The likely reason why the reviewer raises this issue is that a cross-sectional approach to mediation usually is not appropriate. See:

Maxwell, S. E., & Cole, D. A. (2007). Bias in cross-sectional analyses of longitudinal mediation. Psychological Methods, 12(1), 23-44.

Maxwell, S. E., Cole, D. A., & Mitchell, M. A. (2011). Bias in cross-sectional analyses of longitudinal mediation: Partial and complete mediation under an autoregressive model. Multivariate Behavioral Research, 46(5), 816-841.

Therefore, there is no real "justification" for choosing a cross-sectional design for mediation. Perhaps a reasonable argument could be made to say a longitudinal data collection was not feasible in your case (for whatever reason). You could mention this as a limitation of your study and a goal for future research (i.e., to replicate the findings with longitudinal data).

I think the answer by Christian Geiser represents a bit of a "purist" approach. In sociology, the use of mediation in cross-sectional "path analyses" goes back to the earliest examples, including Blau and Duncan on Status Attainment (1967). Since then, there have probably been thousands of studies that use this design -- that doesn't make it right, but it does quite literally make it "acceptable."

I think you could reply that having over-time data would certainly be preferable, even though the practice of assessing mediation in cross-sectional analyses is quite widely used, then go on to mention it as a limitation for your study.

Maedeh Ahadi another ChatGPT answer, huh?

Thank you Dr Christian Geiser for your helpful answer.

Thank you Dr David L Morgan for your fruitful answer.