GSTPi-positive tumour microenvironment-associated fibroblasts are significantly associated with GSTPi-negative cancer cells in paired cases of primary invasive breast cancer and axillary lymph node metastases.

BACKGROUND: Glutathione S-transferase Pi (GSTPi) expression is one of the factors, which is known to be associated with development of resistance to chemotherapeutics in cancer patients, including those with breast cancer. Yet, its expression has been reported to be undetectable in cancer cells in high percent of patients with primary breast cancer. However, GSTPi expression in stromal cells in breast tumour microenvironment, namely cancer-associated fibroblast (CAF), which is recognised to have major roles in cancer progression, remains poorly reported. METHODS: The aim of the study was to determine the expression of GSTPi; vimetin, a fibroblast-associated cytoskeleton protein; and α-smooth muscle actin (α-SMA), a known marker of CAF in breast cancer tissue, by immunohistochemical staining method in consecutive histologic sections of formalin-fixed and paraffin-embedded tissue biopsy specimens from a cohort of 39 paired cases of patients with invasive breast cancer and the corresponding axillary lymph nodes metastases. RESULTS: Ductal and acinar luminal epithelial cells, myoepithelial cells and surrounding fibroblasts exhibited a homogeneous cytoplasmic reactivity with anti-GSTPi antibody in 11 of 11 cases of benign breast tissue biopsies. The vimentin-positive fibroblasts were unreactive with anti-α-SMA antibody. Loss of GSTPi expression was observed in breast cancer cells, at both the primary and metastatic sites, in 31 of 39 paired cases, as compared with benign breast epithelial cells (Fisher's exact test P<0.001). A significant association was observed between GSTPi-positive, vimentin-positive and α-SMA-positive fibroblast in tumour microenvironment at both sites. CONCLUSION: This is an original report of demonstration of a significance association between tumour microenvironment-associated GSTPi-positive CAF (vimentin/α-SMA-positive) and the GSTPi-negative cancer cells in paired cases of primary invasive breast cancer and the corresponding axillary lymph nodes metastases.

Tissue engineered tumor models.

Many research programs use well-characterized tumor cell lines as tumor models for in vitro studies. Because tumor cells grown as three-dimensional (3-D) structures have been shown to behave more like tumors in vivo than do cells growing in monolayer culture, a growing number of investigators now use tumor cell spheroids as models. Single cell type spheroids, however, do not model the stromal-epithelial interactions that have an important role in controlling tumor growth and development in vivo. We describe here a method for generating, reproducibly, more realistic 3-D tumor models that contain both stromal and malignant epithelial cells with an architecture that closely resembles that of tumor microlesions in vivo. Because they are so tissue-like we refer to them as tumor histoids. They can be generated reproducibly in substantial quantities. The bioreactor developed to generate histoid constructs is described and illustrated. It accommodates disposable culture chambers that have filled volumes of either 10 or 64 ml, each culture yielding on the order of 100 or 600 histoid particles, respectively. Each particle is a few tenths of a millimeter in diameter. Examples of histological sections of tumor histoids representing cancers of breast, prostate, colon, pancreas and urinary bladder are presented. Potential applications of tumor histoids include, but are not limited to, use as surrogate tumors for pre-screening anti-solid tumor pharmaceutical agents, as reference specimens for immunostaining in the surgical pathology laboratory and use in studies of invasive properties of cells or other aspects of tumor development and progression. Histoids containing nonmalignant cells also may have potential as "seeds" in tissue engineering. For drug testing, histoids probably will have to meet certain criteria of size and tumor cell content. Using a COPAS Plus flow cytometer, histoids containing fluorescent tumor cells were analyzed successfully and sorted using such criteria.

Ionizing radiation-induced instant pairing of heterochromatin of homologous chromosomes in human cells.

Using fluorescence in situ hybridization with human band-specific DNA probes we examined the effect of ionizing radiation on the intra-nuclear localization of the heterochromatic region 9q12-->q13 and the euchromatic region 8p11.2 of similar sized chromosomes 9 and 8 respectively in confluent (G1) primary human fibroblasts. Microscopic analysis of the interphase nuclei revealed colocalization of the homologous heterochromatic regions from chromosome 9 in a proportion of cells directly after exposure to 4 Gy X-rays. The percentage of cells with paired chromosomes 9 gradually decreased to control levels during a period of one hour. No significant changes in localization were observed for chromosome 8. Using 2-D image analysis, radial and inter-homologue distances were measured for both chromosome bands. In unexposed cells, a random distribution of the chromosomes over the interphase nucleus was found. Directly after irradiation, the average inter-homologue distance decreased for chromosome 9 without alterations in radial distribution. The percentage of cells with inter-homologue distance <3 micro m increased from 11% in control cells to 25% in irradiated cells. In contrast, irradiation did not result in significant changes in the inter-homologue distance for chromosome 8. Colocalization of the heterochromatic regions of homologous chromosomes 9 was not observed in cells irradiated on ice. This observation, together with the time dependency of the colocalization, suggests an underlying active cellular process. The biological relevance of the observed homologous pairing remains unclear. It might be related to a homology dependent repair process of ionizing radiation induced DNA damage that is specific for heterochromatin. However, also other more general cellular responses to radiation-induced stress or change in chromatin organization might be responsible for the observed pairing of heterochromatic regions.

LEA.135 expression: its comparison with other prognostic biomarkers for patients with primary breast carcinoma.

The purpose of this retrospective study was to examine the prognostic value of expression of luminal epithelial antigen (LEA.135) for recurrence and overall survival of patients with primary invasive breast carcinoma by both univariate and multivariate analyses. The possible prognostic value of LEA.135 was also compared with some widely utilized prognostic biomarkers such as c-erbB 2, topoisomerase II.alpha (TPII.alpha), MIB 1, estrogen receptor (ER) and progesterone receptor (PR), as well as age of the patients and clinicopathologic parameters. The study was carried out by immunohistochemical methods on formalin-fixed/paraffin-embedded tissue sections in a series of 225 patients with median follow-up of 8.5 years. Prognostic significance of the biomarkers was determined by two-sided p value. In this series of patients, among the age and clinicopathologic parameters, only age, was significantly associated with a decreased overall survival (logrank p = 0.027). Among the prognostic biomarkers, TPII a expression at high (> 50% positive cells) or moderate (6-50% positive cells) level was associated with an increased rate of recurrence (logrank p < 0.001). However, the association of TPII.alpha expression with a decreased overall survival failed to reach a statistically significance. Expression of c-erbB 2 showed a trend of being associated with an increased probability of recurrence, but the association did not reach statistical significance. The remaining biomarkers were not associated with either the probability of recurrence or overall survival. LEA.135 expression was observed in 163 (72.4%) of the 225 patients. The patients with high (> 50% positive cells) or moderate (6-50% positive cells) level of LEA.135-positive cancer cells showed a significantly decreased probability of recurrence (logrank p < 0.001) and an increased overall survival (logrank p < 0.001) compared with those with LEA.135-negative cancer cells. The association remained significant by multivariate analysis for recurrence (likelihood ratio test p < 0.001) and overall survival (likelihood ratio test p < 0.001) when assessed with other prognostic parameters. Furthermore, the combination of LEA.135 with other prognostic biomarkers stratified four subgroups of patients with distinct clinical outcome. The subgroup of patients who were LEA.135+/TPII.alpha- showed the lowest probability of recurrence and the longest overall survival compared with those who were LEA.135-/TPII.alpha+ (logrank p < 0.001). Interestingly, the patients whose cancer cells were LEA.135+/TPII.alpha+, LEA.135+ MIB.1+ or LEA.135+/c-erbB 2+ experienced a decreased probability of recurrence and an increased overall survival compared with those with LEA.135-/TPII.alpha+, LEA.135- MIB.1+ or LEA.135-/c-erbB 2+ (logrank p < 0.001). The results demonstrated that LEA.135 is an independent and favorable prognostic biomarker for patients with primary invasive breast carcinoma, that the loss of LEA.135 expression is associated with aggressive phenotype of cancer cells during the breast cancer progression, and that its continued expression seems to override the adverse effects of expression of an oncogene or cell proliferation-associated molecules.

LEA.135 expression: an independent and favorable prognostic biomarker for patients with primary invasive breast cancer.

The prognostic significance of LEA.135 expression, detected by immunohistochemistry in formalin-fixed and paraffin-embedded tissue sections, was evaluated and compared with the widely utilized clinicopathological parameters for patients with primary invasive breast carcinomas. Pathological parameters such as tumor size, histological tumor type, histological grade, nuclear grade, lymph node (LN) status, bone marrow (BM) status, as well as age of patient at initial diagnosis together with follow-up in years were available for this group of patients (n = 178). Among these parameters, tumor size, histological tumor type, histological grade, LN status, and BM status were individually and significantly associated with increased probability of recurrence by univariate analysis. By multivariate analysis, however, only tumor size, LN status, and BM status remained statistically significant. LEA.135-positive patients showed a statistically significant probability of not recurring (77 +/- 5% at 5 years after surgery) compared with patients who were LEA. 135-negative (49 +/- 6% at 5 years after surgery) (log-rank p < 0. 001). Furthermore, the association remained statistically significant by multivariate analysis (log-rank p = 0.019), demonstrating that LEA.135 expression independently and significantly identified breast cancer patients with favorable clinical outcome. In addition, there was a statistically significant association between loss of LEA.135 expression and poor prognosis when patients were stratified by pathological parameters. Furthermore, a subgroup of patients who were LEA. 135-positive/LN-negative experienced a decreased rate of recurrence compared with those who were LEA.135-negative/LN-negative (16% vs. 27%, respectively). A similar result was also obtained when BM-negative patients were stratified on the basis of LEA. 135-positive or LEA.135-negative subgroups for recurrence (18% vs. 43%, respectively). Most interestingly, the patients whose cancer cells were LEA.135-positive/LN-positive experienced a much lower rate of recurrence than those whose cells were LEA. 135-negative/LN-positive (29% vs. 57%, respectively). The results clearly demonstrate that LEA.135 expression was a significantly independent and favorable prognostic marker for patients with primary invasive breast carcinoma by both univariate and multivariate analyses.

Analysis of telomerase activity in ovarian cystadenomas, low-malignant-potential tumors, and invasive carcinomas.

OBJECTIVE: Inappropriate telomerase expression has been reported to be associated with the development and/or progression of malignancies. Therefore, the purpose of the study was to determine and evaluate the levels of telomerase activity in normal ovary, cystadenomas, low-malignant-potential tumors, and carcinomas of the ovary. METHODS: In the present study, telomerase activity was examined in frozen tissue specimens of normal ovary (n = 6), ovarian cystadenomas (n = 13), ovarian low-malignant-potential (LMP) tumors (n = 12), and ovarian invasive carcinomas (n = 81). Clinicopathological information including age at diagnosis, histological grade, FIGO stage, presence of distant metastasis at diagnosis, and residual disease was available for all patients with ovarian carcinomas (n = 81). Telomerase activity was assessed by the telomeric repeat amplification protocol (TRAP). Arbitrary values were assigned to processivity of telomerase activities based on the molecular weights of the telomeric repeat DNA ladders, and were graded as "negative," "moderate" (< or =99 bp), or "high" (>100 bp) activities. The specificity of telomerase activity was determined by the pretreatment of telomerase-positive control or tumor samples with RNase that led to the abolition of the activity. In addition, to determine the possibility of false negativity due to the presence of telomerase inhibitors, TRAP assay was performed on each of the telomerase-negative specimens by mixing them individually with the telomerase-positive control. RESULTS: Telomerase activity in the progression of ovarian carcinogenesis was evaluated. In comparison with normal ovary/cystadenoma (32%), a much higher frequency of the moderate activity was observed in LMP tumors (67%) or invasive carcinomas (57%), suggesting a close association between the latter two categories. The results reflect a subpopulation of telomerase-positive LMP tumor cells with the potential to develop invasive carcinomas. None of the specimens of the benign or LMP tumors exhibited high activity. In contrast, 18% of ovarian invasive carcinomas showed high telomerase activity (P = 0.013, Fisher exact test) and further 57%, moderate activity (75% in all). A statistically significant difference was observed in the expression of telomerase activity between normal ovary/benign cystadenomas and ovarian invasive carcinomas (P = 0.001, chi(2) test). CONCLUSIONS: The study showed a high prevalence of telomerase activity in LMP tumors or invasive carcinomas, the high levels of telomerase activity being associated exclusively with the invasive ovarian carcinomas. Therefore, the levels of processivity of telomerase activity and evidence of its statistically significant association with ovarian carcinoma suggest its role in the progression of ovarian carcinogenesis.

Telomerase activity: comparison between fine-needle aspiration and biopsy specimens for the detection of tumor cells.

BACKGROUND: The purpose of the current study was to determine telomerase activity as a sensitive biomarker for the detection of malignant cells in fine-needle aspiration (FNA) specimens. METHODS: FNA specimens with parallel samples of fresh tumor tissue were obtained from surgical specimens after surgical excision. Using a polymerase chain reaction-based telomeric repeat amplification protocol (TRAP) assay, telomerase activity was determined systematically in FNA specimens (n = 21) and corresponding available tissue biopsy specimens (n = 16) containing malignant cells. In addition to a case of myelolipoma, normal counterparts for 3 of 16 cancer cases, including both biopsy and FNA specimens, also were available for the determination of telomerase activity. RESULTS: Telomerase activity was observed in 14 of 16 of the FNA specimens (88%) and 15 of 16 of the corresponding biopsy specimens (94%). Telomerase activity was detected in both the biopsy specimen and the corresponding FNA specimen, with one exception (a case of mucinous adenocarcinoma of the cecum). In contrast, specimens from three normal tissue biopsies and FNA specimens of normal tissue adjacent to the malignant lesions, as well as the myelolipoma, exhibited no telomerase activity. It is interesting to note that both tissue biopsy specimens and FNA specimens from a patient with high grade sarcoma were negative for telomerase activity. The examination of hematoxylin and eosin-stained adjacent tissue biopsy sections or FNA smears revealed similar low populations of lymphocytes, including those cases that were negative for telomerase activity. There was agreement in the detection of telomerase activity between tissue biopsies and their corresponding FNA specimens in 15 of the 16 patients, indicating a 94% concordance rate (95% confidence interval, 70%, 98%). CONCLUSIONS: The results of the current study clearly suggest that the telomerase activity in FNA specimens was comparable to that of their corresponding biopsy specimens, and that this activity was associated with the presence of malignant cells. The TRAP assay has potential for use in the detection of malignant cells in FNA specimens, particularly cases in which the cytology is not characteristically malignant and/or is present in insufficient numbers. Cancer (Cancer Cytopathol)

Purification and characterization of a Hodgkin's and B cell-associated glycoprotein.

A glycoprotein (BLA.36), expressed on the plasma membrane of Hodgkin's cells and also on normal and malignant B lymphocytes and histiocytes, was identified by reaction with a monoclonal antibody. BLA.36 was not detectable on other hematopoietic, carcinoma or melanoma cell lines. BLA.36 was purified to homogeneity by extracting proteins from a Hodgkin's cell line (HDLM-3), followed by immunoaffinity chromatography, utilizing immobilized anti-BLA.36 antibody, and gel filtration on Sephacryl S-100 in the presence of protein dissociating agents. The purified component yielded a single band on sodium dodecyl sulphate/polyacrylamide gel electrophoresis under both non-reducing and reducing conditions, and closely related three isotypes of similar molecular weight and with the apparent isoelectric points that ranged from 5.0 to 5.2 on two-dimensional gel electrophoresis. The purified BLA.36 reacted with the original specific antibody, on both one- or two-dimensional gel electrophoresis, suggesting that antigenic determinant was not adversely affected during purification procedure. Competitive immunoprecipitation analyses and the determination of N-terminus sequence of the first 13 amino acid residues suggest that BLA.36 is unrelated to other known Hodgkin's or hematopoietic cell antigens. Finally, significance of BLA.36 expression on the growth of BLA.36-positive cell lines was studied. Blocking of BLA.36 with anti-BLA.36 antibody led to the in vitro growth-inhibition of BLA.36-positive cell lines. The antibody pre-absorbed with the purified BLA.36 was unable to exert growth-inhibition, demonstrating the specificity of reaction. In addition, the treatment of the BLA.36-positive cell lines with differentiation-inducing agent, alpha-interferon (alpha-IFN), down-regulated BLA.36 expression and also showed in vitro growth-inhibition of BLA.36-positive cell lines. Taken together, these results suggest a growth-related function of BLA.36.

Telomeric DNA: marker for human prostate cancer development?

BACKGROUND: Telomeres that protect chromosomes at both ends are shortened with each somatic cell division through replication-dependent sequence loss at DNA termini. The chromosomes with shortened telomeres tend to become unstable, leading to cell death. Due largely to reactivation/upregulation of telomerase, a ribonucleoprotein that adds nucleotide sequences onto chromosome ends, cancer cells become immortal and neoplastically transformed. METHODS: The purpose of the present study was to study three newly established human prostate cancer cell lines and three prostate-derived fibroblastic cell cultures at different passages for telomeric DNA signal intensity, telomeric restriction fragment length (TRFL), telomerase activity, and spontaneous apoptotic index. RESULTS: Compared with the three fibroblastic cell cultures, the three new prostate cancer cell lines showed: 1) telomerase activity, 2) stronger telomeric signals, 3) relatively longer TRFLs, and 4) much lower apoptotic indices. On the other hand, three fibroblastic cell cultures showed: 1) no telomerase activity, 2) weaker telomeric signals, 3) shorter TRFLs (fibroblasts derived from surrounding tissue of prostate tumor showed intermediate TRFLs), and 4) comparatively higher apoptotic indices. CONCLUSIONS: Based on these results, we conclude that telomeric DNA signal intensity, TRFL, and telomerase activity can be used to distinguish prostate cancer cells from adjacent fibroblasts.

Systematic determination of telomerase activity and telomerase length during the progression of human breast cancer in cell culture models.

The purpose of the study was to determine systematically the expression of telomerase activity and the length of telomere repeat arrays by utilizing two different cell culture models that derive from normal individual donors, and probably represent various stages of human breast oncogenesis in cell culture. The models consist of mortal, non-tumorigenic immortal and tumorigenic immortal human mammary epithelial cell (MEC) lines. Using a recently developed polymerase chain reaction (PCR)-based telomeric repeat amplification protocol (TRAP) assay, telomerase activity was undetectable in mortal MEC cells. In contrast, the immortal MEC that were nontumorigenic or tumorigenic in immunosuppressed athymic mice, showed telomerase activity. The absence of telomerase activity in mortal and its presence in both non-tumorigenic and tumorigenic immortal cell lines did not reflect their proliferative rate, as demonstrated by the similar pattern and intensity of reactivity of these cell lines with anti-Ki 67 antibody which recognizes a human nuclear cell proliferation--associated antigen. Southern blot analyses of Hinf I-digested genomic DNA hybridized with a (TTAGGG)4 probe revealed arrays of telomeric repeat lengths ranging from 3 to 5, 3.5 to 9, 3.2 to 9 or 3 to 15 kilobase pair (kbp) for mortal, nontumorigenic immortal, and tumorigenic immortal or established MEC lines respectively. These results suggest that telomerase activity and stable telomeric repeat lengths may be a molecular phenotype of the early stages in the progression of breast cancer.

A sialoglycoprotein (LEA.135) associated with favourable prognosis of patients with lymph node-negative primary breast carcinoma.

A cell-surface sialoglycoprotein (LEA.135) was identified using a monoclonal antibody that was generated by immunization of Balb/c mice with extracts of normal breast tissue following prior immune-tolerization with mammary carcinoma cell lines. LEA.135 is distinct from other known epithelial cell-associated antigens, including the family of mucins or keratins and epidermal growth factor receptor. Using immunohistochemical staining methods, LEA.135 expression was detected predominantly on the apical plasma membrane of normal and neoplastic mammary and extramammary epithelial cells in freshly frozen or formalin-fixed paraffin-embedded tissue sections. A retrospective study of 111 cases of lymph node-negative patients (TanyN0M0) with primary infiltrating ductal breast carcinoma, with a median follow-up of 7.9 years, was conducted. A comparison of overall survival (O.S.) was made of patients whose tumor cells exhibited reactivity with anti-LEA.135 antibody (O.S. 92.9 +/- 3.3% at 8 years), compared with those whose specimens showed the absence of LEA.135 expression (O.S. 68.3 +/- 10.8% at 8 years). A statistically significant univariant association between LEA.135 expression and O.S. was observed (logrank p < 0.001). In addition, in a subgroup of patients with histologically moderately differentiated tumors (N = 71), LEA.135-positive cases showed an improved O.S. (90.8 +/- 4.6% at 8 years; p < 0.001) compared with those who were LEA.135-negative (O.S. 55.6 +/- 13.6% at 8 years). The association remained statistically significant in a multivariable analysis after adjusting for histological grade, tumor size and age (p < 0.02). Thus, in this series of patients with lymph node-negative primary breast carcinoma, LEA.135 expression was associated with a significant decrease in the rate of recurrence and with an increase in overall survival, independent of tumor size, histologic grade, and patient's age. In contrast to the majority of other prognostic markers which predicts a worse biology, LEA.135 is a unique class of antigen whose expression indicates a lower aggressiveness of the tumor cells.

Evidence for the release of autocrine growth factor(s) by Hodgkin's cell lines.

A cell surface glycoprotein (BLA.36) expressed on Hodgkin's cell and also on normal and malignant B lymphocytes, was identified by reaction with a monoclonal antibody. Expression of BLA.36 was detected on Hodgkin's cell lines, as well as B or pre-B cell lines, but was not detectable on other hematopoietic, carcinoma or melanoma cell lines, both by immunoprecipitation of intrinsically labeled cell lines and by immunocytochemical staining methods. Hodgkin's cell lines which were positive for BLA.36 expression were also responsive to a partially purified Hodgkin's cell growth factor(s) (HCGFs). HCGFs was partially purified by concentrating conditioned medium from a Hodgkin's cell line (HDLM-3), followed by ion-exchange chromatography (DE52) and gel filtration on Sephadex G-50. The active fractions yielded protein bands with apparent molecular weights ranging from 16 to 21 kilodaltons on sodium dodecyl sulphate-polyacrylamide gel electrophoresis under both non-reducing or reducing conditions. Hodgkin's cell lines treated with antiBLA.36 antibody were not growth responsive to exogenously added HCGFs. Addition of alpha-interferon (alpha-IFN) to BLA.36-positive Hodgkin's cell lines resulted in down-regulation of BLA.36, accompanied by inhibition of growth. Cell lines treated in this way failed to respond to exogenously added HCGFs, suggesting a requirement for BLA.36 expression in order for HCGFs to exercise its growth-stimulatory effect.

Use of pH 9.5 Tris-HCl buffer containing 5% urea for antigen retrieval immunohistochemistry.

Successful antigen retrieval (AR) immunohistochemistry is dependent on the temperature, heating time, and pH value of the AR solutions. There is no single standardized AR solution, however, that is suitable for all antibodies "routinely" used in surgical pathology for immunostaining archival tissue sections. We tested a variety of AR solutions varying in pH value, chemical composition, and molarity. Based upon preliminary results, we compared three AR solutions: 0.1 M Tris-HCl buffer, pH 9.5, containing 5% urea, 0.1 M Tris-HCl buffer pH 9.5 without urea, and citrate buffer, pH 6.0. Each AR solution was tested with a panel of 34 antibodies using microwave heating for antigen retrieval. The heating conditions were standardized at 10 min and an automated stainer was used to standardize the immunostaining method. The Tris-HCl containing urea was superior to pH 6.0 citrate buffer for 22 antibodies. In 12 cases, Tris-HCl with urea was also superior to Tris-HCl alone. In 12 cases, the intensity was similar for all three retrieval solutions. The staining obtained with Tris-HCl with urea was equal to or better than with pH 6.0 citrate buffer in all cases. The Tris-HCl with urea solution is satisfactory for AR of most antibodies employed in routine surgical pathology.

Comparison of two microwave based antigen-retrieval solutions in unmasking epitopes in formalin-fixed tissue for immunostaining.

The present study compared two microwave based antigen-retrieval solutions in their ability to unmask antigenic determinants in formalin-fixed and paraffin-embedded tissues for Immunostaining. In this regard, two widely used antigen-retrieval solutions, namely 0.05 M glycine-HCl buffer, pH 3.6, containing 0.01% (w/v) (EDTA) and 0.1 M sodium citrate buffer, pH 6.0, were evaluated for (1) their effectiveness in unmasking a wide range of antigenic determinants (2) their ability to yield reproducible results (3) the lack of deleterious effects in any antibody antigen systems of interest. Both of these antigen-retrieval solutions resulted in greatly improved immunostaining following microwave-heating of dewaxed tissue sections for 2 x 5 min. Glycine-HCl buffer solution resulted in stronger immunostaining with antibodies to nuclear antigens [androgen receptor (AR), estrogen receptor (ER), progesterone receptor (PR), p53, proliferating cell nuclear antigen (PCNA), Ki-67 and MIB-1], cytoplasmic antigens (actin and factor-VIII) and cell-surface antigens [Cu-18, epithelial membrane antigen (EMA) and MT-1 (CD43)], whereas sodium citrate buffer yielded superior immunostaining with antibodies to vimentin, and some cell-surface antigens [common leukocyte antigen (CLA) (CD45) and UCHL-1 (CD45RO)]. The effect of unmasking the epitopes recognized by antibody to PCNA was equally effective with either of the antigen-retrieval solutions. Antibodies to pan-keratin, prostatic acid phosphatase (PAP), B lymphocyte antigen (BLA.36, CD20CY) and L26 (CD20) exhibited no enhancement in the intensity of staining with either of the antigen-retrieval solutions.

Antigen retrieval immunohistochemistry under the influence of pH using monoclonal antibodies.

Antigen retrieval (AR) incorporating high-temperature microwave (MW) heating of tissue sections before immunostaining is a revolutionary technique that can unmask the antigens in formalin-fixed tissue sections, thus making them available for immunohistochemical staining. Although high temperature is believed to be the primary mechanism in retrieval of antigens, a variety of chemical solutions have been tested to define an optimal AR solution. We tested the hypothesis that pH of the AR solution may influence the quality of immunostaining by using seven different AR buffer solutions at a series of different pH values ranging from 1 to 10. We evaluated the staining of monoclonal antibodies to cytoplasmic antigens (AE1, HMB45, NSE), nuclear antigens (MIB-1, PCNA, ER), and cell surface antigens (MT1, L26, EMA) on routinely formalin-fixed, paraffin-embedded sections under different pH conditions with MW heating for 10 min. The intensity of immunostaining was graded in a blinded fashion. The pH value of the AR buffer solution was carefully measured before, immediately after, and 15 min after the AR procedure. The influence of pH on AR immunohistochemical staining can be summarized into three patterns. Some antigens (L26, PCNA, AE1, EMA, and NSE) showed excellent retrieval throughout the pH range. Other antigens (MIB1 and ER) showed strong intensity of immunohistochemical staining at very low pH and at neutral to high pH, but a dramatic decrease in the intensity of the AR immunostaining at moderately acidic pH (pH 3-6). Still others (MT1 and HMB45) showed increasing intensity of the AR immunostaining with increasing pH, but only weak immunostaining at low pH. Among the seven buffer solutions at any given pH value, the intensity of AR immunostaining was very similar. However, Tris-HCl buffer tended to produce better results at higher pH, compared with other buffers. Although high-temperature heating is believed to be the most important factor for the AR technique, the pH value of the AR solution is an important co-factor for some antigens. Optimization of the AR system should therefore include optimization of the pH of the AR solution. Our results indicate that AR immunostaining of Tris-HCl or sodium acetate buffer at pH 8-9 may be suitable for most antigens, although certain nuclear antigens show optimal staining at low pH.

Generation of a murine monoclonal antibody to normal mammary epithelium using mice rendered immune-tolerant to malignant mammary epithelium.

A monoclonal antibody (MAb) that distinguishes normal from malignant mammary epithelia in tissue or cell lines was generated using a procedure that involved immune-tolerization before immunization. Immune-tolerance to two transformed mammary epithelial cell lines (MCF.7 and MDA.MB.231 cell lines combined) was induced in neonatal mice within 24 hr of birth. Successful induction of immune-tolerance was determined by an indirect immunohistological method, testing sera from mice against the tolerogen (i.e., the MCF.7 and MDA.MB.231 cell lines). Mice lacking antibodies in their sera against the immune-tolerogen were subsequently immunized with an extract of normal breast epithelium. One mouse was selected for hybridoma production based on evidence of serum antibody that showed reactivity with normal mammary epithelial cells (MEC) but not with invasive breast carcinoma cells, as determined by an indirect immunohistological method. Spleen cells from the selected mouse were fused with a mouse myeloma cell line to generate MAb. After extensive screening, one MAb was further studied on the basis of reactivity with normal MEC in tissue and absence of staining of malignant MEC in tissue or tumorigenic MEC lines. This test of specificity of reactivity revealed that the antigen detected by the specific antibody was expressed on the apical plasma membrane of normal glandular epithelia that included breast, cervix, colon, lung, pancreas, and stomach, but not on their malignant counterparts in tissue sections. The antigen recognized by the MAb was termed luminal epithelial antigen with an apparent MW of 92 KD (LEA.92). This study illustrates the practical usefulness of the immune-tolerization/immunization approach in the generation of antibodies with particular specificity requirements, as in the identification of an antigen that is differentially expressed in two tissues (e.g., normal and malignant) which otherwise have a multiplicity of antigens in common.

Identification of a cell-surface glycoprotein associated with normal mammary and extramammary epithelial cells.

The goal of the study was to identify any normal genes that may become inactivated in malignant cells, with associated modifications or loss of gene products. Consequently, attempts were made to identify such products by generating monoclonal antibodies using an immune tolerisation-immunisation procedure. Using such a technique, a plasma membrane-associated glycoprotein with an apparent molecular weight of 92 kDa was identified. The glycoprotein was termed luminal epithelial antigen (LEA.92). The pattern of expression of LEA.92 was demonstrated by an indirect immunostaining technique. Using an in vitro model system representing various stages of breast oncogenesis, LEA.92 was detected on normal or immortalised mammary epithelial cell (MEC) lines which were dependent on epidermal growth factor (EGF) and anchorage formation for growth and non-tumorigenic in nude mice. In contrast, LEA.92 was undetectable on oncogenically transformed or established lines of mammary carcinoma cell lines which were independent of EGF or anchorage formation for growth and were highly tumorigenic. The results appear to suggest a correlation between the down-regulation of LEA.92 and the development of tumorigenicity in malignant MEC lines. Furthermore, the patterns of expression of LEA.92 on breast cells in tissue mirrored those of breast epithelial cells in cell cultures. LEA.92 was detected on the surface of normal but not malignant epithelial cells, which included breast, cervix, colon, lung, pancreas and stomach. LEA.92 appeared to be distinct from receptor for epidermal growth factor, antigens associated with milk fat globule membrane and the family of epithelium-specific keratins.

Strategies for improving the immunohistochemical staining of various intranuclear prognostic markers in formalin-paraffin sections: androgen receptor, estrogen receptor, progesterone receptor, p53 protein, proliferating cell nuclear antigen, and Ki-67 antigen revealed by antigen retrieval techniques.

Different variations of the antigen retrieval technique using different retrieval solutions have been evaluated for their effectiveness in restoring the antigenicity of six intranuclear antigens, each of which is a potentially valuable prognostic indicator in formalin-fixed, paraffin-embedded tissue sections. The results of immunohistochemical staining for estrogen receptor, progesterone receptor, androgen receptor, p53 protein, proliferating cell nuclear antigen, and Ki-67 antigen were compared following the different antigen retrieval approaches. The strongest immunostaining signal with the clearest background was obtained by microwave heating of dewaxed paraffin sections for 10 minutes in 0.05 mol/L glycine HCl (pH 3.5) or in citrate buffer solution (pH 6). Urea solution, distilled water, and lead thiocyanate solution yielded improvements with some antigens, but less consistently and less impressively than glycine HCl buffer or citrate buffer. Following antigen retrieval nuclear staining was sharply defined and could be achieved consistently in a variety of tissues after formalin fixation for as long as 7 days. The duration of fixation, however, was an important variable; generally, the longer the fixation time the more vigorous the retrieval procedure required. This study demonstrates the ability to stain a variety of intranuclear antigens, which are not readily demonstrable otherwise, in formalin-paraffin sections with a high degree of consistency and reproducibility. The availability of methods that are effective in paraffin sections may facilitate studies of the possible value of these markers as prognostic indicators for predicting the response of major tumors to different forms of therapy. This study also provided insight into the basic principles of the antigen retrieval method, which may be helpful in attempts to develop a more uniformly standardized technique applicable to many different antigen systems.

Development of tumorigenicity and rearrangement of chromosome 1 correlates with down-regulation of cell-surface glycoproteins in human mammary carcinoma cell line.

In the current study, attempts were made to identify any products of normal breast cell genes, that may become inactivated in the malignant counterparts. Using an immune-tolerization/immunization procedure of generating antibody, two different cell-surface glycoproteins termed luminal epithelial antigen, LEA.92 and LEA.135 were identified. LEA.92 and LEA.135 expressions on MEC in a culture model-system, that reflect various steps of neoplastic transformation were detected on normal or immortalized MEC lines that were non-tumorigenic in nude mice. Furthermore, no rearrangement of chromosome 1 was observed in those cells. In contrast, both glycoproteins were undetectable on oncogenically transformed or established lines of mammary carcinoma cells that were tumorigenic. LEA.92 or LEA.135 negative cell lines exhibited a partial deletion of their chromosome 1. In tissue sections, LEA.92 expression was detected on the apical plasma membrane of normal and hyperplastic but not on the malignant mammary or extramammary epithelial cells (MEC). However, unlike LEA.92, LEA.135 was detected on certain cases of primary breast carcinoma cells, irrespective of morphological differentiation, in tissue sections.

Identification of a cell-surface antigen (LEA.135) associated with favorable prognosis in human breast cancer.

The present study was undertaken with a rationale that loss of certain "normal tissue" antigens might have prognostic significance, reflecting inactivation of the corresponding genes during neoplastic progression. An attempt was made to identify such antigens by means of generating monoclonal antibodies using a tolerization/immunization procedure. A monoclonal antibody generated by immunization of BALB/c mice with normal breast tissue extract, following prior tolerization with mammary carcinoma cells, recognized a cell-surface glycoprotein, luminal epithelial antigen, with an apparent molecular weight of 135,000 (LEA.135). The pattern of expression on LEA.135 was determined by immunohistochemical-staining techniques on frozen and formalin-fixed and paraffin-embedded tissue sections. LEA.135 was demonstrable on the apical plasma membrane of normal and nonneoplastic epithelial cells in breast and other tissues. Studies have shown that LEA.135 is distinct from receptors for epidermal growth factor and from known antigens associated with epithelial cells, including the family of keratins. In a retrospective study, with a follow-up ranging from 5 to 15 years, patients whose breast tumor cells expressed LEA.135 had a superior overall survival rate (78 0.139% at > 5 years; P = 0.025). Furthermore, in patients with histologically poorly differentiated tumors, LEA.135-positive cases had a better prognosis (80 0.179% at > 5 years; P = 0.013) compared with LEA.135-negative cases. In addition, in patients with aneuploid tumors, LEA.135-positive cases again showed an improved survival (90 0.001% at > 5 years; P = 0.039) compared with those that were with LEA.135 negative. The results suggest that the expression of LEA.135 provides a useful indication of clinical outcome in patients with breast carcinomas.