Victor,

Within our article, we concluded that: 1) BORIS mRNA is undetectable in most human breast cell lines and high grade tumors by RT-PCR, 2) levels of BORIS mRNA are thus miniscule compared to those of CTCF mRNA, and 3) amplification of chromosome 20q13 in specific cell lines has not led to detectable increases in BORIS expression. From your comments, it is clear that we have not convinced you of these. As these findings are discordant with work published from your own lab, I am not surprised that you have questions. I do disagree with your introductory comment that “it is generally accepted that proving a negative result is nearly impossible, yet the paper uses this as a punch line.” We are not telling a joke, but communicating a finding that we feel is important to this field. Obtaining a significant negative result is not impossible, just historically difficult to publish.

As you, and we in our article, have discussed, there are several published articles describing BORIS expression in breast cells/tissues. Most comparable to ours, in that BORIS was evaluated by RT-PCR, are 1) D’Arcy et. al. [1] (although in this report, the RT-PCR analysis was constrained to a limited number of clinical samples), and 2) Vatolin et. al. [2], (in which BORIS was analyzed by RT-PCR in 11 primary breast tumors and 8 breast cell lines, as well as cell lines and tumors of other tissue types). We agree with you that it is easier to compare assays using cell lines as a measure of comparison; which is why we chose to contrast our data to that reported by Vatolin et. al. I must point out that the article you referenced to support the claim of abnormal BORIS expression in breast cancer; i.e., “…the levels of BORIS transcripts were significantly greater than zero in 88.5% of 87 breast cancer patients (D'Arcy, Abdullaev et al. 2006),” reports the expression of BORIS within leukocytes from patients with breast cancer, and not within the patients’ tumors - a key distinction you omitted.

You have argued that the number of cell lines and tumors we analyzed is inadequate to substantiate our claims. However, the number of breast cell lines and tumor samples evaluated by Vatolin et. al. (which you reference to support the expression of BORIS in breast cancer/lines) is roughly equivalent to the numbers of samples we analyzed. Also, you have incorrectly stated that we claim two of 11 breast cancer cell lines to be positive. To clarify, we observed one of 10 breast cancer lines to be positive. This was MDA-MB-436, which was found to express BORIS near the limit of detection, at 0.033% of TBP. The other breast cancer cell lines we analyzed were: MDA-MB-231, MCF-7, T47D, SUM185PE, UACC-812, AU-565, BT474, SK-BR-3, and MDA-MB-453. As we discussed in our paper, MCF10A and MCF12A are derived from non-malignant tissues, and MDAMB435 is considered to be melanoma-derived. Nevertheless, we feel our sample size does support our claims.

Another of your criticisms is that we “must have been aware that BORIS is aberrantly activated in cancers but expressed at low levels, similarly to many other CTAs (cancer testis antigens)” as published by Simpson, Caballero et al. [3]. In fact, we were aware of this publication. There is, however, a significant difference between “being expressed at low levels” and being “completely undetectable by RT-PCR.” In fact, the claim that CT antigens are expressed at low levels refers to data from their earlier work [4] in which these authors screened/measured CT expression in normal, disease-free, tissues by endpoint RT-PCR. They did find BORIS to be expressed at greatest levels in testis>pancreas>prostate>thymus>kidney. Notably, these authors used a total of 35 PCR cycles in their assay. As we discuss briefly below, there is a detection limit of PCR, and one risks detecting noise/contaminating DNA when an extreme number of PCR cycles is used, especially when qPCR is not used to detect the amplification.

Within our article, we have described at length all controls, the sensitivity, and specificity of our assays. One of your critiques however, is that “these primers and/or RNA preps were likely inadequate, because the Ct value (27.35) for BORIS expression in testis (Table 1) did not come even close to Ct value around 18-22 one routinely gets for 100 ng of total testes RNA.” However, the TBP Ct value from this sample indicates that both the RNA and RT reaction were adequate – a point you raise regarding our clinical samples. Therefore, I fail to see how you come to this conclusion. Just above, you criticized the conclusions we have made from the 13 cell lines and 8 tumors (as well as all the controls we have used) because the sample size was too small in your opinion, although a single Ct value is enough for you to draw conclusions on the entirety of our data. Since you have experience with real-time PCR, you should know that Ct values are dependent on many factors, and you shouldn’t make this type of conclusion based solely on Ct values without normalization. Specifically, this is clinical material, which may itself be enough to explain a different Ct value due to a possible low expression of BORIS in this patient’s testis. In addition, different real-time PCR machines, PCR primer sets, and PCR chemistries will also lead to a variance in Ct values. In fact, all software used by the various real-time PCR instruments allows the user to analyze amplification profiles post-run so that baseline amplifications and thresholds can be set accurately. This too will lead to raw Ct values that vary among users. In addition, the data in our publication was obtained using Taqman probes; using probes will typically push back Ct values 2-5 cycles when directly compared to Sybr Green chemistry (depending on the specific primer set and probe/quencher used). Are the values you are quoting from Sybr Green reactions (do you have a reference for the Ct values which you are referring)? Although not described within our manuscript, we have also evaluated BORIS expression in several breast cell lines using SYBR Green chemistry. In these experiments, the average Ct values for TBP were roughly 21 cycles whereas BORIS signal was still undetected. In addition, we were able to detect endogenous BORIS in 5’aza-dC treated MCF-7 cells (a control you suggested), as well as in all positive controls. I’m sorry, but your argument that our technique is questionable just doesn’t hold water.

I feel compelled to comment on your reference to the article by Compérat et al, 2007 [5]. You stated that most people report TBP values of 25-26, that “some scientists consider the tissue samples to be eligible for evaluation only if the TBP Ct value was between 25 and 26, suggesting an appropriate starting amount and quality of total RNA. (Compérat et al, 2007)”. The “some scientists” refers only to Compérat et al, 2007 [5]. It is noteworthy that these investigators used SYBR Green chemistry to validate their cut-off, and this is the level they considered to be valid for their equipment/ experiments. However, no objective data were presented that would justify generalizing their criterion. You may also want to consider that this same group of authors published data in which their reported TBP values were between 28-29 [6]. Also, our TBP values do fall inside the range reported by Applied Biosystems (between 25-30 when using 100ng - 2ug total RNA). Most importantly, the slope of our standard curve for BORIS was -3.3219, which means the amplification efficiency was 100% over a 6 log dynamic range (typically the Cts are between 18-35). In the next lowest dilution, only 1/4 replicates produced signals, as would be expected. This represents the assay’s limit of detection. As we stated in our manuscript, this assay was sensitive enough to detect down to a 24 copies of BORIS. When you amplify past this linear range, any detected signal will not be correlated to the amount of input template. Thus, one should be careful in interpreting data using an extreme number of PCR cycles; i.e., 40-45 cycles. This was the key driving force behind the development of the quantitative real-time PCR method.

I must also disagree that the depth of analysis is inadequate. We have used a stringent set of controls, used 4 PCR assays to analyze BORIS expression, and have found that the level of expression is not consistent with other reports in the literature. One other important finding that you left out of your critique was that we find the levels of BORIS (undetectable in our case) pale in comparison to the abundant levels of CTCF (above the levels of TBP) in these same samples. This is clearly not consistent with the endpoint PCR data presented by Vatolin et.al (figure 1), which show BORIS mRNA levels are equal to, and in some cases greater than those of CTCF in the breast cell lines and tumors. Even if your comments regarding our inability to detect a scant level of BORIS were valid, this discrepancy would still remain. Notably, Renaud et.al. [7] concluded that that CTCF acts as a repressor of BORIS transcription, and our data within both breast samples and testis are in accordance with this model.

Based on our results, we do believe that BORIS is not expressed in most human breast cell lines and high grade breast tumors. We do not claim that these findings for breast can be extended to other cancer types, as we did not critically evaluate the levels in any other tissue but breast. Let me assure you that we did not “jump to these conclusions,” as they were made after a long and thorough investigation.


1. D'Arcy V, Pore N, Docquier F, Abdullaev ZK, Chernukhin I, Kita GX, Rai S, Smart M, Farrar D, Pack S et al: BORIS, a paralogue of the transcription factor, CTCF, is aberrantly expressed in breast tumours. Br J Cancer 2008, 98(3):571-579.

2. Vatolin S, Abdullaev Z, Pack SD, Flanagan PT, Custer M, Loukinov DI, Pugacheva E, Hong JA, Morse H, 3rd, Schrump DS et al: Conditional expression of the CTCF-paralogous transcriptional factor BORIS in normal cells results in demethylation and derepression of MAGE-A1 and reactivation of other cancer-testis genes. Cancer Res 2005, 65(17):7751-7762.

3. Simpson AJ, Caballero OL, Jungbluth A, Chen YT, Old LJ: Cancer/testis antigens, gametogenesis and cancer. Nat Rev Cancer 2005, 5(8):615-625.

4. Scanlan MJ, Simpson AJ, Old LJ: The cancer/testis genes: review, standardization, and commentary. Cancer Immun 2004, 4:1.

5. Comperat E, Bieche I, Dargere D, Ferlicot S, Laurendeau I, Benoit G, Vieillefond A, Verret C, Vidaud M, Capron F et al: p63 gene expression study and early bladder carcinogenesis. Urology 2007, 70(3):459-462.

6. Bieche I, Laurendeau I, Tozlu S, Olivi M, Vidaud D, Lidereau R, Vidaud M: Quantitation of MYC gene expression in sporadic breast tumors with a real-time reverse transcription-PCR assay. Cancer Res 1999, 59(12):2759-2765.

7. Renaud S, Pugacheva EM, Delgado MD, Braunschweig R, Abdullaev Z, Loukinov D, Benhattar J, Lobanenkov V: Expression of the CTCF-paralogous cancer-testis gene, brother of the regulator of imprinted sites (BORIS), is regulated by three alternative promoters modulated by CpG methylation and by CTCF and p53 transcription factors. Nucleic Acids Res 2007, 35(21):7372-7388.