Nitric Oxide Modulation of Erythropoiesis in Rats

To the Editor:

We have read with interest Ponka's article1 reviewing distinct control mechanisms of iron metabolism and heme synthesis in erythroid cells. Erythroid cell growth and differentiation may be regulated by many autocrine and paracrine factors secreted by progenitor cells and other cells present in the bone marrow microenvironment. The author also described intriguing links of nitric oxide (NO) to heme synthesis and erythropoiesis. Actually, such views followed solely from in vitro studies using established erythroleukemic cell lines.2-5 We have recently investigated whether pharmacological modulations of the L-arginine-NO system could affect erythropoiesis in vivo.

Eight-week-old Sprague-Dawley rats weighing 170 to 200 g (n = 34) were used for all experiments, as described previously.6 The animals were fed a regular diet containing 0.16% iron and 1.5% L-arginine and allowed free access to water. They were divided into the following four groups. Group 1 (n = 18) represented control rats. Group 2 (n = 6) received subcutaneous injections of an NO generator, isosorbide dinitrate (Eizai Co Ltd, Tokyo, Japan), at 1.0 mg/kg every day. Group 3 (n = 5) received a potent NO synthase inhibitor, N_G-nitro-L-arginine methyl ester HCl (L-NAME; 50 mg/dL of drinking water). Animals of group 4 (n = 5) were treated with L-NAME (50 mg/dL) together with a substrate of NO synthesis, L-arginine (2.0 g/dL).

At the end of the 2-week study, body weight, tail-artery blood pressure, hematocrit, and urinary excretion of NO metabolites (NOx⁻; brucine reaction) were determined.6 Data were expressed as mean ± SD. The effect of treatment was examined for significance using ANOVA followed by Scheffè test. A P value <.05 denoted the presence of statistical significance.

After 2 weeks of treatment, all rats appeared well and fit, and there was no significant difference in body weight among the different groups. As shown in Fig 1, oral administration of L-NAME induced a significant increase in systolic blood pressure compared with that in animals of other groups, the effect of which was partially inhibited when L-NAME was combined with a large dose of L-arginine. Interestingly, the hematocrit significantly decreased in isosorbide dinitrate-treated rats and significantly increased in L-NAME rats as compared with the other groups. The polycythemic effect of L-NAME was completely abolished by simultaneous administration of L-arginine. Urinary excretion of NOx⁻ was significantly increased by the treatment with isosorbide dinitrate and significantly reduced by L-NAME (but not L-NAME plus L-arginine) relative to control, confirming the effects of these drugs on the NO system.

These data suggest that erythropoiesis may be regulated by the levels of NO (or its related compounds) in rats. It has been proposed that NO indirectly controls levels of ferritin and transferrin receptor expression through activation of the iron-regulatory protein.2,3 It has also been suggested that NO suppresses erythroid-specific gene expression4 or inhibits hemoglobinization at the aminolevulinic acid synthase step.5 In another study, NO appeared to suppress erythropoietin production in vitro.7 Thus, the entire response to NO may serve to decrease cellular heme synthesis and suppress erythropoiesis. It is apparent from our present experiments that such NO modulation of erythropoiesis can occur in the in vivo setting and that the rat model of chronic depletion or excess of NO presented here may be useful for further study in this field. Our observations also may be relevant for understanding the pathophysiology of anemia of inflammation and infection where excessive NO production associated with immune activation may result in decreased erythropoiesis.