Primary cell cultures from human renal cortex and renal-cell carcinoma evidence a differential expression of two spliced isoforms of Annexin A3

Abstract

Primary cell cultures from renal cell carcinoma (RCC) and normal renal cortex tissue of 60 patients have been established, with high efficiency (more than 70%) and reproducibility, and extensively characterized. These cultures composed of more than 90% of normal or tumor tubular cells have been instrumental for molecular characterization of Annexin A3 (AnxA3), never extensively studied before in RCC cells although AnxA3 has a prognostic relevance in some cancer and it has been suggested to be involved in the hypoxia-inducible factor-1 pathway. Western blot analysis of 20 matched cortex/RCC culture lysates showed two AnxA3 protein bands of 36 and 33 kDa, and two-dimensional Western blot evidenced several specific protein spots. In RCC cultures the 36-kDa isoform was significantly down-regulated and the 33-kDa isoform up-regulated. Furthermore, the inversion of the quantitative expression pattern of two AnxA3 isoforms in tumor cultures correlate with hypoxia-inducible factor-1alpha expression. The total AnxA3 protein is down-regulated in RCC cultures as confirmed also in tissues by tissue microarray. Two AnxA3 transcripts that differ for alternative splicing of exon III have been also detected. Real-time PCR quantification in 19 matched cortex/RCC cultures confirms the down-regulation of longer isoform in RCC cells. The characteristic expression pattern of AnxA3 in normal and tumor renal cells, documented in our primary cultures, may open new insight in RCC management.

Figure 1

Phenotypic characterization of normal cortex and RCC primary cultures. A: Representative micrographs of contrast phase morphology (×40 magnification) and of immunofluorescence staining (×400 magnification) of indicated markers in normal cortex (top panels) and RCC primary cultures (bottom panels). The coexpression of CD13 (in green) and calbindin (in red) have also been evaluated. Representative FACS analysis (middle panels) for tested markers is also reported. B: Evaluation of nonepithelial cells in normal cortex and RCC primary cultures by immunofluorescence (top and bottom panels) and FACS analysis (middle panels). DAPI counterstains the nuclei in blue. Percentage of positivity, evaluated by FACS analysis, is reported as mean ± SE of several experiments. In the case of CD14/CD11b the mean percentage of coexpression is reported.

Figure 2

Western blot analysis of AnxA3 and HIF-1α protein pattern in cortex and RCC primary cultures. A: Cell lysates from matched primary cultures established from patients n. 43 and n. 53 run on 10% SDS-PAGE probed with anti-AnxA3 antibodies (left) and run on 4 to 12% gradient gel probed with anti-AnxA3 and anti-HIF-1α antibodies (right). B: AnxA3 and HIF-1α protein pattern in cell lysates of 18 matched primary cultures established from corresponding patients and analyzed by 4% to 12% gradient gel. Anti–β actin immunoblotting assessed loading. C: Densitometric analysis of total AnxA3 bands, normalized for the corresponding β-actin band of 20 cortex, 20 RCC, 13 RCC HIF-1α–positive, 7 RCC HIF-1α negative primary culture lysates. D: Densitometric analysis of the 36-kDa and 33-kDa AnxA3 bands, normalized for the corresponding β-actin band, in the same lysates evaluated in C. Means ± SE. All cultures are from RCCcc patients (Pt) except n. 36, n. 41 RCCpap, and n. 14, n. 57 RCCmixed. C indicates cortex culture; T, RCC culture. *P = 0.01; **P < 0.01; NS indicates not significant respect to the matched normal cortex samples (Student t test).

Figure 3

A: Immunodetection of the AnxA3 isoforms on the 2-DE Western blot (2D-WB) of matched primary cell culture lysates of patients n. 53 and n. 43 and corresponding pattern of isoforms obtained by 1-DE Western blot (1D-WB), after separation on 4% to 12% gradient gel. B: Sypro-stained membrane obtained after 2-DE separation on 4% to 12% gel of a pool of protein lysates from different normal cortex primary cultures. In the enlargement the immunoreactivity of the AnxA3 isoforms on the same membrane is shown. The isoelectric point (pI) and molecular weight markers are reported; the arrows show the specific spot that after gel-excision and MS analysis has been identified as AnxA3 protein.

Figure 4

Western blot analysis of AnxA3 in cortex and RCC tissues. A: AnxA3 protein pattern in cell lysates of matched normal and tumor tissues obtained from five patients, after separation on 4% to 12% gradient gel. Anti–β actin immunoblotting assessed loading. B: Densitometric analysis of 36-kDa and 33-kDa AnxA3 bands, normalized for corresponding β-actin band. Means ± SE of the five matched tissues analyzed. All patients (Pt) are RCCcc except n. 36 RCCpap. C indicates cortex tissue; T, RCC tissue. *P = 0.01, Student t test.

Figure 5

Selected images of tissue microarray analysis performed on histopathological samples of 51 RCC and 33 matched non-neoplastic kidney tissues. AnxA3 immunoreactivity with corresponding score and intensity values. Original magnifications: ×40 and ×400 (insets).

Figure 6

Detection of two AnxA3 transcripts in cortex and RCC primary cultures by RT-PCR analysis. A: Schematic representation of the exon structure of the AnxA3 gene. The localizations of the specific forward and reverse primers used in the PCR assays are indicated by arrows. Primer b spans noncontiguous sequences belonging to different exons. B: The a–c primer pair amplified a band of 973 bp of AnxA3 encompassing the full-length coding sequence containing exon III in both cortex and RCC cDNA samples. The b–c primer pair amplified in the same cDNA samples a band of 885 bp corresponding to the full-length coding sequence of AnxA3 lacking exon III. C: The d--e primer pair amplified in cortex and RCC cDNA samples, two bands of 165 and 77 bp corresponding to the AnxA3 transcripts respectively encompassing and lacking exon III. D (top): Partial nucleotide sequence of the AnxA3 transcript isoform containing exon III, the corresponding amino acid sequence that starts with ATG in exon II is enumerated; bottom: Partial sequence of the AnxA3 isoform lacking exon III. The translation of this spliced transcript starts from the “ATG” coding for the first methionine in exon IV (M₄₀ of the unspliced isoform). The tga stop codon (underlined) raised in exon IV by the frame shift attributable to the spliced event, when the atg start codon (underlined) in exon II is used, is shown.

Figure 7

Relative quantification by real-time PCR of AnxA3 transcript containing exon III in 19 matched normal cortex and RCC cultures obtained from corresponding patients. The value expressed as 2^−ΔΔCt represent the fold of transcript expression respect to a pool of all normal cortex cDNA samples, used as calibrator with a value of 1. Each sample has been analyzed in duplicate. All patients are RCCcc except n. 36, n. 41 RCCpap and n. 14, n. 57 RCCmixed.