Response to Chen et al . re: “ β -Caryophyllene Reduces DNA Oxidation and the Overexpression of Glial Fibrillary Acidic Protein in the Prefrontal Cortex and Hippocampus of d -Galactose-Induced Aged BALB/c Mice”

Dear Editor,

Thanks for sharing with us the letter written by Chen et al. ¹ regarding our recently published article within the Journal of Medicinal Food. ² We agree with the authors of the letter in that the description of some data was missed in the experimental design of our article and we apologize for this issue. However, from our point of view, as well as the reviewers', we consider that these omissions were not relevant to understand or refute any part of the publication, including experimental design, results, discussion, and conclusions. Besides, Chen et al. ¹ stated several suggestions that are interesting but not essential to this article; in fact, the results presented in Chávez-Hurtado et al. unveil several opportunities to keep evaluating and demonstrating whether or not β-caryophyllene (BCP) may represent a feasible preventive or therapeutic drug in aging.

Concerning the experimental design, the BCP oil used in this study was obtained as previously described in Gertsch et al. ³ ; it was 95% pure (gas chromatography - mass spectrometry measurement) with β-oxide and α-humulene as the major impurities as previously reported. ⁴ In contrast, we used 4-week-old male BALB/c mice, as stated in the article, and these weighted 22 ± 3 g at the beginning of the study.

Since the study published by Song et al., who first described a cognitive impairment after d-galactose systemic administration, ⁵ it is commonly known and widely reported that a standardization of the d-galactose aging model may vary due to the use of different animals, strains, dosage, administration periods, and routes of administration; however, none of these difficulties may disqualify this well-accepted animal model. Recently, Sadigh-Eteghad et al. explored this issue and published an analysis of the several experimental parameters that seem to determine the effectiveness of this model. ⁶ To date, we can find a broad spectrum of the dosage for d-galactose: from 50 to 1250 mg/kg, but, interestingly, none of these doses have been reported as lethal. ⁶ In our study, we also observed similar findings and a 100% survival rate when we administered 100, 300, 600, and 1200 mg/kg of d-galactose i.p. (supplementary data of Chávez-Hurtado et al. ² ).

Moreover, it is interesting to state that even the same dosage of d-galactose used in one study to induce aging could not develop the same degree of damage in others. ^7,8 In this regard, we have hypothesized that the main factors involved in this variability may include the route of administration and the mouse strain used in this model. We coincide with Chen et al. ¹ that this issue needs to be addressed to establish a reproducible and reliable aging model after chronic administration of d-galactose. Indeed, in the first version of our article, we discussed these points; however, we omitted them due to reviewer's suggestions and space limitation.

Chen et al. ¹ questioned the lack of effects found in locomotor activity, long-term memory, and neuronal loss, but not in cognitive flexibility, suggesting the need of reproducibility of this model. We kindly disagree with their point of view and actually believe that taken together, our findings of cognitive flexibility impairment without alterations in long-term memory indicate that our model indeed “mimics early human aging” as also discussed in Chávez-Hurtado et al. ² Remarkably, we find quite interesting a model that produces mild and progressive changes in the expression of glial fibrillary acidic protein in hippocampus (HIP) and an increase in the oxidation of the DNA in prefrontal cortex (PFC) that, at this initial stage, is not aggressive enough to provoke neuronal death in neither HIP nor PFC, which may explain the lack of long-term memory impairment. ²

To the best of our knowledge, there is no single molecular, cellular, or functional alteration that has been causally associated with aging. Thus, Chen et al. ¹ suggestions to evaluate antioxidant enzymes (SOD, GSH-Px, CAT, and T-AOC) as well as neurotransmitter-related enzymes (AChE and MOA) are valid recommendations that we might take into account for future studies.

It has been previously reported that the neuroprotective effects of BCP could be exerted through different signaling pathways, and at least one of them could be responsible for the results that we had reported in Chávez-Hurtado et al. ¹ BCP's possible mechanisms in d-galactose aging model can include not only neuritogenesis ⁹ and Nrf2/HO-1^10–12 pathway as mentioned by Chen et al. ¹ , but also PGC-1α/PPAR-γ ¹³ and AMPK/CREB/BDNF ¹⁴ ; these pathways have been previously associated with the antioxidant, anti-inflammatory, and cell survival effects of BCP as mentioned in the references 13–16 of our published article. ² However, the analysis of any signaling pathway was out of the scope of this exploratory study. It would be interesting to investigate which of them are involved in the effects exerted by BCP in the d-galactose model, as we already stated in our article.

Moreover, Chen et al. ¹ asked whether the administration of BCP at doses higher or lower than 10 mg/kg still effectively control DNA oxidation or neuroinflammation-associated disability? and if there is a therapeutic window in which the BCP administration decreases the changes in both: astrocytes and DNA oxidation. Although both of them are valued questions, our experimental design was not focused on addressing any of these. We choose the dose of 10 mg/kg of BCP since it was of our interest to evaluate the activation of the cannabinoid receptor type 2 (CB2R). It has been previously demonstrated that higher doses of BCP also exert neuroprotective effects in different models through the activation of not only CB2R ^10,15,16 ; therefore, we did not discard that higher doses of BCP could exert neuroprotective effects in d-galactose aging model and that they could exert a greater protection than 10 mg/kg that even could be reflected in a significant improvement of cognitive flexibility.

Therefore, we strongly believe that, although interesting, the arguments stated by Chen et al. ¹ are only complementary to the discussion section of our article, and some of them could be addressed with different experimental designs that were out of the scope of our original study.