The detection of trans gene fragments of hEPO in gene doping model mice by Taqman qPCR assay

Introduction

Doping is an act of enhancing competitive abilities to achieve success by using substances or methods prohibited in sports (JADA, 2019). Doping in sports, particularly in festivals such as the Olympic Games and world or local championships for various competitions, is illegal and against the spirit of the game. The World Anti-Doping Agency (WADA) was established in 1999, and has been involved in scientific research on doping, anti-doping education, development of anti-doping strategies, and monitoring of the World Anti-Doping Code (hereafter, the Code) ( WADA, 2019) to ensure the soundness and fairness of competitive sports worldwide.

With the rapid progress of genetic engineering technology and gene therapy, the WADA has been strongly alerted against gene doping. Since its early days, “gene doping” has been on the list of prohibited actions by WADA. In 2004, the WADA created a panel of experts on gene doping to investigate the latest advances in the field of gene therapy and methods for detecting doping (WADA, 2016). In January 2018, the WADA extended the ban on gene doping to include all forms of gene editing. Therefore, the list of prohibited substances currently includes “gene editing agents designed to alter genome sequences and/or the transcriptional or epigenetic regulation of gene expression” (WADA, 2018). Numerous gene editing systems have been established, including virus vectors (68.3%), CRISPR/CAS9 (0.4%), and short interfering RNA (0.2%), among others. In these systems, naked plasmids (15.4%) are used for gene therapy (Gene therapy Clinical Trials, 2019). They are safe compared to virus vectors and are most commonly used in gene therapy clinical trials because they do not cause infection. Naked plasmids have been used in gene therapy to treat human diseases such as obstructive arteriosclerosis and hemophilia (Shigematsu et al., 2010; Roth et al., 2001). In, generally, naked plasmids in vivo were hardly transferred to targeting cells and organs. However, recently, naked plasmids induced technique is improving (Hu et al., 2017; Ito et al., 2019; Kanemura et al., 2008). Additionally, gene doping and gene therapy are linked in terms of enhancing or silencing of gene expression, resulting in alterations in biological functions. Thus, they may be used for gene doping. Additionally, there are no established standard methods for detecting or preventing gene doping by naked plasmids.

In recent years, functional genes that enhance exercise performance were reported. Particularly, human erythropoietin (hEPO) was brought to the attention of the WADA. hEPO is a glycoprotein hormone mainly produced by peritubular fibroblasts in the kidney and act on bone marrow to stimulate the production of red blood cells (Jelkmann, 2011) Red blood cells can deliver oxygen throughout the body, and thus increase exercise performance (Haile et al., 2019). Endogenous hEPO production is strongly regulated by hypoxia (Jelkmann, 2011; Baker & Parise, 2016). A previous study showed that high-altitude training increases hEPO levels and improves exercise capacity by increasing red blood cell and hematocrit levels (Park et al., 2016). Therefore, the WADA added hEPO to the prohibited doping list as peptide hormones, growth factors, related substances, and mimetics. While a method is available for evaluating exogenous hEPO, there is no established method for detecting increased production of endogenous hEPO by gene editing. Therefore, we developed a detection method for detecting transgenes delivered by injection of naked plasmids in a mouse model that mimics gene doping.

Materials & Methods

Animal experiments

Animal experiments in this study were approved by the Animal Care Committee, University of Tsukuba (approval numbers: 19-163 and 19-425). An overview of the experiments is shown in Fig. 1. Four-week-old C57BL/6 male mice, weighing 10.3–13.9 g, were purchased from the Central Laboratories for Experimental Animals (Tokyo, Japan). The mice were bred and maintained in an air-conditioned animal house under specific pathogen-free conditions and subjected to 12∕12−h light and dark cycles. The mice were fed standard mice pellet and water ad libitum. When the mice were 5 weeks of age, 50–100 μL whole blood was collected from the tail tip from an approximately one mm cut into a microtube containing EDTA 2Na while the mice were under inhalation-general anesthesia by isoflurane. One piece of stool was also collected from the mice. The mice were returned to normal breeding after blood collection. Blood and stool samples were stored at −20 °C until further analysis. After 3 weeks, when the mice were aged 8 weeks, plasmids containing hEPO gene were injected as described below.

Results

TaqMan qPCR specifically and accurately detected the transgene

After confirming the specificity and accuracy of the Taqman qPCR assay for detecting trans-gene fragments, a specific amplification curve was detected in pooled blood-DNA and stool-DNA samples, while no amplification was observed in any human cell lines or DW as the negative controls (Fig. 2). In agarose gel electrophoresis, a single band for the amplicon was detected in all whole blood-DNA samples by Taqman qPCR (Fig. 3). In contrast, stool-DNA samples showed two bands for the amplicon, indicating that one was the hEPO gene and the other was a non-specific amplicon. No amplicon band was detected in any of the human cell lines or DW as negative controls (Fig. 4). Further analysis to confirm the specificity and accuracy of Taqman qPCR was conducted by Sanger sequencing. In the whole blood samples, a separated single peak was detected in the IV, IM, and IP specimens (Fig. 5). In the stool samples, no single peak was detected in any groups (Fig. 4). To confirm the specificity and accuracy of the amplicon, Sanger sequencing was conducted. In whole blood samples, separated and single peak sequences were detected in the IV, IM, and IP specimens. In the stool samples, no single sequence was detected in any group. Concordances between hEPO-reference and amplicon sequences in all samples were high (Figs. 4 and 5).

Discussion

In recent years, gene editing systems have been improved and developed for gene therapy and gene therapy for various diseases is being evaluated in clinical trials (ex. critical limb ischemia, soft tissue sarcomas, retinal dystrophy, muscular dystrophy) (Shigematsu et al., 2010; Xiao et al., 2017; Russell et al., 2017; Mendell et al., 2015). However, with the development of gene therapy technology, the risk of gene doping is increasing. Particularly, hEPO is a target gene for gene doping because it may improve endurance capacity and exercise performance (Haile et al., 2019; Durussel et al., 2013). An hEPO formulation has been developed for doping which shows a different molecular weight from endogenous hEPO, enabling its detection (Salamin et al., 2018). However, detection remains difficult because there is no difference between the genome product and transgene product. Thus, we aimed to detect the transgene in a gene doping mouse model using naked plasmids.

We developed a method for detecting the trans-gene delivered by a naked plasmid coding for hEPO and examined the differences in detection sensitivity depending on the route of administration, as various organs may be targeted by the gene. We detected the transgene fragment following IV, IP, or IM injection by two qPCR methods and TaqMan assay. Tozaki et al. (2018) reported that plasmid injection into the muscle could be detected by droplet digital PCR (ddPCR) until 2–3 weeks after injection in micro mini pigs. In contrast, hEPO transgene fragments were detected in IP blood samples until 48 h after injection in this study. There are three reasons for this difference: (i) the injection amount was lower than that used in the previous study (previous: 250 µg, this study: 50 μg); (ii) species differences (metabolic speed can differ because of enzyme activity differences); and (iii) the PCR method and sensitivity show some differences (previous: ddPCR/high sensitivity, this study: Taqman qPCR/normal). However, ddPCR is more costly than Taqman qPCR, and thus ddPCR cannot be used to analyze large numbers of samples. In this study, we compared the Taqman qPCR and SYBR Green PCR assay to detect gene doping. As a result, Taqman qPCR detected higher levels of the gene than the SYBR Green assay and for a longer period of time. Therefore, this model can be used for gene doping detection using naked plasmids.

Based on the results of qPCR, stool samples are useful for gene doping inspection because they can be collected non-invasively. The collection of blood samples is invasive; however, transgene fragments were detected from a smaller sample volume. Hence, blood and stool samples are useful for this detection.

We evaluated the accuracy of the TaqMan assay by three methods using blood and stool samples, which confirmed the presence of (i) specific amplifications in the TaqMan assay, (ii) transgene molecular size by electrophoresis, and (iii) transgene DNA sequence by Sanger sequencing. The hEPO amplification products and DNA sequence agreed with the reference in all blood samples. In contrast, in the stool sample, several non-specific bands were observed in agarose gel electrophoresis and waveform disturbances were detected in the Sanger method. The stool sample may have contained a high level of enterobacteria DNA. Although Sanger sequencing did not provide accurate results, TaqMan qPCR was specific and thus may be adapted for transgene detection.

Primer dimers formed in SYBR Green reactions cause decrease the sensitivity of real-time PCR. This problem can be avoided by using a specific detection system with TaqMan MGB probes, where non-specific amplification is not detected because of the binding of the specific MGB probe only to a 100% complementary nucleotide sequence (Orlando, Pinzani & Pazzagli, 1998).

There was a limitation to this study. We did not examine protein expression levels, the number of red blood cells, or hemoglobin levels. Thus, whether the plasmid functioned after injection remains unclear and its gene-doping require further analysis. However, we detected the transgene in blood and stool samples in a specific and accurate manner. These results will contribute to the development of gene doping detection methods.

Conclusions

In the present study, a transgene fragment was detected in whole blood DNA and stool DNA samples by qPCR. Particularly, TaqMan qPCR is specific and accurate compared to the SYBR Green assay. Additionally, stool and whole blood sample collection is relatively non-invasive and can be used for gene doping inspection. These results may lead to the development of standard detection methods for gene doping and deter the use of gene doping by athletes.

Supplemental Information