All share a xanthine ring, the core chemical framework of caffeine and related plant-derived purine alkaloids, but the compounds displayed markedly different biological effects in the injury models. The experiment deliberately used two complementary approaches: excessive L-glutamate, abbreviated L-GLU, to mimic excitotoxic neuronal injury, and hydrogen peroxide to generate oxidative stress. This design revealed that protection depended on the type of damage, the severity of injury, the concentration used and the duration of exposure.

Before testing protection, the team established concentrations that did not themselves reduce cell viability. Cytotoxicity was measured with the MTT colorimetric assay, in which metabolic conversion of a tetrazolium salt to formazan provides a readout associated with viable cells. Each concentration was evaluated in six replicates, and the selected non-toxic or maximum safe concentrations differed markedly among compounds: they ranged from 0.1 micromolar for Jas 1 to 100 micromolar for Jas 10.

These compound-specific ranges are important because the later percentages describe protection relative to matched injury controls, not equivalent absolute doses or evidence that every member is equally potent. In the L-GLU model, cells were challenged with 60 millimolar L-GLU after pretreatment with the test compounds, and viability was assessed by MTT after a further 24 hours. Jas 9 was the only derivative to produce statistically significant protection.

At 50 micromolar, it improved viability by approximately 7% relative to the L-GLU-only group, while 100 micromolar produced an improvement of about 10%; the reported significance levels were p < 0.05 and p < 0.01, respectively. The other nine compounds did not significantly improve survival under those conditions. The result indicates that modifying caffeine’s structure did not automatically yield broad protection against excitotoxicity.

That contrast became more pronounced in the hydrogen-peroxide experiments, which were conducted at two injury levels. A severe model used 6 micromolar hydrogen peroxide and reduced viability by roughly 50%, while a mild model used 5 micromolar and produced an approximately 20% loss. With two hours of compound pretreatment, Jas 2, Jas 5, Jas 6 and Jas 8 showed statistically significant cytoprotection in the severe model.

The largest reported increases were about 18% for Jas 8 at 0.1 micromolar and Jas 5 at 0.1 micromolar. These gains were measured relative to the damaged control and did not restore viability to the level of untreated cultures. Extending pretreatment from two hours to 24 hours did not produce a uniformly stronger response.

Under severe oxidative injury, Jas 8 at 0.1 micromolar and Jas 5 at 1 micromolar remained protective, each corresponding to an improvement of approximately 7%. The authors interpreted the continued activity of Jas 8 as evidence that its effect could be stable under prolonged exposure, but the measured benefit under the severe, extended-preratment condition was smaller than some effects seen with shorter pretreatment. The experiment therefore does not support a simple claim that longer treatment always produces greater protection.

In the mild oxidative-stress model, significant protection was reported for Jas 2, Jas 7, Jas 8 and Jas 10 at selected concentrations. Jas 8 at 0.1 and 1 micromolar, Jas 2 at 1 micromolar and Jas 10 at 10 micromolar were significant at p < 0.01; Jas 7 at 10 micromolar and Jas 2 at 10 micromolar were also reported as significant. Among them, Jas 8 produced the largest increase in cell viability.

The fact that different compounds were active in mild and severe injury, at different concentrations and with different pretreatment schedules, points to a narrow and context-dependent activity profile rather than a straightforward linear dose-response relationship. The authors consequently identified Jas 8 as a lead candidate for further pharmacological optimization and evaluation in more advanced experimental systems. The researchers then examined how the leading compounds affected reactive oxygen species using the fluorescent probe DCFH-DA.

Cells were treated with the test compounds and exposed to 6 micromolar hydrogen peroxide for either 30 or 60 minutes. Jas 8 significantly lowered fluorescence at 0.1, 1 and 10 micromolar in both assays. At 1 micromolar, the compound reduced the signal by approximately 50% after 30 minutes of hydrogen-peroxide exposure and by about 70% after 60 minutes, compared with cells exposed to hydrogen peroxide alone.

These observations provide experimental evidence that Jas 8 can reduce the intracellular oxidative signal detected by this assay under the tested conditions. The results were reported as means with standard deviations from three independent experiments, with each condition tested in six replicates. Jas 8’s overall profile made it the study’s lead candidate, whereas Jas 9 was distinguished by its activity against L-GLU injury and Jas 5 showed selected protective effects.

The xanthine structure provides a possible mechanistic connection to adenosine-receptor modulation, because caffeine is a nonselective adenosine-receptor antagonist, and the authors suggest that this pathway might contribute alongside antioxidant activity. The experiments reported here did not directly test that mechanism. They also do not establish that the compounds cross the blood–brain barrier, reach therapeutic concentrations in the human brain, avoid toxicity or improve symptoms in people.

Their relevance to neurodegenerative disease is therefore limited to the cellular findings and the experimental rationale for further investigation, rather than demonstrated clinical benefit. Overall, the study supports a mechanism-specific conclusion: caffeine-derived structures can exhibit different protective profiles depending on whether cells face oxidative or excitotoxic injury. Jas 8 produced the most consistent activity in the oxidative-stress experiments, combining improved viability with substantial reductions in the DCFH-DA signal, while Jas 9 alone was significant in the L-GLU model.

The results identify Jas 8 as a candidate for further optimization, but they do not yet establish its effectiveness, safety or mechanism in more complex biological systems.

Subject of Research: In vitro neuroprotective and antioxidant effects of caffeine-derived compounds in human neuroblastoma cells

Article Title: Caffeine-derived 8-α-methylthioglycolic acid compounds as promising neuroprotective agents in cellular models of neurotoxicity

Article References: Stefanova, D., Mitkov, Y., & Tzankova, V. (2026). Caffeine-derived 8-α-methylthioglycolic acid compounds as promising neuroprotective agents in cellular models of neurotoxicity. Pharmacia, 73, Article e195774. https://doi.org/10.3897/pharmacia.73.e195774

Image Credits: AI Generated

DOI: 10.3897/pharmacia.73.e195774

Keywords: caffeine derivatives, neuroprotection, oxidative stress, excitotoxicity, SH-SY5Y cells, xanthines, reactive oxygen species, cell viability, hydrogen peroxide, laboratory research, Caffeine-derived, methylthioglycolic