The development and evaluation of drugs to elevate fetal hemoglobin in the
treatment of the genetic diseases of hemoglobin would be facilitated by the
availability of reliable cell assays. We have used rear-time, quantitative
polymerase chain reaction (PCR) analyses of globin messenger RNA (mRNA) le
vels in a biphasic, erythropoietin-dependent primary culture system for hum
an adult erythroid cells in order to assay compounds for their ability to m
odulate levels of adult (beta) and fetal (gamma) globin mRNA, Complementary
DNA synthesized from total RNA extracted at timed intervals from aliquots
of cells were assayed throughout the period that the culture was studied. g
amma-globin mRNA levels were found to be much lower (less than 1%) than bet
a-globin mRNA levels. At concentrations of agents chosen for minimal effect
on cell division, we find that the 3 drugs studied, 5-azacytidine (5 mu mo
l/L), hydroxyurea (40 mu mol/L), and butyric acid (0.5mmol/L) significantly
increase gamma-globin mRNA levels. Interestingly, hydroxyurea also had a s
mall stimulatory effect on beta-globin mRNA levels, while butyric acid caus
ed a twofold inhibition of beta-globin mRNA levels, and 5-azacytidine had l
ittle effect on beta-globin mRNA levels. The net result of all 3 drugs was
to increase the gamma/(gamma + beta) mRNA ratios by threefold to fivefold.
These data suggest that the mechanism is distinct for each drug. The profil
e of butyric-acid-induced changes on globin gene expression is also quite d
istinct from changes produced by trichostatin A, a known histone deacetylas
e inhibitor. Quantitative PCR analyses of human erythroid cells should prov
e useful for studying the mechanism(s) of action of known inducers of gamma
-globin and identifying new drug candidates.
(C) 2000 by The American Society of Hematology.