Human population data on air pollution and its effects on the immune system are scarce. A survey was conducted within the framework of the Central European Study of Air Quality and Respiratory Health (CESAR) to measure a panel of immune biomarkers in children of Bulgaria, Czech Republic, Hungary, Poland, Romania, and Slovakia. Seventeen cities were chosen to represent a wide range of exposure to outdoor air pollution. In each, ambient particulate matter of less than 10 microns diameter and less than 2.5 microns diameter (PM10 and PM2.5) were measured with a Harvard impactor. Blood was collected from 366 school children aged 9 to 11 yr between 11 April and 10 May 1996. The percentage of B, total T, CD4+, CD8+, and natural killer (NK) lymphocytes was determined by flow cytometry (Becton Dickinson); total immunoglobulins of class G, M, A and E (IgG, IgM, IgA, and IgE) were measured in serum using nephelometry (Behring). Associations between PM and each log-transformed biomarker concentration were studied by linear regression, in a two-stage model. The yearly average concentrations varied from 41 to 96 micrograms/m3 for PM10 across the 17 study areas, from 29 to 67 micrograms/m3 for PM2.5, and from 12 to 38 micrograms/m3 for PM10-2.5 (coarse). Number of B, CD4+, CD8+, and NK lymphocytes increased with increasing concentration of PM, having adjusted for age, gender, parental smoking, laboratory of analysis, and recent respiratory illness. Differences in lymphocyte number were larger and statistically significant for exposure to PM2.5. Similar results were found when we examined the association between PM and lymphocyte number separately for each laboratory. Total IgG was increased with increasing concentration of PM, significantly in the case of PM2.5. When we repeated the analyses with two other statistical approaches the results did not differ from those reported here. The effect of coarse PM on lymphocyte numbers appears small in comparison to PM2.5. One possible interpretation of our findings is that long-term exposure to airborne particulates leads to inflammation of the airways and activation of the cellular and humoral immune system.

We evaluated humoral immunity by measuring IgG, IgA, IgM, and IgE concentrations in 274 male workers in an iron foundry in Cracow, Poland. There were two groups: 199 coke oven workers and 76 cold-rolling mill workers. The groups were similar with respect to age, length of work (average 15 years), and smoking habits. Exposure to polycyclic aromatic hydrocarbons (PAHs), assessed by personal and area monitoring, ranged from 0.2 to 50 micrograms/m3 benzo[a]pyrene in coke plant workers and was of 3-5 magnitudes higher than in the cold-rolling mill employees. Comparison of the two groups revealed a marked depression of mean serum IgG and IgA in coke oven workers (p < 0.001, Student's unpaired t-test). In the same subjects, serum IgM had a tendency to decrease, whereas serum IgE showed a trend toward higher values. Thus, workers exposed chronically to complex mixtures of air pollutants, composed primarily of PAHs, develop immunosuppression. It remains to be established whether the immunosuppression described here is related to the frequent development of lung cancer reported in coke plant employees. Workers exposed chronically to PAHs should have serum immunoglobulins monitored regularly.ImagesFigure 1.

No abstract

Five asymmetric hybrid plants were obtained between Medicago sativa (2n = 4x = 32) and Medicago arborea (2n = 4x = 32) through sexual reproduction and the use of a cytoplasmically male sterile M. sativa genotype. Over 2,000 pollinations were made to obtain these hybrids. Amplified fragment length polymorphism (AFLP) analysis showed that in the most studied hybrid (WA2273), 4% of the bands unique to the M. arborea parent were present, versus 72% for the unique M. sativa bands. This suggests that only a single M. arborea chromosome or chromosome parts has been transferred. WA2273 had 7% of AFLP bands which were not present in either parent, which is suggestive of chromosome rearrangements as would be expected if only chromosome parts or a single part had been transferred from M. arborea. Phenotypic evidence for hybridity was obtained for pod coiling (1.4 coils in WA2273 versus three coils in the M. sativa parent and its self and testcross populations, and one coil in M. arborea), and Colletotrichum trifolii race 2 resistance (transferred from the resistant M. arborea parent, as the M. sativa parent and the self populations were highly susceptible). The hybrids were self sterile, but were female fertile to a high level when crossed with 4x, but not 2x, M. sativa, indicating they were at or near 4x. Both the pod coiling trait and anthracnose resistance segregated in the progeny of testcrosses between WA2273 and M. sativa. The work demonstrates that agronomically useful traits can be introgressed into M. sativa from M. arborea by use of male sterile M. sativa and sexual reproduction.

In eastern Australia and California, USA, one of the major lethal fungal diseases of lucerne (Medicago sativa) is Stagonospora root and crown rot, caused by Stagonospora meliloti. Quantitative trait loci (QTL) involved in resistance and susceptibility to S. meliloti were identified in an autotetraploid lucerne backcross population of 145 individuals. Using regression analysis and interval mapping, we detected one region each on linkage groups 2, 6 and 7 that were consistently associated with disease reaction to S. meliloti in two separate experiments. The largest QTL on linkage group 7, which is associated with resistance to S. meliloti, contributed up to 17% of the phenotypic variation. The QTL located on linkage group 2, which is potentially a resistance allele in repulsion to the markers for susceptibility to S. meliloti, contributed up to 8% of the phenotypic variation. The QTL located on linkage group 6, which is associated with susceptibility to S. meliloti, contributed up to 16% of the phenotypic variation. A further two unlinked markers contributed 5 and 8% of the phenotypic variation, and were detected in only one experiment. A total of 517 simple sequence repeat (SSR) markers from Medicago truncatula were screened on the parents of the mapping population. Only 27 (6%) SSR markers were polymorphic and could be incorporated into the autotetraploid map of M. sativa. This allowed alignment of our M. sativa linkage map with published M. truncatula maps. The markers linked to the QTL we have reported will be useful for marker assisted selection for partial resistance to S. meliloti in lucerne.