The aged HSC pool has a lower frequency of T-competent cells and fewer highly proliferative T-competent clones. (A) Two hundred fifty sorted 2-month-old or 18- to 24-month-old HSCs were plated on OP9-DL1 stromal layers and analyzed on days 17 to 19. Data shown represent 5 independent experiments (*P < .001). (B) Limit-dilution analysis was performed on purified HSCs in vitro by OP9-DL1 coculture. Poisson statistics determined frequencies of lineage-competent cells (T lineage young: 1 in 15 [95% confidence interval 1 in 11-22], aged: 1 in 57 [1 in 33-99], P < .001; myeloid lineage young: 1 in 10 [1 in 7-14], aged: 1 in 14 [1 in 10-19], P = .2). Twenty-four wells were analyzed for each cell dose. (C) Single HSCs were plated on OP9-DL1 layers with cytokines and analyzed by flow cytometry 18 days later. Data shown are restricted to those clones that generated detectable T-lineage progeny and are thus T competent. Data were analyzed using the binomial distribution probability mass function. Colonies with greater than 10 000 progeny were considered highly proliferative (denoted by dashed line), and the frequency of highly proliferative T-competent clones was greater in the young HSC pool than aged (young: 25 highly proliferative colonies out of 50 total T colonies; aged: 10 out of 32; P = .015). Differences were also significant when other arbitrary cutoff values from 5000 to 100 000 were used. For the myeloid lineage, aged HSCs had a higher frequency of highly proliferative myeloid producing clones compared with young HSCs (young: 47 highly proliferative colonies out of 72 total myeloid colonies; aged: 39 out of 52; P = .04). Average colony sizes are indicated by horizontal bars.