Physiological PTEN expression in peripheral T-cell lymphoma not otherwise specified

Peripheral T-cell lymphomas not otherwise specified (PTCL/NOS) is the commonest subtype of PTCL.1 This is a complex entity, characterized by great morphologic, immunophenotypic and clinical variability, whose molecular pathology is still largely unknown.2

Recently, gene expression profiling (GEP) studies allowed the identification of PTCL/NOS-associated signatures, leading to the better understanding of its histogenesis, pathogenesis and prognostication.2 Interestingly, proliferation pathways were found to be commonly altered in PTCL,2,3 highly proliferative cases being characterized by poorer prognosis.4,5

The tumor suppressor PTEN (phosphatase tensin homolog), is a critical regulator for multiple cellular processes including proliferation. In addition to its well-defined role in signaling in the cytoplasm, its nuclear localization seems to contribute to its physiological tumor suppressor activity. Interestingly, somatic mutations of PTEN, including null or missense mutations, and truncations, occur in multiple sporadic tumors. Furthermore, deletions of 10q22-25 have been reported in ~5–10% of B-NHLs, raising the possibility of PTEN involvement in the pathogenesis of lymphoid malignancies.6 Importantly, the great majority of such mutations abolish (81%) or greatly decrease (10%) PTEN phosphatase activity, by determining the absence of the encoded product or the presence of a mutated, nonfunctional protein, finally leading to tumorigenesis.

Interestingly, experimental evidence showed that mice with loss of heterozygosity of the wildtype mPTEN allele develop T-cell lymphomas.7 Furthermore, down-regulation of this molecule was found to be essential for the formation of the typical nuclear lobules in adult T-cell leukemia/lymphoma8 and its alterations may be important in the pathogenesis and progression of mycosis fungoides9 and T-cell acute lymphoid leukemia.10 Notably, partial or complete loss of PTEN was detected in 66.7% of anaplastic large cell lymphomas and in 12.5% of the few other mature T-/NK-cell lymphomas so far studied.11 Nevertheless, the number of PTCLs/NOS was indeed limited and no definitive conclusion can be drawn concerning the possible involvement of PTEN in the pathogenesis of such tumors. The present study was designed to evaluate the possible occurrence of PTEN aberration in PTCL/NOS. We analyzed GEP data of 28 PTCLs/NOS, and 20 samples of normal T cells. Technical details have been previously reported (http://www.ncbi.nlm.nih.gov/projects/geo/).3,12

Genotypes were determined by PCR amplification and direct sequencing. In particular, PCR products of all nine exons and exon-intron junctions of PTEN were directly sequenced in 72 PTCLs/NOS. Primers and relative conditions of amplification are detailed in Table 1.

Finally, we studied PTEN expression by immunohistochemistry (IHC) on tissue micro-arrays (TMAs) containing 34 PTCL/NOS cases.4 PTEN cytoplasmatic expression was tested by a mouse monoclonal antibody, PTEN/MMAC1 Ab-4 (clone 17.A, Thermo Scientific), at a 1:20 dilution, while PTEN nuclear expression was tested by a specific mouse monoclonal antibody, PTEN Ab-6 (clone 28H6, Thermo Scientific), which was applied at a 1:10 dilution. The sections underwent antigen retrieval in citrate buffer (pH=6.0) in a micro-waver at 900W (3 cycles lasting 5' each) and revealed by the EnVision and APAAP techniques, respectively.

First, GEP showed no significant differences in PTEN expression when PTCL/NOS cases were compared to normal T-lymphocytes (Figure 1).

Secondly, direct sequencing did not reveal any significant abnormality. Similarly, PTEN locus was found to be intact in the course of a high-density karyotyping study carried out by the Affymetrix 250k SNPs-array in a series of PTCL/NOS (Hartmann et al., manuscript in preparation).

We then studied the protein expression and localization in a series of 34 PTCL/NOS cases. Importantly, IHC documented physiological nuclear and cytoplasmic expression in 31/31 and 23/26 evaluable cases, respectively (Figure 1). Notably, the 3 cases with nuclear positivity alone lacked genomic abnormalities: they were then considered to be normal as far as PTEN expression is concerned. Possibly, in these cases, the cytoplasmatic negativity reflected a defective immunohistochemical reaction.

Table 1.Primers for PTEN PCR analysis and sequencing, and relative annealing temperatures.

Figure 1.(A) Normalized expression values of PTEN in PTCL/NOS cases vs. samples of normal tissues evaluated by gene expression analysis on DNA microarrays. Median values are indicated by bars. No significant differences in PTEN expression was observed in PTCL/NOS cases compared to normal tissues. (B) CD3 immunostaining in a PTCL/NOS sample (APAAP Technique; x400). Note the high amount of CD3+ neoplastic elements. (C) Cytoplasmatic PTEN immunostaining in a PTCL/NOS sample (Immunoperoxidase Envision+ Technique; Gill’s hematoxylin counterstaining; Olympus BX41 microscope, Olympus CAMEDIA C-7070 camera, magnification x400, colors corrected after acquisition with Adobe Photoshop). (D) Nuclear PTEN immunostaining in a PTCL/NOS sample (APAAP Technique; Olympus BX41 microscope, Olympus CAMEDIA C-7070 camera, magnification x400, colors corrected after acquisition with Adobe Photoshop).

In a previous study, PTEN loss was found by IHC in only 3/24 mature T/NK lymphomas, suggesting the possibility of occasional PTEN involvement in PTCL/NOS pathogenesis. However, the number of PTCL/NOS with such a pattern was not specified, genomic analyses were not performed, and possible IHC reaction failure was not discussed. Thus, based on the present data, obtained in the largest series of PTCLs/NOS so far studied with this purpose, we can conclude that PTEN seems to have physiological distribution in PTCL/NOS. In fact, PTEN was transcriptionally regulated in tumors as in normal T-lymphocytes, lacked structural abnormalities and presented with regular expression in both cytoplasmatic and nuclear compartments. Anyway, it would be of interest to study matched PTCL/NOS cases collected at diagnosis and relapse, respectively, in order to evaluate a possible involvement of PTEN in PTCL/NOS progression.

Acknowledgments:

the authors are grateful to Dr. Maria Teresa Sista for tissue micro-array construction and to Dr. Maria Rosaria Sapienza for gene sequencing.

Footnotes

• ↵* These authors equally contributed
• Funding: this work was supported by BolognAIL, Ateneo 60% (Prof. S. A. Pileri), Centro Interdipartimentale per la Ricerca sul Cancro “G. Prodi”, AIRC, Fondazione CARISBO, Progetto Strategico d’Ateneo (Prof. SA Pileri/Dott. PP Piccaluga) grants.