1. Two primer sequences used for semiquantitative PCR analysis have previously been published. The primer sequence for ctr1 detection is from the work of Tennant et al. (Tennant J, Stansfield M, Yamaji S, Srai SK, Sharp P (2002) Effects of copper on the expression of metal transporters in human intestinal Caco-2 cells. FEBS Lett 527: 239-44), while the primer sequence for nhe1 is from the work of Miraglia et al. (Miraglia E, Viarisio D, Riganti C, Costamagna C, Ghigo D, Bosia A (2005) Na+/H+ exchanger activity is increased in doxorubicin-resistant human colon cancer cells and its modulation modifies the sensitivity of the cells to doxorubicin. Int J Cancer 115: 924-9).
2. The primer sequence used for the amplification of nhe1 has not been listed correctly in Table 1. Instead of forward primer sequence ccagtcattgccttctacc, the correct sequence, as previously published in paper by Miraglia et al., should be ccagCtcattgccttctacc.
3. Method for determination of apoptosis has not been correctly described. Combination of propidium iodide and ethidium bromide staining cannot discriminate live from dead (apoptotic or necrotic) cells. The authors should have described the method in which fluorescein diacetate and propidium iodide are used, if this was the method they used for apoptosis determination, since they mention it in figures 1A and 8C.
4. The authors cannot claim that the “mRNA expression of the influx pumps CTR1 and NHE1 was reduced by 50% and 30%, respectively (Fig. 4A)” (i.e. express the level of downregulation in percentages), since they have not performed statistical analysis of densitometry analysis from 3 independently isolated RNAs and independently performed RT-PCRs that they have shown representatives from. The representative figures essentially show duplicate PCR products since the same cDNA was used in concurrently performed PCR reactions. Although the authors mention in Materials and Methods section that all data were analyzed by unpaired Student’s t-test and expressed as the mean ±standard error, this is not the case in Figure 4.
5. Although, based on the data depicted in Figure 1, the authors made conclusion that carboplatin resistance in 7T cells is associated with reduced induction of apoptosis, in figures 5B (tempol pretreatment) and 6A (salubrinal pretreatment) only results of MTT assay are shown and thus definite conclusion that “cell death mechanism in 7T cells is independent of ROS induction and activation of ER stress”, indicated in the abstract of the article, could not be made. In addition to MTT assay, determination of apoptosis (i.e. cell death) should have been performed, since MTT assay essentially measures cell viability, not cell death, and therefore reflects decrease in cell proliferation which apart from cell death can also be the consequence of cell cycle arrest. Furthermore, MTT assay measures the formation of formazan-product which is the result of cellular oxidative status/mitochondrial activity that could be influenced by compounds used in pretreatment experiments.
6. The authors have not presented a clear-cut evidence that the endoplasmic reticulum stress is not involved in response of 7T cells to carboplatin treatment. It could be that they have not observed this effect in 7T cells because they did not use high enough carboplatin concentration for the treatment of 7T cells. Namely, in majority of experiments they use a single carboplatin concentration for both HEp2 and 7T cells, although they have shown in Figure 1A that two times higher concentration of carboplatin is needed to achieve the same level of apoptosis in 7T cells. Furthermore, although they showed data from several time dependent experiments, the obtained results are sometimes not comparable. For example, in Figure 1B they present PARP and procaspase-3 cleavage, but although in HEp2 cells procaspase-3 cleavage was already observed after 16 h of treatment, in 7T cells procaspase-3 cleavage is not present even after 48 h of treatment. However, at the same time period (48 h of treatment), PARP cleavage was clearly observed in 7T cells, which is a discrepancy that warrants further explanation. In addition, in order to make well-founded conclusion that CHOP RNA silencing „significantly decreased the percentage of apoptotic cell population upon carboplatin treatment in HEp2 cells, while this effect was absent in 7T cells“ (a conclusion drawn from results depicted in figure 8C), the authors should have used higher concentrations of carboplatin for treatment of 7T cells because it is more challenging to observe a significant decrease when there are only 10% of apoptotic cells detected in total cell population. Instead, a comparable fraction of apoptotic cells should have been targeted (around 50%, like in HEp2 cells), and this could have been achieved by treating 7T cells with higher concentration of carboplatin.
7. The authors do not refer to the paper by Rak et al. which shares the same experimental model - human laryngeal carcinoma HEp2 cells and their subline resistant to carboplatin (7T). Namely, in both papers, by Rak et al. (Carboplatin resistant human laryngeal carcinoma cells are cross resistant to curcumin due to reduced curcumin accumulation. Toxicol In Vitro. 2013 Mar;27(2):523-32. doi:10.1016/j.tiv.2012.11.003. Epub 2012 Nov 9.) and by Brozovic et al., it is indicated that the development of the 7T subline resistant to carboplatin has been published previously by Osmak et al. (1995), thus confirming that the same experimental model was used in the research. Here are some comparisons of the results published by Rak et al. and the ones published by Brozovic et al.:
Both papers are dealing with the molecular mechanisms responsible for the resistance of 7T subline to a certain drug – in paper by Rak et al. this drug is curcumin, while in paper by Brozovic et al. the drug is carboplatin.
In both papers a conclusion is made that probably the main mechanism of the resistance to curcumin and carboplatin, respectively, is reduced drug accumulation. Notably, in the discussion section of paper by Brozovic et al. the authors conclude “despite the fact that we did not measure the protein expression of different transporter proteins, our data suggest that altered transport of CBP, especially an increased efflux of Pt out of the cells is probably the main mechanism of resistance to CBP in 7T cells”.
However, while Rak et al. are careful in their estimation of the contribution that the effects downstream from drug accumulation have in the entire mechanism of curcumin resistance, since obviously the difference in all the mechanisms that are downstream can result from the observed difference in curcumin accumulation, Brozovic et al. did not made clear evidence that this is not the case in terms of endoplasmic reticulum stress and carboplatin resistance. Namely, while Rak et al. have determined equitoxic concentrations of curcumin for apoptosis induction, as well as curcumin accumulation in the cells, and used these equitoxic concentrations in further experiments, Brozovic et al. have not made similar experiments on HEp2 and 7T cells in terms of their response to carboplatin and the molecular mechanisms studied. And this should have been performed in order to make clear conclusion that endoplasmic reticulum stress is involved in the response of human laryngeal carcinoma cells to carboplatin but is absent in carboplatin-resistant cells, as indicated in the title of this article.

Tamara Cimbora Zovko, PhD