Genetic diversity of KIR natural killer cell markers in populations from France, Guadeloupe, Finland, Senegal and Réunion

Killer cell immunoglobulin-like receptors (KIRs) belong to a diverse family of natural killer (NK) cell receptors recognizing human leukocyte antigen (HLA) class I molecules. Due to this functional link, KIR molecules are expected to display a high polymorphism, such as their HLA ligands. Moreover, many studies conducted in mouse and human models have shown that NK-KIR receptors play an important role in haematopoietic stem cell transplantation (HSCT). A beneficial impact of peculiar KIR ligand (HLA) mismatching has been reported suggesting a role to this combinatory HLA-KIR polymorphism. It is thus important to investigate KIR diversity in various human populations. To this end, we used polymerase chain reaction-sequence-specific primers to evaluate KIR gene in five selected populations (France, Guadeloupe, Senegal, Finland and Réunion). Genotypic and haplotypic frequencies were computed, as well as genetic distances and dendrogram (phylip package). These data illustrate the genetic relationship of these five populations through the KIR polymorphism. Results revealed a wide diversity in KIR gene frequencies in Guadeloupe and Réunion, and a high specificity in Senegal. The obtained dendrogram indicated small genetic distances between France, Guadeloupe and Réunion as well as between France and Finland. Senegal showed a distant genetic relationship with the other countries and, interestingly, an inverted ratio of coding/non-coding (KIR2DS4/1D) alleles compared with Caucasians. These data expose the broad diversity in KIR genes worldwide and show that KIR genes are pertinent tools in human population genetics. If the role of KIR donor-recipient incompatibilities is confirmed, KIR diversity according to ethnicity should be taken into account during the selection of HSCT donors.     

Autologous and allogeneic HLA KIR ligand environments and activating KIR control KIR NK‐cell functions

NK‐cell function is regulated by a balance between inhibitory and activating killer cell immunoglobulin‐like receptors (KIR) that specifically recognize HLA class I molecules. Using KIR‐specific mAb to discriminate between KIR2DS1 and KIR2DL1 receptors, we show that KIR2DS1⁺ NK cells are C2‐alloreactive only from C2^? individuals. Moreover, using an in vitro model of NK‐cell expansion, we show here that the frequency of KIR2DL1⁺ NK cells is significantly higher in the absence of C2 ligand on stimulator EBV‐B cells than in its presence. This observation was made regardless of the presence or absence of the autologous C2 ligand, suggesting that the C2^? EBV‐B stimulator cells used in this in vitro model could activate unlicensed KIR2DL1⁺ NK cells. In the case of KIR2DL1⁺/S1⁺ genotyped individuals, KIR2DS1⁺ NK‐cell frequency was increased after stimulation with C2⁺ compared with C2^? stimulator B cells, but only from C2^? individuals. Altogether, these data highlight the C2 alloreactivity of KIR2DS1⁺ NK cells that is only observed in C2^? individuals. These results provide new insights into the way in which NK KIR cell expansion might be regulated in an allogeneic environment.     

Preparative two-step purification of human IL-2 by HPLC and hydrophobic affinity chromatography

Interleukin 2 (IL-2) has been purified by a protocol using gel filtration high performance liquid chromatography (HPLC) and hydrophobic affinity chromatography with blue-trisacryl M. Peripheral blood lymphocytes or tonsillar lymphocytes were stimulated with phytohemagglutinin (PHA). Serum free conditioned medium (CM) containing IL-2, other lymphokines and residual PHA molecules was analyzed after 3 variations of ammonium sulfate (AS) precipitation: (1) precipitation of CM with 50% AS yielded a precipitate containing most of the residual PHA but also a fraction of IL-2. (2) Precipitation with direct 80% AS of crude CM yielded both IL-2 and residual PHA. (3) A double step procedure (50% AS followed by 80% AS) yielded a precipitate containing IL-2 but free of residual lectin.

HPLC purification of these various AS-precipitated materials or of lyophilized crude CM yielded 2 peaks with mitogenic activity as assayed with the CTLL2 murine clone or IL-2-dependent human Con A-stimulated lymphoblasts. IFN was easily separated from IL-2 and PHA, but BCGF still copurified with IL-2. Peak I (25 kDa) was enriched 400-fold for IL-2 while peak II (68 kDa) contained the residual PHA.

The IL-2-containing fractions eluted from HPLC were further purified by blue-trisacryl M chromatography. The IL-2 eluted with 0.4 M NaCl. The entire protocol (HPLC followed by blue-trisacryl) led routinely to 8000-fold IL-2 enrichment. Preparative HPLC directly applied to lyophilized crude (CM) enriched IL-2 activity 400-fold with yield averaging 60% of the IL-2 input. The final material was free from interferon and IL-1, but BCGF still copurified with IL-2. The 2-step purified material (HPLC and blue-trisacryl) gave 2 bands in SDS-PAGE both of which contained IL-2.     

Human bladder cancers and normal bladder mucosa present the same hot spot of heterozygous chromosome-9 deletion

Loss of heterozygosity (LOH) on chromosome 9 is the most frequent genetic alteration in bladder cancer identified to date, suggesting the presence of key gene(s) for this pathology. In this study, we examined 44 bladder tumors and 21 normal bladder samples for LOH on both arms of chromosome 9. Sixteen microsatellite markers, 12 on the short arm (encompassing 9p21-22) and 4 on the long arm (encompassing 9q33-34), were chosen for their highly frequent alterations in bladder cancer. LOH for at least one marker was identified in 42 tumor samples (95.5%), and 14 tumors (32%) displayed LOH for all informative tested markers. Detailed analysis showed that 2 markers on chromosome 9p (D9S157 and D9S156) had the highest frequencies of allelic loss (about 70%), independent of tumor grade and stage. The same study was performed on the 21 normal bladder mucosa samples: 50% of informative cases presented a single specific LOH at the D9S156 locus. Normal samples showing LOH at this locus were therefore screened with 3 novel microsatellite markers in the 810-kb region incorporating D9S156. Using this marker, we found no further heterozygous loss in this region. This result allows different interpretations of the D9S156 loss in normal bladder mucosa, and suggests that D9S156 may be more an indicator of bladder epithelium impairment than a tumor-initiation marker. Similarly, this unexpected result calls in question the interpretation of LOH studies. Int. J. Cancer 77:821–824, 1998.© 1998 Wiley-Liss, Inc.     

The pathology of familial breast cancer: histological features of cancers in families not attributable to mutations in BRCA1 or BRCA2.

Breast cancers arising in carriers of mutations in the breast cancer susceptibility genes, BRCA1 and BRCA2, differ histologically from each other and from breast cancers unselected for a family history. However, a substantial proportion of families with multiple cases of breast cancer is not attributable to these two genes (non-BRCA1/2 families). We have now characterized the pathology of 82 breast cancers from non-BRCA1/2 families. Breast cancers in non-BRCA1/2 families were of lower grade (P = 0.0018), showed fewer mitoses (P < 0.0001), less nuclear pleomorphism (P = 0.0014), less lymphocytic infiltrate (P < 0.0001), a lesser extent of the tumor with a continuous pushing margin (P = 0.004), a lesser extent of the tumor composed of solid sheets of cells (P = 0.0047), less necrosis (P = 0.002), and wereparison with BRCA2 tumors, non-BRCA1/2 tumors were lower grade (P = 0.017) and exhibited less pleomorphism (P = 0.01) and more tubule formation (P = 0.05). In comparison with control breast cancers unselected for a family history of the disease, non-BRCA1/2 tumors were of significantly lower grade (P = 0.001), showed less pleomorphism (P = 0.0002), and had a lower mitotic count (P = 0.003). The results indicate that non-BRCA1/2 breast cancers differ histologically from both BRCA1 and BRCA2 breast cancers and are overall of lower grade. They also suggest that non-BRCA1/2 breast cancers differ from nonfamilial breast cancers, but these differences may be attributable to various types of bias.     

Influence of di- and tri-phenylethylene estrogen/antiestrogen structure on the mechanisms of protein kinase C inhibition and activation as revealed by a multivariate analysis

We have performed a systematic study of the interaction of 36 di- and tri-phenylethylene derivatives (DPEs and TPEs) with protein kinase C (PKC). The results were submitted to a multivariate analysis in order to identify the structural features that might be implicated in interference with the activity of three PKC subspecies under three enzyme activation conditions. Four groups of test-compounds, each with common chemical features, could be distinguished clearly. The first group comprised all TPEs substituted with at least one basic dialkylaminoethoxy side-chain. These inhibited type alpha, beta and gamma PKC subspecies activated by Ca2+ and phosphatidylserine (PS) with or without diolein (DO) at micromolar concentrations but did not inhibit protamine sulfate phosphorylation. The other effectors, which all possessed a 1,1-bis-(p-hydroxyphenyl) ethylene moiety, influenced PKC activity at high concentrations (30-200 microM) and could be divided into two groups. One group constituted PKC inhibitors in the TPE series and inhibited PKC activated by Ca2+, PS and DO, as well as protamine sulfate phosphorylation. The other group constituted dual-type inhibitors/activators in the DPE series and stimulated PKC in the presence of Ca2+ and low PS concentrations but inhibited the enzyme in the simultaneous presence of DO. The fourth group of compounds was inactive and had, for the most part, one or two substituents with weak steric hindrance. In agreement with previous data for six lead compounds, this study suggests that, in these chemical series, a basic amino side-chain leads to interaction with phospholipid and the regulatory domain of PKC, whereas a 1,1-bis-(p-hydroxyphenyl) ethylene moiety leads to interaction with the catalytic domain of the enzyme.     

Recurrent chromosome aberrations in human lung squamous cell carcinomas

Cytogenetic study of seven cases of previously untreated lung squamous cell carcinomas (SQC) is reported. Chromosome numbers vary from 38 to 538, with a majority of hypotriploid karyotypes with complex rearrangements. The numbers of recurrent imbalances were evaluated in considering the average number of chromosomes or chromosome segments in each analyzed metaphase and for each case. In decreasing order of frequency, deficiencies for 3p, 5q, 8p, Y, 5p, 10p, 13, and, to a lesser degree, for 8q, 9, 10q, 11pter, 14, 15, and 21 were observed; the excesses principally involve 1q, 3q, and 7q. In three tumors, homogeneously staining regions were observed at various chromosome sites. Most chromosome rearrangements occurred after breakage in constitutive heterochromatin, and no recurrent breakpoints were found in euchromatin except 11p15. The major consequences of these anomalies may be chromosomal imbalances, leading to hemizygosity and perhaps related to gene dosage, rather than to alterations of genes.     

Alternating radiotherapy and chemotherapy schedules in small cell lung cancer, limited disease

Sixty-three evaluable patients with limited small cell lung carcinoma were entered into two pilot studies alternating 6 cycles of combination chemotherapy (Doxorubicin 40 mg/m2 d 1; VP16213 75 mg/m2 d 1, 2, 3; Cyclophosphamide 300 mg/m2 d 3, 4, 5, 6; and Methotrexate 400 mg/m2 d 2--plus folinic acid rescue--or Cis-Platinum 100 mg/m2 d 2) with 3 courses of mediastinal radiotherapy as induction treatment. The first course of radiotherapy started 10 days after the second cycle of chemotherapy; there was a 7 day rest between chemotherapy and radiotherapy courses. This 6 month induction treatment was followed by a maintenance chemotherapy. The total mediastinal radiation dose was increased from 4500 rad in the first study to 5500 rad in the second. Both protocols obtained a complete response (CR) rate of greater than 85% (with fiberoptic bronchoscopy and histological verification). Local control at 2 years was 61% in the first study and 82% in the second. Relapse-free survival at 2 years was 32 and 37%, respectively. Toxicity was acceptable. We conclude that our results justify further clinical research in alternating radiotherapy and chemotherapy schedules.