The same drug brought different maze results in male and female mice

Since the previous research note · Partly supported

Track continuity

Since the previous research note: What happened when treatment began before memory problems in mice?

Previous expectation: An early senescence-targeting intervention in mice might help preserve memory before substantial impairment develops, subject to translation, timing, and safety.

What this study adds: This study tested ABT-263 in a different atherosclerotic mouse model and included both preventive and later treatment windows; it found memory benefits in males but no equivalent benefit, and some adverse outcomes, in females.

---

What changed in this study

A drug intended to clear senescent cells improved some measures of learning and memory in middle-aged male mice with atherosclerosis. In female mice from the same model, the treatment did not provide the same benefit and was associated with slower learning. The contrast is the central finding of a study testing whether senescent cells might contribute to vascular cognitive impairment.

The compound was ABT-263, also called navitoclax. In this experiment, a senolytic drug was being tested as a way to remove senescent cells that might be affecting blood vessels and brain tissue. The results do not show that the drug improves cognition in people. They show that, in this particular mouse model, sex was closely tied to the direction of the cognitive response.

What the experiment compared

The researchers used male and female atherosclerotic ATX mice and tested two treatment windows. In the preventive window, mice received ABT-263 from 3 to 6 months of age. In the later treatment window, mice received it from 9 to 12 months. Each course lasted three months and involved repeated five-day dosing cycles separated by two weeks of rest. Other mice received the matching placebo.

The middle-aged result therefore refers to mice treated between 9 and 12 months, not to a human age equivalent. The animals were tested in a water maze to assess how quickly they learned the location of a hidden platform. After learning, the platform was removed and the mice were tested one hour later and again 72 hours later. Time spent in the former platform area was used to assess short-term and delayed memory retention.

This design allowed the researchers to ask two related questions. Could ABT-263 help when cognitive and vascular problems were already present in the model? And did the answer look different when treatment began earlier? The comparison between males and females was part of both questions rather than an afterthought.

Male mice learned faster after the later treatment

Among 12-month-old male mice, ABT-263 accelerated water-maze learning. By the fifth training day, the average time to reach the target was about half as long in treated males as in placebo males, 15 seconds compared with 33 seconds. The study also found improved short-term memory retention in treated males.

The delayed probe test added a more qualified memory result. The overall delayed discrimination measure tended to improve in males, although the reported p-value was not below the conventional threshold. Treated males nevertheless spent significantly more time in the target quadrant than in the opposite quadrant, a pattern not seen in the placebo group. The authors interpreted this as evidence of delayed memory retention after treatment.

The improvement was not readily explained by the mice simply moving differently. Mean swimming speed over the learning period was not affected by sex or treatment, so the faster maze performance was not attributed to a locomotor difference in the tested groups.

Female results followed another pattern. Placebo-treated females began to improve their platform-finding performance earlier in training, while treated females showed a delay of about three days before their learning process became evident. By the fifth day, the average times were similar between the groups, but the treatment had not produced the faster learning seen in males. ABT-263 also failed to improve female short-term memory retention, and the delayed probe pattern was consistent with a harmful rather than helpful effect in that group.

The younger mice make the result harder to reduce to age

The preventive groups showed that the contrast was not limited to the later treatment window. In young male ATX mice, ABT-263 improved spatial memory retention. The treatment was also associated with greater sensitivity of cerebral arteries to shear stress, more CD31-positive endothelial cells in the hippocampus, and lower activation of astrocytes and glial cells.

Young females did not show the same package of findings. Their delayed memory retention tended to improve, but the result was weaker than the male memory effect. ABT-263 also increased aortic plaque area by about one-third in young females, while brain measurements included greater astrocyte activation and lower expression of CD31-positive cells. The preventive treatment did not produce a clear endothelial benefit in this group.

The middle-aged groups had their own differences. The later treatment slightly reduced aortic lesion size in males and did not change it in females. These vascular findings do not by themselves explain the maze results, but they show why the outcome cannot be summarized as a simple effect of getting older. Treatment timing and sex both shaped what was observed, and a result that looked favorable in one group could be neutral or adverse in another.

Vascular and inflammatory markers did not move in one direction

The middle-aged male memory improvement was accompanied by changes in several biological measurements. In cerebrovascular tissue, markers associated with inflammation and senescence-related damage tended to be lower after ABT-263. The treatment was also associated with a higher density of CD31-positive cells and lower activation of glial cells.

Yet the measured endothelial function of isolated pial cerebral arteries did not improve in the middle-aged males. In the female mice, the pattern was almost reversed for one vascular measure: ABT-263 increased the maximum dilation response to shear stress. That improvement in arterial dilation was not accompanied by better learning or memory.

Female tissue measurements also moved in a concerning direction. In the cerebrovascular fraction, ABT-263 increased p21 and several inflammation-associated factors, including PAI1. In the brain parenchymal fraction, p21 increased substantially as well. The study linked this rise in markers with the lack of cognitive benefit, but the measurements remain associations within the experiment. They do not establish that any one marker caused the slower learning.

The mismatch between vascular function and cognition is important. A larger response in an isolated artery does not necessarily represent better coordination between blood flow and neural activity throughout the living brain. Conversely, improved behavior can occur alongside no detectable change in the particular endothelial measure used here. The findings therefore support a sex-dependent response without supplying a single biological switch that explains it.

What the study can identify about senescent cells

The paper used p21 and several senescence-associated secretory phenotype markers, including IL-6, TNF-alpha, PAI1, and Angptl2, to examine tissue changes. It also assessed markers associated with endothelial cells, astrocytes, and microglia. These measurements provide clues about inflammation, vascular tissue, and glial activation, but they do not identify which cells were senescent.

There is no universal cell-specific marker of senescence, and p16 was not detectable in the brain samples examined in this study. As a result, the measurements cannot show that ABT-263 directly removed a defined population of senescent brain or vascular cells. They also cannot determine whether the male benefit and female harm arose from different cell populations, different drug sensitivity, hormonal factors, or another biological pathway.

The vascular test had a separate boundary. Researchers measured flow-mediated dilation in isolated pial arteries outside the body. That is useful for examining a property of those arteries, but it does not capture the full in-vivo cerebrovascular system, including the microvascular processes involved in matching blood flow to neural activity. Direct measurements of cerebral blood flow and neurovascular coupling would address a different level of biology.

The study also lacked age-matched wild-type controls, and the animals used for cognitive testing had been pre-trained. The authors considered pre-training to have had limited effects in their data, but these design features still matter when interpreting how broadly the result should be applied. Most importantly, this is one controlled animal study in one atherosclerotic model. It does not establish a human treatment response, a consistent benefit across sexes or ages, or a general rule that senolytic therapy is helpful or harmful for everyone.

Could cell-clearing treatments help protect memory?

The possible future value of this line of research is more precise treatment for people whose vascular disease puts cognition at risk. If the sex-dependent response seen in these mice can be understood, and if a senolytic approach proves safe and effective in people, it could help guide therapies designed to preserve memory while reducing the chance of harm in groups that respond differently.

That capability remains conditional on much more than a positive maze result. The mouse findings identify a potentially important treatment divide, but they do not yet provide a human dose, a confirmed target cell, or evidence that clearing senescent cells will protect cognition outside this model. Their contribution is to show why those questions need to be answered before a cell-clearing strategy can be treated as a broadly useful route to protecting memory.

Key Sources

• Primary source: The senolytic ABT-263 improves cognitive functions in middle-aged male, but not female, atherosclerotic LDLr<sup>-/-</sup>;hApoB<sub>100</sub><sup>+/+</sup> mice.
  DOI or Journal Page · Peer-Reviewed Paper · 2025-02-21