Immunohistochem. 1995; 3: 91-98
        • Backlund M.G.
        • Mann J.R.
        • Holla V.R.
        • Buchanan F.G.
        • Tai H.H.
        • Musiek E.S.
        • Milne G.L.
        • Katkuri S.
        • DuBois R.N.
        15-Hydroxyprostaglandin dehydrogenase is down-regulated in colorectal cancer.
        J. Biol. Chem. 2005; 280: 3217-3223
        • Yan M.
        • Rerko R.M.
        • Platzer P.
        • Dawson D.
        • Willis J.
        • Tong M.
        • Lawrence E.
        • Lutterbaugh J.
        • Lu S.
        • Willson J.K.
        • Luo G.
        • Hensold J.
        • Tai H.H.
        • Wilson K.
        • Markowitz S.D.
        15-Hydroxyprostaglandin dehydrogenase, a COX-2 oncogene antagonist, is a TGF-β-induced suppressor of human gastrointestinal cancers.
        Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 17468-17473
        • Backlund M.G.
        • Mann J.R.
        • Dubois R.N.
        Mechanisms for the prevention of gastrointestinal cancer: the role of prostaglandin E2.
        Oncology. 2005; 69: 28-32
        • Kutchera W.
        • Jones D.A.
        • Matsunami N.
        • Groden J.
        • McIntyre T.M.
        • Zimmerman G.A.
        • White R.L.
        • Prescott S.M.
        Prostaglandin H synthase 2 is expressed abnormally in human colon cancer: evidence for a transcriptional effect.
        Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 4816-4820
        • Kirschenbaum A.
        • Liu X.
        • Yao S.
        • Levine A.C.
        The role of cyclooxygenase-2 in prostate cancer.
        Urology. 2001; 58: 127-131
        • Huang M.
        • Stolina M.
        • Sharma S.
        • Mao J.T.
        • Zhu L.
        • Miller P.W.
        • Wollman J.
        • Herschman H.
        • Dubinett S.M.
        Non-small cell lung cancer cyclooxygenase-2-dependent regulation of cytokine balance in lymphocytes and macrophages: up-regulation of interleukin 10 and down-regulation of interleukin 12 production.
        Cancer Res. 1998; 58: 1208-1216
        • Wolff H.
        • Saukkonen K.
        • Anttila S.
        • Karjalainen A.
        • Vainio H.
        • Ristimaki A.
        Expression of cyclooxygenase-2 in human lung carcinoma.
        Cancer Res. 1998; 58: 4997-5001
        • Soslow R.A.
        • Dannenberg A.J.
        • Rush D.
        • Woerner B.M.
        • Khan K.N.
        • Masferrer J.
        • Koki A.T.
        COX-2 is expressed in human pulmonary, colonic, and mammary tumors.
        Cancer. 2000; 89: 2637-2645
        • Subbaramaiah K.
        • Norton L.
        • Gerald W.
        • Dannenberg A.J.
        Cyclooxygenase-2 is overexpressed in HER-2/neu-positive breast cancer: evidence for involvement of AP-1 and PEA3.
        J. Biol. Chem. 2002; 277: 18649-18657
        • Muller-Decker K.
        • Berger I.
        • Ackermann K.
        • Ehemann V.
        • Zoubova S.
        • Aulmann S.
        • Pyerin W.
        • Furstenberger G.
        Cystic duct dilatations and proliferative epithelial lesions in mouse mammary glands upon keratin 5 promoter-driven overexpression of cyclooxygenase-2.
        Am. J. Pathol. 2005; 166: 575-584
        • Elston C.W.
        • Ellis I.O.
        Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up.
        Histopathology. 1991; 19: 403-410
        • Spizzo G.
        • Gastl G.
        • Wolf D.
        • Gunsilius E.
        • Steurer M.
        • Fong D.
        • Amberger A.
        • Margreiter R.
        • Obrist P.
        Correlation of COX-2 and Ep-CAM overexpression in human invasive breast cancer and its impact on survival.
        Br. J. Cancer. 2003; 88: 574-578
        • Wang D.
        • Dubois R.N.
        Prostaglandins and cancer.
        Gut. 2006; 55: 115-122
        • Duncan R.E.
        • El-Sohemy A.
        • Archer M.C.
        Statins and cancer development.
        Cancer Epidemiol. Biomark. Prev. 2005; 14: 1897-1898
        • Kusama T.
        • Mukai M.
        • Tatsuta M.
        • Matsumoto Y.
        • Nakamura H.
        • Inoue M.
        Selective inhibition of cancer cell invasion by a geranylgeranyltransferase-I inhibitor.
        Clin. Exp. Metastasis. 2003; 20: 561-567
        • Katz M.S.
        • Schapira L.
        • Harisinghani M.G.
        • Hughes K.S.
        Palpable right breast mass in a pregnant woman.
        Nat. Clin. Pract. Oncol. 2005; 2: 218-221
        • Jakobisiak M.
        • Golab J.
        Potential antitumor effects of statins.
        Int. J. Oncol. 2003; 23: 1055-1069
        • Fritz G.
        HMG-CoA reductase inhibitors (statins) as anticancer drugs.
        Int. J. Oncol. 2005; 27: 1401-1409
        • Mossler J.A.
        • Barton T.K.
        • Brinkhous A.D.
        • McCarty K.S.
        • Moylan J.A.
        • McCarty Jr., K.S.
        Apocrine differentiation in human mammary carcinoma.
        Cancer. 1980; 46: 2463-2471
        • Eusebi V.
        • Betts C.
        • Haagensen Jr., D.E.
        • Gugliotta P.
        • Bussolati G.
        • Azzopardi J.G.
        Apocrine differentiation in lobular carcinoma of the breast: a morphologic, immunologic, and ultrastructural study.
        Hum. Pathol. 1984; 15: 134-140
        • Hirsch F.R.
        • Lippman S.M.
        Advances in the biology of lung cancer chemoprevention.
        J. Clin. Oncol. 2005; 23: 3186-3197
        • Waters D.D.
        What do the statin trials tell us?.
        Clin. Cardiol. 2001; 24: 3-7