Rocco Fraioli

Expression of Endoplasmic Reticulum Aminopeptidases in EBV-B Cell Lines from Healthy Donors and in Leukemia/Lymphoma, Carcinoma, and Melanoma Cell Lines

Peptide trimming in the endoplasmic reticulum (ER), the final step required for the generation of most HLA class I-binding peptides, implicates the concerted action of two aminopeptidases, ERAP1 and ERAP2. Because defects in the expression of these peptidases could lead to aberrant surface HLA class I expression in tumor cells, we quantitatively assayed 14 EBV-B cell lines and 35 human tumor cell lines of various lineages for: 1) expression and enzymatic activities of ERAP1 and ERAP2; 2) ER peptide-trimming activity in microsomes; 3) expression of HLA class I H chains and TAP1; and 4) surface HLA class I expression. ERAP1 and ERAP2 expression was detectable in all of the EBV-B and tumor cell lines, but in the latter it was extremely variable, sometimes barely detectable, and not coordinated. The expression of the two aminopeptidases corresponded well to the respective enzymatic activities in most cell lines. A peptide-trimming assay in microsomes revealed additional enzymatic activities, presumably contributed by other unidentified aminopeptidases sharing substrate specificity with ERAP2. Interestingly, surface HLA class I expression showed significant correlation with ERAP1 activity, but not with the activity of either ERAP2 or other unidentified aminopeptidases. Transfection with ERAP1 or ERAP2 of two tumor cell lines selected for simultaneous low expression of the two aminopeptidases resulted in the expected, moderate increases of class I surface expression. Thus, low and/or imbalanced expression of ERAP1 and probably ERAP2 may cause improper Ag processing and favor tumor escape from the immune surveillance.

Altered expression of endoplasmic reticulum aminopeptidases ERAP1 and ERAP2 in transformed non-lymphoid human tissues

The endoplasmic reticulum (ER) aminopeptidases ERAP1 and ERAP2 contribute to generate HLA class I binding peptides. Recently, we have shown that the expression of these enzymes is high and coordinated (with each other and with HLA class I molecules) in immortalized B cells, but variable and imbalanced in human tumour cell lines of various non‐lymphoid lineages. Herein, this issue was investigated in vivo by testing ERAP1 and ERAP2 expression in normal non‐lymphoid tissues and their malignant counterparts. ERAP1 and ERAP2 were detected exclusively in the epithelial cells of over half of the tested normal tissues. Four ERAP1/ERAP2 phenotypes (+/+, −/−, +/− and −/+) were detected, and the presence of either or both enzymes was not necessarily associated with HLA class I expression. In more than 160 neoplastic lesions, the expression of either or both aminopeptidases was retained, lost (most frequently, particularly ERAP1) or acquired as compared to the normal counterparts, depending on the tumour histotype. The double‐negative (−/−) phenotype was the most frequent, and significantly (P = 0.013) associated with a lack of detectable HLA class I antigens. In selected neoplastic lesions, ERAP1 and ERAP2 were also tested for their enzymatic (peptide‐trimming) activities. Expression and function were found to correlate, indicating that immunohistochemistry detects active enzymes in vivo. Thus, dissociation in the expression of ERAP1, ERAP2 and HLA class I may already be present in some normal tissues, but malignant transformation causes additional losses, gains and imbalances in specific tumour histotypes, and these alter the peptide‐trimming ability of tumour cells in vivo. J. Cell. Physiol. 216: 742–749, 2008,

Expression of the c‐Kit receptor and its ligand SCF in non‐small‐cell lung carcinomas

Increasing experimental evidence indicates that stem‐cell factor (SCF) and its cognate receptor c‐Kit may participate in the growth control of various solid human malignancies. In the present study, we have extended this analysis to non‐small‐cell lung carcinomas (NSCLC). The results of an immunohistochemical analysis demonstrated that c‐Kit/SCF are expressed by 30%/58% of adenocarcinomas, 15%/37% of squamous‐cell carcinomas and by 40%/30% of undifferentiated carcinomas respectively. In 15% of primary and 18% of metastatic tumors, co‐expression of the receptor and its ligand was documented. Western‐blot assays of tumor extracts demonstrated that both molecules exhibit features of the normal receptor and ligand. Since biologically active SCF is physiologically present in the bloodstream, our data indicate that SCF is available to c‐kit‐expressing NSCLC cells via autocrine, paracrine or endocrine mechanisms. Thus activation of c‐Kit in these tumors may contribute critically to their progression. Int. J. Cancer 75:171–175, 1998.

Antibody–drug conjugates: targeting melanoma with cisplatin encapsulated in protein-cage nanoparticles based on human ferritin

A novel antibody-drug conjugate (ADC) was synthesized incorporating ferritin-based nanoparticles. An average of three molecules of monoclonal antibody (mAb) Ep1 to the human melanoma-specific antigen CSPG4 were conjugated to a single ferritin cage encapsulating about 50 cisplatin molecules (HFt-Pt-Ep1). The HFt-Pt-Ep1 nanoparticle had an estimated molecular size of about 900 kD and 33 nm, and flow cytometry demonstrated specific binding to a CSPG4(+) melanoma cell line, but not to a CSPG4(-) breast carcinoma cell line. As compared to the cisplatin-containing ferritin nanoparticle alone (HFt-Pt), which inhibited thymidine incorporation more efficiently in breast carcinoma than melanoma cells, the mAb-derivatized HFt-Pt-Ep1 nanoparticle had a 25-fold preference for the latter. A similar preference for melanoma was observed upon systemic intravenous administration of HFt-Pt-Ep1 to nude mice xenotransplanted with pre-established, palpable melanoma and breast carcinoma tumors. Thus, we have been able to determine precise combinations and stoichiometric relationships between mAbs and nanoparticle protein cages, whereby the latter lose their tropism for ubiquitously distributed cellular receptors, and acquire instead remarkably lineage-selective binding. HFt-Pt-Ep1 is therefore an interesting model to improve the therapeutic index of antiblastic therapy in a tumor such as melanoma, which at its advanced stages is totally refractory to mono- and combination-chemotherapy.

Artichoke polyphenols induce apoptosis and decrease the invasive potential of the human breast cancer cell line MDA‐MB231

The human breast cancer cell line, estrogen receptor negative, MDA‐MB231, was used to evaluate the antitumor effect of polyphenolic extracts from the edible part of artichokes (AEs). Treatment of cancer cells reduced cell viability and inhibited cell growth in a dose‐dependent manner. Importantly, AEs did not have any effect on normal breast epithelial cell line, MCF10A. Chlorogenic acid (ChA), the most representative component of the polyphenolic fraction of artichoke, had no prominent effects on the cell death rate of MDA‐MB231 cells. The addition of AEs to the cells, rather than ChA, triggered apoptosis via a mitochondrial and a death‐receptor pathway, as shown by the activation of caspase‐9 and caspase‐8, respectively. Furthermore, an increase of the Bax:Bcl2 ratio and up‐regulation of cyclin‐dependent kinase inhibitor, p21WAF1, crucial apoptotic players, were documented. According to our data on activation of caspase‐9, a loss of mitochondrial transmembrane potential (Ψm) was shown. Cell motility and invasion capabilities were remarkably inhibited by AEs‐treatment in highly invasive MDA‐MB231 cells. In addition, a significant decrease of proteolytic activity of metalloproteinase‐2 protein (MMP‐2), involved in degrading components of the extracellular matrix, was detected. Our findings indicate that AEs reduced cell viability, inhibited cell growth, triggered apoptotic mechanisms, and showed inhibitory properties against the invasive behavior of MDA‐MB231 cancer cell line. Altogether, these data indicate the potential chemopreventive activity of artichoke polyphenolic extracts. J. Cell. Physiol. 227: 3301–3309, 2012.

Impaired Assembly Results in the Accumulation of Multiple HLA-C Heavy Chain Folding Intermediates

Class I MHC H chains assemble with β2-microglobulin (β2m) and are loaded with peptide Ags through multiple folding steps. When free of β2m, human H chains react with Abs to linear epitopes, such as L31. Immunodepletion and coimmunoprecipitation experiments, performed in this study, detected a preferential association of L31-reactive, β2m-free H chains with calnexin in β2m-defective cells, and with calreticulin and TAP in β2m-expressing cells. In β2m-defective cells, the accumulation of calnexin-bound H chains stoichiometrically exceeded their overall accumulation, a finding that supports both chaperoning preferences and distinct sorting abilities for different class I folds. No peptide species, in a mass range compatible with that of the classical class I ligands, could be detected by mass spectrometry of acidic eluates from L31-reactive HLA-Cw1 H chains. In vitro assembly experiments in TAP-defective T2 cells, and in cells expressing an intact Ag-processing machinery, demonstrated that L31 H chains are not only free of, but also unreceptive to, peptides. L31 and HC10, which bind nearly adjacent linear epitopes of the α1 domain α helix, reciprocally immunodepleted free HLA-C H chains, indicating the existence of a local un-/mis-folding involving the N-terminal end of the α1 domain α helix and peptide-anchoring residues of the class I H chain. Thus, unlike certain murine free H chains, L31-reactive H chains are not the immediate precursors of conformed class I molecules. A model inferring their precursor-product relationships with other known class I intermediates is presented.

HLA-A, -B, -C genotyping and expression in human nonlymphoid tumor cell lines.

A combination of molecular genotyping and protein biochemistry methods was used to assess the HLA-A, -B, -C genotyping and expression of six tumor cell lines. Four cell lines had been previously HLA typed by conventional serologic methods. Two could not be typed by serology because deficient in the surface expression of HLA-A, -B, -C molecules. As shown herein, all the 25 alleles carried by the six tested cell lines were typed at the DNA level. In addition, discrepancies between the previous serologic and the present DNA typing results were detected in 9 of the 21 tested serologic specificities. Typing at the protein level by isoelectric focusing and allele-specific monoclonal antibodies confirmed the DNA typing data. Our results exemplify the limits of the serologic typing procedures and demonstrate that molecular methods are highly desirable to conduct functional experiments and identify HLA losses in neoplastic cells at single allele level.

Monoclonal antibodies to HLA‐E bind epitopes carried by unfolded β2m‐free heavy chains

Since HLA‐E heavy chains accumulate free of their light β2‐microglobulin (β2m) subunit, raising mAbs to folded HLA‐E heterodimers has been difficult, and mAb characterization has been controversial. Herein, mAb W6/32 and 5 HLA‐E‐restricted mAbs (MEM‐E/02, MEM‐E/07, MEM‐E/08, DT9, and 3D12) were tested on denatured, acid‐treated, and natively folded (both β2m‐associated and β2m‐free) HLA‐E molecules. Four distinct conformations were detected, including unusual, partially folded (and yet β2m‐free) heavy chains reactive with mAb DT9. In contrast with previous studies, epitope mapping and substitution scan on thousands of overlapping peptides printed on microchips revealed that mAbs MEM‐E/02, MEM‐E/07, and MEM‐E/08 bind three distinct α1 and α2 domain epitopes. All three epitopes are linear since they span just 4–6 residues and are “hidden” in folded HLA‐E heterodimers. They contain at least one HLA‐E‐specific residue that cannot be replaced by single substitutions with polymorphic HLA‐A, HLA‐B, HLA‐C, HLA‐F, and HLA‐G residues. Finally, also the MEM‐E/02 and 3D12 epitopes are spatially distinct. In summary, HLA‐E‐specific residues are dominantly immunogenic, but only when heavy chains are locally unfolded. Consequently, the available mAbs fail to selectively bind conformed HLA‐E heterodimers, and HLA‐E expression may have been inaccurately assessed in some previous oncology, reproductive immunology, virology, and transplantation studies.