Anemia in older adults is highly prevalent, affecting 10% of adults over age 65 and 20% of adults over 85 in the NHANES III study. At least a third of those anemia cases are likely to have some underlying inflammatory component (Guralnik, et al, 2004). Circulating levels of the inflammatory cytokine interleukin-6 (IL-6) are elevated in older adults (Harris et al, 1999), and serum IL-6 levels negatively correlate with hemoglobin concentration in frail older adults (Leng, et al, 2002). Given its known negative impact on a variety of tissue types in older adults, we hypothesize that IL-6 contributes to poor red blood cell production in aging organisms. To test this hypothesis we compared indices of inflammation, erythropoiesis, and iron metabolism in normal C57BL/6 (B6) mice and IL-6 knock-out (KO) mice.
We investigated features of the peripheral blood, extramedullary hematopoiesis (EH), and iron homeostasis in 2 and 24 month (m) old B6 mice and compared them to groups of IL-6 KO mice of the same age. We observed an inflammatory phenotype in aged B6 mice, as circulating neutrophils were higher in 24m than in 2m mice (1.13 ± 0.70 vs 0.39 ± 0.20 K/μl, p=0.001 T-test). Monocytes were also higher in aged mice (0.30 ± 0.13 in 24m vs 0.14 ± 0.09 K/μl in 2m, p=0.001 T-test). Circulating IL-6 was higher in 24m old mice than 2m (15.4 ± 7.0 vs 3.1 ± 2.4 pg/ml), and Keratinocyte-derived Cytokine was also higher in aged mice (153.8 ± 25.3 vs 107.5 ± 32.7 pg/ml, p< 0.05 for both T-test).
Age-dependent anemia was attenuated in the IL-6 KO mice. Hemoglobin concentration was higher (14.4 ± 0.7 g/dL) in 2m old B6 mice than 24m B6 mice (10.8 ± 1.8 g/dL, p < 0.001 T-test), while the difference in IL-6 KO mice was not as severe (13.9 ± 0.5 in 2m vs 12.0 ± 0.9 g/dL in 24m, p=0.02 T-test). Two-way factorial ANOVA of data from the four groups indicated a significant interaction between genotype and age for hemoglobin concentration (p=0.02). We defined EH as an increase in both the spleen weight and the percentage of committed splenic erythroid progenitors (Ter119+ splenocytes). Spleen weight was three-fold higher in 24m B6 mice (182 ± 81 vs 61 ± 5 mg, p= 0.002 T-test) but less than two-fold higher in 24m IL-6 KO (66 ± 14 vs 40 ± 3 mg, p= 0.002 T-test; two-way factorial ANOVA for age by genotype interaction, p=0.01). Percent Ter119+ splenocytes were two-fold higher in 24m B6 mice (74 ± 5 vs. 38 ± 4%, p< 0.001) but there was no significant difference between 2m and 24m IL-6 KO (49 ± 8 vs 46 ± 14%). Thus, EH was less severe in the aged IL-6 KO mice (two-way factorial ANOVA p < 0.001). In keeping with these measures of erythropoietic stress, we found that aged B6 mice had a higher proportion of detectable Epo levels [65 pg/mL (range = 0–289)] compared to their 2m counterparts, which had undetectable levels (Fisher's exact test p=0.026). 2m and 24m IL-6 KO had largely undetectable levels of serum Epo.
We found a trend of 1.5 fold increased liver hepcidin mRNA expression in 24m B6 vs 2m B6 mice (T-test, p=0.06), and a trend toward a reduction in hepcidin expression in IL-6 KO mice at 24m compared to B6 mice at 24m (T-test, p = 0.125). Consistent with hepcidin levels, a trend toward lower serum iron in the 24m B6 mice than in their 24m IL-6 KO counterparts was also observed (140 ± 13 μg/dL vs. 168 ± 19 μg/dL, p = 0.07 T-test).
Aside from specific effects on erythropoiesis, a decline in circulating lymphocyte number in the context of increased circulating neutrophils may be indicative of myeloid skewing. While lymphocytes were higher (3.3 ± 2.1 K/mL) in 2m B6 than 24m B6 (1.7 ± 1.0 K/mL, p=0.05 T-test) there was not a significant difference in lymphocyte numbers between aged and young IL-6 KO (two-way factorial ANOVA, p = 0.02).
Our data suggest that IL-6 is a significant modifier of the anemia observed in an aging mouse model. We conclude that aging contributes to the development of anemia, and that aged animals lacking IL-6 appear to have reduced erythropoietic stress. Hepcidin results were somewhat inconclusive, suggesting IL-6 may have hepcidin-independent effects on erythropoiesis with age. Our results show that while IL-6 is not the sole mechanism behind the anemia associated with aging, it may yet present a potential therapeutic target to alleviate some of the deficits associated with this common condition in older adults. Future experiments should investigate whether hepcidin is required for the development of anemia with age and whether therapeutics directed against IL-6 improve erythropoiesis in aged B6 mice.
Langdon:Celgene Corporation: Research Funding. Roy:Celgene Corporation: Research Funding.
Asterisk with author names denotes non-ASH members.