Sensitive and precise regulation of haemoglobin after gene transfer of erythropoietin to muscle tissue using electroporation


Electroporation-based gene transfer (electro gene transfer (EGT)) is gaining increasing momentum, in particular for muscle tissue, where long-term high-level expression is obtainable. Induction of expression using the Tet-On system was previously established; however, attempts to reach a predefined target dose – a prescription, have not been reported. We set three target haemoglobin levels (10, 12 and 14 mmol/l, base level was 8.2 mmol/l) and aimed at them by transferring the erythropoietin (EPO) gene to mouse tibialis cranialis (TC) muscle, and varying (1) DNA amount, (2) muscle mass transfected and (3) induction with the Tet-On system. Results showed that (a) using GFP, luciferase and EPO low DNA amounts were needed. In fact, 0.5 μg of DNA to one TC muscle led to significant Hgb elevation – this amount extrapolates to 1.4 mg of DNA in humans, (b) three prescribers hit the targets with average Hgb of 10.5, 12.0 and 13.7 mmol/l, (c) different approaches could be used, (d) undershooting could be corrected by retransferring, and (e) overshooting could be alleviated by reducing dose of inducer (doxycycline (dox)). In conclusion, this study shows that using EGT to muscle, a preset level of protein expression can be reached. This is of great interest for future clinical use.


Gene transfer by electroporation is gaining increasing momentum as a non-viral approach in gene therapy. Electro gene transfer (EGT) is obtained when an external electric field, which transiently destabilises the membrane, renders the cells accessible for cDNA entry.1 Recent developments in the EGT procedure have provided highly efficient pulses with minimal perturbation of cell physiology and membrane integrity, when using a combination of one high-voltage (HV) pulse followed by one low-voltage (LV) pulse.2, 3, 4

Electroporation has been used for several years in the clinic for enhanced drug delivery to tumours, showing that electroporation-based treatments are feasible and safe in the clinical setting.5, 6, 7 Moreover, the first clinical trials with EGT to tumours have been undertaken.8 EGT to skeletal muscle is particularly advantageous since the muscle is capable of producing large amounts of the transgenic product and can act as an artificial endocrine organ.9

There are several reports on successful in vivo EGT of erythropoietin (EPO) into skeletal muscles in both mice10, 11, 12 and rats,13, 14 and EPO expression has been detected more than a year after EGT in mice.15 The delivery of EPO was followed by a marked increase in haematocrit (Hct) levels both in normal and uraemic rats.16, 17, 18 In these studies, EPO expression was controlled by constitutive promoters, which induced high expression in the first months after transfection, followed by a reduction in the expression level to a lower steady-state phase.19

In mice, the half-life of erythrocytes is 12 days compared to 120 days in humans.20 Month-long transgenic EPO expression does, therefore, have great regulatory effect on the Hgb level. Thus the EPO–Hgb system is an excellent model for studying regulation of transgene expression in an inducible promoter system as the effect and control of EPO expression is easily measured from blood samples, which can be taken repeatedly.

In the clinical setting, control of gene expression is of utmost importance. Systems to induce expression of transgenes have previously been established,21 and the most utilised of these seems to be the Tet-On system, which uses doxycycline (dox) as inducer drug.22 The Tet-On system comprises of a chimeric transactivator, which activates transcription from a silent tetO promoter in the presence of dox in a dose-dependent manner.22 The system has been hampered by high basal activity; however, this has been solved by two approaches: the development of a transcription silencer23 and new generations transactivators.24, 25 The inducer molecule dox is an orally available antibiotic, which induces gene expression in the range of 0.01–0.2 mg/ml. This dosing regimen for induction of gene expression is comparable to levels used in antibiotic therapy.21 It has been shown that expression can be induced, turned-off and re-induced at a later time.24 Other variables affecting the level of gene expression could be the amount of DNA transferred, but very few dose-dependent studies have been carried out as far as DNA dose is concerned. Finally, electroporation makes it possible to delineate precisely the area to be transferred, as transfer beyond the electrodes is minimal.26 However, only few data on ‘dose’ with respect to amount of transferred tissue are available.

As long-term high-level gene expression can be obtained after EGT,15, 27 treatment of, for example, protein deficiency syndromes is envisaged. However, to proceed with such a strategy, precise regulation of expression with options for changing level of expression both upward and downward is paramount.

We therefore engaged in an experiment to explore the possibilities of regulating expression of electro gene transferred genes by (1) amount of DNA, (2) area of muscle transfected, and (3) induction of promoter. Before starting with the actual ‘prescription’ experiment the influence of the three different variables were examined.


Optimisation of plasmid usage for local gene expression

The initial studies for optimising the amount of pTetO-Luc (plasmid encoding luciferase under a tetO inducible promoter) and pTet-On (plasmid encoding the tetO transactivator) needed for maximal gene expression, were performed by evaluating luciferase expression in muscles transfected with pTetO-Luc and pTet-On in C57Black/C mice. We found that 10 μg of each plasmid per muscle was sufficient to ensure maximal inducible gene expression (Figure 1a); however, a significant reduction in the gene expression response (P<0.05, Student's t-test) was not observed until the plasmid amount was decreased to 2.5 μg. Using enhanced green fluorescent protein (EGFP) for qualitative description, the plasmid amount could be reduced to 5 μg, before the maximal gene expression starts to decrease (Figure 1a). Importantly, the luciferase plasmid (9.5 kb) is almost twice as big as the EGFP plasmid (5.1 kb), this means that in molar concentrations, similar amounts (1.6 pmol) of both plasmids were needed for maximal gene expression.

Figure 1

Optimisation for pTetO-EGFP amount for maximal gene expression. (a) Combinations of three plasmids pTet-On, pTetO-Luc and pTet-tTS in the ratio 1:1:0.5 were electrotransferred into TC muscles by combinations of HV and LV pulses. The amount of pTet-On and pTetO-Luc is given in the figure. The control group was electrotransferred with 5 μg pTetO-Luc alone. All mice received drinking water containing 0.2 mg/ml dox from the day of treatment. The bars show the mean Luc expression (n=12), while the pictures show representative samples of EGFP transfected muscles. There was no significant difference between the groups (n=12) treated with 5–20 μg (P>0.05), while there was significant difference between the groups treated with 2.5 and 10 μg (P<0.05) of each plasmid according to Student's t-test. (b) Combinations of three plasmids, pTet-On (5 μg), pTetO-Luc (5 μg) and pTet-tTS (0–10 μg) were electrotransferred into TC muscles. The control group was injected with ptetO-Luc alone. Animals were given drinking water with (light grey bars) or without (black bars) 0.2 mg/ml dox. Luciferase expression was evaluated by luminometry 48 h after treatment. The pictures show representative EGFP expression in muscles derived from the corresponding groups, while the bars represent the mean and standard deviation of 5–6 muscles (no dox) and of 12 muscles (with dox). Dox induction was significant in all groups tested (P<0.01, Man–Whitney rank test). Inhibition with 5 μg pTetO-Luc was significantly more effective than 2.5 μg pTet-tTS (P<0.05, Man–Whitney rank test), which caused no significant reduction in luciferase expression (P>0.05, Man–Whitney rank test). No further inhibition was achieved by increasing from 5 to 7.5 or 10 μg pTet-tTS.

In the non-induced state, we found high basal activity in the Tet-On system as has been reported previously by others.11, 23, 28 Addition of pTetS (plasmid encoding a silencer) to the injected plasmid solution reduced the basal luciferase expression in a dose-dependent way. pTetS (2.5 μg) reduced the basal activity, limiting activity to a few muscle fibres (Figure 1b), whereas luciferase expression was completely eliminated using 5, 7.5 or 10 μg pTetS. In the dox-induced state, pTetS had no effect on luciferase expression at any concentration, where expression stayed high. Studies using EGFP confirmed these results (Figure 1b).

When using the Tet-On system, the level of gene expression can be regulated by dox in a dose-dependent manner.22 Maximal gene expression of EGFP was achieved using 0.2 mg/ml. Reducing dox to 0.1 mg/ml resulted in 50% of the maximal gene expression and 0.01 mg/ml dox in the drinking water resulted in 25% expression (data not shown).

Dose–response correlation between the electrotransferred plasmids and haemoglobin levels

To determine the minimal effective dose of EPO plasmid, C57Black/C mice (n=10) were electrotransferred with 1, 2.5 or 5 μg of pUHD-10.3–mEPO (plasmid encoding EPO under a tetO inducible promoter), pTet-On and pTetS in either one or both hind legs. Gene expression was maximally induced by 0.2 mg/ml dox (Figure 2). Electrotransfer of 1 and 2.5 μg of the plasmids in one leg led to significant increases in Hgb to 12.6±2.7 and 14.6±2.6 mmol/l, respectively. Interestingly, EGT of 5 μg of each plasmid did not increase Hgb. In fact, the electrotransfer resulted in an initial increase in Hgb to 10.3±2.4 mmol/l, but after day 28 the Hgb levels started to decline and returned to control levels at day 56.

Figure 2

Serum haemoglobin under stable dox induction. Mice were electrotransferred in the TC muscle with 1, 2.5 or 5 μg of pUHD10-mEPO, pTet-On and pTets in either the right leg (left panel) or both legs (right panel) and induced with 0.2 mg/ml. Each group consisted of 10 mice and the development in median Hgb levels with s.d. values is depicted. Electrotransfer with 1 μg (P=0.0027) and 2.5 μg (P=0.014) induced significantly higher Hgb levels than the control mice. There was no significant difference in Hgb between the mice transfected with 1, 2.5 or 5 μg. Similar patterns were observed after electrotransfer to both legs with significantly higher Hgb levels than the control level after transfer of 1 μg (P=0.00016) and 2.5 μg (P=0.00009), respectively, but no significant difference between the two or when compared to electrotransfer with 5 μg.

Electrotransfer of 1, 2.5 or 5 μg of each plasmid to both legs showed same pattern as transfer to one legs. One and 2.5 μg both induced significant increases in Hgb to 15.9±2.9 and 15.6±2.1 mmol/l, respectively, while electrotransfer of 5 μg induced an initial increase in Hgb, followed by a decline to control level after day 28. The levels of Hgb were slightly higher than those after transfer to one leg, but not as high as might be expected from transfer of the double amount of plasmid.

Highly elevated Hgb levels with long-term dox induction

To determine the duration of increased Hgb levels, groups of 10 mice were electrotransferred with 1 μg of pUHD-10.3–mEPO, 1 μg pTet-On and 1 μg pTetS in both legs, or simply injected with the same amounts of plasmid. All mice received 0.2 mg/ml dox throughout the experiment for maximal expression. In the EGT-treated group, the Hgb increased to 14.2±2.5 mmol/l and this level was maintained for 85 days, hereafter the median Hgb slowly decreased (Figure 3). The decrease in median Hgb was caused by some mice with rapidly declining Hgb (initial responders), while others maintained the elevated Hgb (responders). The Hgb stabilised at 9.5±2.5 mmol/l, which is the mean of responders and initial-responders. The actual mean level of the responders was 13.3±2.4 mmol/l, and expression in this subgroup was detected for at least 257 days. The observed decreases in Hgb were correlated with low serum iron levels (see below). DNA injection alone induced an initial increase in Hgb, which was lost within the first weeks. This increase might be due to transient transfection of single muscle fibres following the direct DNA injection.

Figure 3

Long-term elevation in haemoglobin under stable dox induction. After electrotransfer of 1 μg in both legs, mice were administrated 0.2 mg/ml dox drinking water and followed for 256 days. The transfected animals were divided in responders (n=5, closed circles) and initial responders with decreasing Hgb levels (n=3, open circles) and median Hgb with s.d. values are depicted. The DNA injected and control groups consisted of eight mice.

Tight correlation between serum EPO level and haemoglobin level during dox regulation

To characterise the responsiveness in gene expression towards dox administration, two groups of four Wistar rats were electrotransferred with 2.5 μg of pUHD-10.3-mEPO, pTet-On and pTetS. Group A was induced with 0.2 mg/ml dox for 8 weeks, followed by 7 weeks of induction with 0.01 mg/ml dox, while group B did not receive dox. From week 15, both groups (A+B) were induced with 0.05 mg/ml dox (Figure 4). A non-transfected control group (C) was also included, receiving dox throughout the experiment. During the first induction at 0.2 mg/ml in group A, the Hct levels increased significantly (P=0.017) in all four rats ranging from 49 to 74 (9.9–15.5 mmol/l) with a control level of 42.1±1.1 (9.1±0.44 mmol/l). Reducing the dox concentration to 0.01 mg/ml led to a significant drop in the Hgb with three rats returning to control level (Hct: 46±8.2, Hgb: 9.1±0.44 mmol/l), while the Hgb of one rat remained elevated. From week 15, when gene expression was induced with 0.05 mg/ml dox in all eight transfected rats, a secondary increase was observed in group A, which had already been induced, while a significant increase (P=0.0024) was observed in two of the four rats in group B, which were induced for the first time. Serum EPO levels were measured in all animals and the levels were closely correlated to the Hct levels (Figure 4). In fact the serum EPO level in the two non-responder rats was not detectable, explaining why no significant increases in Hct levels were observed.

Figure 4

Correlation between Hct and serum EPO during dox regulation. Groups of four rats (A+B) were electrotransferred with 2.5 μg plasmid, while group C served as control. Groups A and C were induced with 0.2 mg/ml dox for the first 8 weeks, whereafter the dox concentration was decreased to 0.01 mg/ml. From week 15, all three groups were induced with 0.05 mg/ml. The curves depict the development in mean Hct with s.d. values, while the table shows the serum EPO concentration (pg/ml) measured at the indicated time points. As predicted, there was a tight correlation between the serum EPO concentrations and the Hct.

Dox drinking water intake

Dox drinking water was administrated without sucrose to ensure normal levels of drinking;29 indeed the mice receiving dox drank between 4.0 and 5.7 ml/day, which was not significantly different from the control groups (4.1 ml/day). There was no correlation between the drinking water intake and the dox concentration in the drinking (R2=0.11); in fact the drinking water intake was more closely correlated to the median Hgb level in the groups (R2=0.61). This is not surprising as thirst is partly controlled by haemoconcentration.

Treatment strategy for obtaining preset haemoglobin levels

The experiment presented in Figure 5 was performed as a prescription experiment where the three authors were to prescribe an EGT treatment to obtain a level of 10, 12 or 14 mmol Hgb/l, respectively. The authors could control the following parameters: the transfected area (one or both tibialis cranialis (TC) muscles), the plasmid amount, the administered dox concentration and finally they could prescribe re-transfection.

Figure 5

Prescribing EGT for precise control of Hgb levels. Three ‘doctors’ prescribed EGT treatments for obtaining 10, 12 and 14 mmol Hgb/l, respectively. Each group consists of eight mice and the median Hgb+s.d. is depicted. (a) For obtaining 10 mmol Hgb/l, group 1 was transfected with 1 μg/leg, group 2 with 0.25 μg/leg, while group 3 was transfected with 0.25 μg/2 legs and re-transfected at day 64 with 0.5 μg/leg. Dox prescription is given in the figure with each prescription period lasting 4 weeks. (b) For obtaining 12 mmol Hgb/l, group 1 was transfected with 2 μg/leg, group 2 was transfected with 1 μg/leg and group 3 was transfected with 0.5 μg/leg. (c) For achieving 14 mmol Hgb/l, group 1 received 1 μg/2 legs, group 2 received 3 μg/leg and group 3 with 1.5 μg/leg. For full prescriptions, see the text.

For 10 mmol/l (Figure 5a), small amounts of plasmids, 1 μg/leg, 0.25 μg/leg and 0.25 μg/2 legs, respectively, were electrotransferred. In group 1, which was electrotransferred 1 μg/leg followed by dox induction at 0.1 mg/ml, a significant increase in Hgb to 13.9±1.17 mmol/l was observed after first prescription. Turning down the dox concentration to 0.05 mg/ml did not alter the Hgb level; however complete withdrawal of dox led to a decrease to control level. Thus 1 μg induced too high increases in Hgb, but importantly the Hgb could be reduced to control level by withdrawing the dox. In the other two groups with smaller amounts of DNA transferred, no significant increases in Hgb were observed after first prescription. Two strategies were employed to amend this. In group 2, the dox concentration was increased first to 0.2 mg/ml dox, and then 0.3 mg/ml dox; however this was not enough to reach 10 mmol/l at any time. In group 3, the mice were re-transfected at the third prescription with 0.5 μg plasmids/2 legs followed by dox induction at 0.2 mg/ml dox; this led to an increase in Hgb to 10.8±1.03 mmol/l.

For 12 mmol/l (Figure 5b), slightly higher concentrations of plasmid were used with submaximal dox concentration to regulate the gene expression through the dox administration. Group 1 was electrotransferred with 2 μg/leg followed by dox induction at 0.1 mg/ml. This led to an increase in Hgb to 14.2±0.95 mmol/l. Reducing the dox concentration to 0.05 mg/ml led to a decrease in median Hgb to 12.5±2.57 mmol/l, which was close to the predetermined value. Continuous induction at 0.05 mg/ml was not sufficient, however, to maintain the elevated Hgb level, as the Hgb dropped to control level (8.25±2.89 mmol/l). Group 2 was electrotransferred with 0.5 μg/leg with dox induction at 0.1 mg/ml, leading to an increase in Hgb of 11.0±1.60 mmol/l, increasing the dox concentration to 0.2 mg/ml dox at the second prescription resulted in exactly the target of 12.0±1.60 mmol/l. This level, however, could not be maintained with continuous induction with 0.2 mg/ml dox. Group 3 was electrotransferred with 1 μg/leg with dox induction at 0.05 mg/ml, resulting precisely in the 12.0 (±1.87) mmol/l. Despite continuous induction with 0.05 mg/ml dox a minor reduction in Hgb was observed.

The averages of the three treatment strategies were 12.4 mmol/l after the first prescription, and precise 12.0 mmol/l after the second prescription. However after the third prescription, the mean Hgb was reduced to 9.88 mmol/l as none of the groups were capable of maintaining the elevated Hgb.

For 14 mmol/l (Figure 5c), higher amounts of plasmids were electrotransferred but submaximal dox concentrations were still administered to give room for adjustments. Group 1 was electrotransferred with 3 μg/leg followed by dox induction at 0.1 mg/ml. This led to an increase in Hgb to 13.5±0.54 mmol/l. Increasing the dox concentration to 0.2 mg/ml led to an increase in Hgb to 15.2±3.26 mmol/l. This level was maintained throughout the experiment. Group 2 was electrotransferred with 1.5 μg/leg followed by dox induction at 0.1 mg/ml. This led to an increase in Hgb to 13.2±1.58 mmol/l. Increasing the dox concentration to 0.2 mg/ml did not increase the Hgb level. At 12 weeks, the Hgb dropped off sharply nearly to control level (9.0±2.29 mmol/l). This could, however, be explained by low serum iron levels (see below). Group 3 was electrotransferred with 1 μg/2 legs followed by dox induction at 0.2 mg/ml. This increased Hgb to 14.5±1.17 and this level was maintained throughout the experiment.

The averages for the three strategies were 13.7 and 14.5 mmol/l after the first two prescriptions, respectively, showing precise regulation within the range of the preset Hgb values.

Loss of EPO responsiveness after continuous stimulation

Most mice were capable of maintaining the elevated Hgb levels; however some mice lost this ability over time. Possible explanations of this loss of responsiveness towards EPO transfer could be serum iron deficiency or silencing of transgenic EPO transcription. In investigating these possible explanations, mice from the prescription experiment were divided into a group with decreased Hgb level (mean=8.4±0.48 mmol/l), a group with elevated Hgb (mean=12.8±1.38 mmol/l) and the control group (mean=8.3±0.32 mmol/l).

Serum iron levels

Serum iron contents were measured from terminal blood samples. The serum iron levels in mice losing the ability to maintain elevated Hgb levels were 15.97±2.23 mmol/l (n=14), which were significantly lower (P=0.0006, Student's t-test) than control level (21.03±1.19 mmol/l, n=4), indicating that the decrease in Hgb could be caused by low serum iron levels. The serum iron levels in mice maintaining the elevated Hgb levels were 25.03±3.72 mmol/l (n=6), which were not significantly different from the control group.

Transcription of transgenic EPO

To determine whether the transcription rate of EPO was altered between the mice with low or high Hgb, Q-PCR was performed on RNA isolated from the transfected muscles. There were no significant differences in the mRNA levels between the two groups (data not shown).


Electroporation is already in use in clinical trials to promote the uptake of chemotherapeutics in malignant tumours (reviewed in Gothelf et al., 200330) and several studies have documented the feasibility of this approach.5, 6, 31 It is likely that this efficient mean of non-viral gene transfer will expand into gene therapy of diseases of organs, as well as organisms. Of particular interest is gene transfer to muscle tissue, as long-term, high levels of expression of the transferred genes can be obtained.15, 32, 33 Gene transfer of, for example, anti-tumour agents to the muscle is a very promising field, while other examples of interesting pre-clinical experiments include gene transfer of EPO for correction of anaemia16, 17, 18 or β-thalassemia.12, 34

Clinical relevance

Being able to regulate expression of the transgene is of utmost importance for clinical use. The long-term expression obtainable is an excellent opportunity to offer treatments for protein deficiency syndromes – but is also a challenge, as the expression of the inserted gene needs to be controlled. In this study, we demonstrate a therapeutic approach with transfer of low amounts of DNA and determination of biological activity.

  • If the level of biological activity is satisfying, the gene expression can be sustained over a long period, for example, regulated through the inducible promoter.

  • If the level of biological activity is too low, new transfers can be undertaken until a satisfactory level is reached.

  • Too high levels of biological activity can be adjusted by deactivating or downregulating the inducible promoter.

  • Finally, the electroporation procedure itself selectively targets the area encompassed by the electrodes, and thus electrode placement can be used as an additional restrictor of the anatomical extent of the transfected host tissue, thus of the number of cells expressing the transgene.

Prescription of precise levels of Hgb

To our knowledge, no attempts have yet been made at devising treatment strategies for the use of EGT to obtain specific levels of the transgene product. We carried out a very simple experiment: by giving the three authors of this paper privileges to prescribe any combination of (1) DNA amount, (2) amount of tissue transfected (one or two muscles), and (3) dose of induction drug (dox), we explored the possibilities of prescribing drugs delivered by EGT with inducible control systems. Interestingly, the prescription experiment showed that several strategies could be employed, resulting in the same range of Hgb. Most interestingly, calculating the average Hgb of the three strategies in each subexperiment, demonstrated a close approximation to the preset Hgb level (Figure 5).

Another advantage of the system is the possibility to regulate stepwise transgene expression. Blood sampling following EGT offers the opportunity to adjust gene expression, while the transgenic expression is proceeding. Undershooting the therapeutic level can be corrected by turning up the dox concentration or re-transferring new DNA amounts, while, more importantly, overshooting the therapeutic level can be amended by withdrawing dox, leading to return of Hgb to control level (Figure 5a). For clinic use, it is of utmost importance that the transgenic expression can be turned on and off – indeed turning off the transgene expression is imperative.

The first parameters to determine in relation to EGT are the DNA amount and the area transfected.

DNA amount

Several factors affect the level of transcriptional activity from a plasmid including the promoter strength.35, 36 Previous studies using the Tet-On system to control EPO production have introduced 10 μg of plasmid.11 We found that it was feasible to reduce the DNA amount to 0.5–1 μg plasmid for obtaining therapeutic levels of Hgb. In fact, exceeding 5 μg led to decreased expression. The decreased efficacy of transferring higher amounts of DNA might be due to local toxicity of the DNA in the muscle. This is supported by our findings that electrotransfer of 2.5 μg in 2 legs caused significant increases in Hgb, while electrotransfer of 5 μg in 1 leg (same total amount of DNA) led to small Hgb increases, which declined to control level within 42 days. It must be noted that no decrease in gene expression was found when introducing higher levels of plasmids encoding GFP or luciferase.

Extrapolating these findings in mice to adult humans suggests that 1.4 mg DNA would be enough to stimulate a significant increase in Hgb in humans. Certainly, this proves the clinical relevance of the EGT, as this amount of DNA is indeed manageable in the clinical setting.

Transfected area

The particular characteristic of EGT is that transfection only occurs in the area injected with DNA and delineated by the electrodes,1 whereas other transfection systems, for example viral vectors and lipofection rely on systemic or intra-organ injections with little control of the tissue and area transfected. In this study, the transfected area was controlled by transfection to one or both TC muscles. However in larger animals defined areas in the muscle could be transfected independently of the rest of the muscle. Thus EGT to a new area within the same muscle will be possible. Another interesting aspect is the possibility of obtaining measurable systemic levels of the transgene product, as well as therapeutic effects, by transfecting a muscle as small as the mouse TC muscle, weighing approximately 50 mg (0.2% of total mouse body weight). This is extremely encouraging for future clinical use of the technology.

Post-transfectional, the level of gene expression may be controlled through dox induction and re-transfection.

Dox induction

The Tet-On system has had problems with high levels of basal activity; however in this study, we were capable of eliminating the basal activity by increasing the concentration of the TetS silencer. A very close correlation was found between dox concentration in the drinking water and transgene expression. In the prescription experiments, it was found that expression levels can be adjusted over a relatively wide dynamic range by changing dox concentrations and that expression can be turned-off completely by withdrawal of dox. Finally, it was shown in another experiment that gene expression can be induced long time after EGT has taken place.11


The potential to re-transfer new amounts of DNA, if efficient levels of gene expression are not obtained, is unlimited. Indeed several studies have shown significant EPO expression and increases in Hgb after re-transfection.11, 12, 13

Transgenic EPO expression

As the main regulator of erythropoiesis, EPO was expected to stimulate increases in Hgb after EGT. Indeed we found that there was a tight correlation between serum EPO level and Hgb level (Figure 4). This close correlation indicates that the transgenic EPO production can supersede the endogenous control of Hgb. In some mice, we observed decreases in the Hgb levels, which correlated with low levels of serum iron. Exhaustion of iron stores is a common side effect of recombinant EPO treatment;37 therefore in future experiments, iron supplementation may eliminate this problem.

In agreement with other studies,18, 27 we found no evidence of quenching of the transgene expression, as has been reported with viral vectors38 due to immune responses.

Less is more: reducing DNA amounts

Our studies with both GFP, luciferase and EPO confirm that increasing the amount of electrotransferred plasmid does not correlate with increasing expression efficacy. We found that reducing the amount of plasmid to 5 μg ensures optimal GFP expression in the muscle, whereas 1 μg was enough to stimulate EPO production resulting in significant increases in the Hgb levels. Generally, other groups have used larger amounts of plasmids, ranging from 10 μg and onwards.39, 40, 41, 42 The rationale for reducing the amount of DNA is both for practical purposes including expenses and administration volumes, as well as the fact that DNA is cytotoxic when introduced in vivo in large amounts.43, 44 DNA-dependent muscle damage with necrotic myofibres and infiltration of inflammatory cells was observed after EGT of 50 μg supercoiled endotoxin-free plasmid DNA.43

Gene transfer to muscle: different requirements depending on aim

In this study, we found that the DNA requirement for obtaining optimal gene expression was different between locally expressed proteins (GFP and luciferase) and the systemically expressed EPO, even though the promoter was the same in the three plasmids. This suggests that for proteins with endocrine/paracrine action, the effect of transgene production is amplified in the body through the signalling cascades induced by the hormone. Therefore EGT of a small amount of DNA encoding a hormone may have significant effect on systemic scale, whereas larger amounts of DNA are required for proteins, whose action is not amplified by the innate signalling pathways.45, 46, 47, 48 For these proteins, the amount of DNA required to obtain a physiological response might be correlated with whether the effect is local or systemic (Figure 6).

Figure 6

Model of DNA requirement for physiological response depending on transfected protein type. Depending on the action of the protein produced different levels of protein production are needed to obtain a physiological response. The level of protein production is determined by the amount of DNA transferred. One μg of the endocrinal EPO was enough for obtaining the most efficient physiological response. For the locally expressed genes (EGFP and luciferase), 5 μg of plasmid was required for maximal local expression, while other studies have shown that larger amounts of plasmid are required for obtaining systemic effect of proteins without endocrine effect.


In conclusion, this study shows that expression of a transfected gene can be tightly controlled, making it possible to reach and maintain a certain preset value. As control is an important prerequisite for bringing the method to the clinic, this study represents an important step towards clinical application of EGT for gene therapy.

Methods and materials

Animals and muscle preparation

All animal experiments were conducted in accordance with the recommendations of the European Convention for the Protection of Vertebrate Animals used for Experimentation and after permission from the Danish Animal Experiments Inspectorate. Mice experiments were performed at University of Copenhagen at Herlev Hospital on 8 to 10 weeks old female C57Black/C mice from Taconic (Tornbjerggaard, Denmark), while rat experiments were carried out at University of Aarhus, Department of Physiology and Biophysics on 4 to 5 weeks old male or female Wistar rats (own breed). The animals were maintained in a thermo-stated environment under a 12-hour light/dark cycle and had free access to food (Altromin pellets, Spezialfutter-Werke, Germany) and drinking water. The animals were killed by cervical dislocation, and the rats were then decapitated. Intact TC without tendons were dissected out and quickly frozen on dry ice and absolute alcohol.

Plasmid constructs and in vivo EGT

The plasmids pTet-On, encoding the reverse tet-responsive transcriptional activator (rtTA) transactivator,49, 50 pTetS, encoding the tS silencer,24 pBI-EGFP, encoding the enhanced green fluorescent protein and pBI-Luc, encoding luciferase, the last two both under the control of an rtTA-dependent promoter, were all obtained from Clontech (Palo Alto, CA, USA). PUHD-10.3-mEPO encoding murine EPO under the control of an rtTA-dependent promoter was kindly provided by Dr Fattori.11 All DNA preparations were performed using Qiafilter Plasmid Maxiprep kits (Qiagen, Germany), and the concentration and quality of the plasmid preparations were controlled by spectrophotometry and gel electrophoresis.

The animals were anaesthetised 15 min before electroporation (EP) by injection of Hypnorm (0.4 ml/kg, Janssen Saunderton, Buckinghamshire, UK) and Dormicum (2 mg/kg, Roche, Basel, Switzerland). For mice, 20 μl plasmid solution was injected intramuscularly (i.m.) along the fibres into the TC muscle using a 29G insulin syringe. The rats were injected with 10 μl plasmid solution in the TC also using an insulin syringe. Plate electrodes with 5-mm gap were fitted around the hind legs. Good contact between electrode and skin was ensured by hair removal and use of electrode gel. The electric field was applied using the Cliniporator™ (IGEA, Italy), applying a combination of a HV (1000 V/cm (applied voltage=400 V), 100 μs) pulse followed by a LV (100 V/cm (applied voltage=40 V), 400 ms) pulse. Induction of gene expression was obtained by administering drinking water containing dox (doxycycline hyclate, Sigma-Aldrich, Denmark) at various concentrations (0.01–0.2 mg/ml) in distilled water.

Evaluation of EGFP expression

Forty-eight hours after treatment the mice were euthanised and muscles were removed and evaluated by fluorescence stereomicroscopy (Nikon SMZ1500, Japan) in blinded and randomised order. Gene expression was scored based on the area of muscle fibres expressing EGFP on a scale from 0 to 5. No EGFP expression was set at 0, less than 20% of the muscle expressing EGFP was given the mark 1, 20–40%=2, 40–60%=3, 60–80%=4, while more than 80% of the muscle surface expressing EGFP was rewarded 5.

Evaluation of luciferase expression

Forty-eight hours after treatment, the mice were euthanised and the muscles were removed and quickly frozen in dry ice and 100% alcohol. Muscle extracts were prepared by homogenising the muscles with a S8N-5G homogeniser (IKA-werke, Germany) in 1 ml cell lysis buffer (Roche, Germany). Cell debris was spun down, and luciferase activity was measured on the supernatant in a luminometer (LumiStar, Ramcom A/S, Denmark).

Protocol for experiment presented in Figure 5

The three authors of this paper acted as ‘doctors’, trying to prescribe an EGT treatment, which would result in 10, 12 or 14 mmol Hgb/l, respectively. Each author had three groups of eight mice at their disposal. The following parameters could be controlled during the prescription: EGT to one or both TC muscles, the amount of plasmid electrotransferred, the administered dox concentration and finally the mice could be re-transfected in one or both TC muscles. Blood samples were taken every 4 weeks, whereafter the ‘doctors’ had 3 days to make adjustments in their prescriptions.

Evaluation of EPO expression and determination of haemoglobin and serum iron levels

In mice, blood samples (20–30 μl) were drawn from the retro-orbital sinus and haemoglobin levels were measured from a drop of blood using the HemoCue Hb201+ (HemoCue AB, Sweden). The precision of the apparatus was ±0.1 mmol/l, controlled by repeated measurements from the same blood sample. At the termination of the experiments, 300 μl blood were drawn from the mice and free iron was measured using the Vitros 950 Chemistry System. In rats, blood samples (ca. 200 μl) were drawn from the tail vein and Hct level was determined by hairpin centrifugation. The Hct levels are reported, but were also converted to haemoglobin levels by multiplying the Hct value with 0.207, and expressed in mmol/l. The remaining blood sample was spun down, and serum was isolated for determination of EPO levels.

Detection of EPO expression after DNA electrotransfer

Total muscle RNA was isolated from frozen muscles by muscle homogenisation and RNA extraction using the TRIzol reagent (Invitrogen Life Technologies, Denmark). RT-PCR was performed using random p(dN)6 primers (Roche, Germany) and AMV reverse transcriptase (Roche, Germany). EPO was amplified from cDNA using the Brilliant SYBR Green Q-PCR kit 600548 (Stratagene, Cedar Creek, TX, USA). The sequence of the pUHD1-mEPO primers were: sense: 5′-IndexTermatgtcgcctccagataccac, and antisense: 5′-IndexTermcctctcccgtgtacagcttc and PCR and detection was performed by Q-PCR (MX3000P, Stratagene).

Detection of EPO in rat serum

The levels of EPO in serum samples were determined by ELISA (Quantikine Mouse Erythropoietin Kit, R&D systems). Shortly, 50 μl undiluted serum samples were incubated in monoclonal anti-EPO antibodies coated wells for 2 h. After washing, secondary anti-EPO peroxidase-conjugated antibodies were added for 2 h incubation. Following incubation and washing, a colour reaction was developed and EPO concentration was determined by optical density measurements and reported as pg EPO/ml. The sensitivity was determined by the minimum detectable dose, which was 12.5 pg/ml, while no cross-reactivity was observed with other cytokines or other factors.


  1. 1

    Gehl J . Electroporation: theory and methods, perspectives for drug delivery, gene therapy and research. Acta Physiol Scand 2003; 177: 437–447.

    CAS  Article  Google Scholar 

  2. 2

    Bureau MF, Gehl J, Deleuze V, Mir LM, Scherman D . Importance of association between permeabilisation and electrophoretic forces for intramuscular DNA electrotransfer. Biochim Biophys Acta 2000; 1474: 353–359.

    CAS  Article  Google Scholar 

  3. 3

    Satkauskas S, Bureau MF, Puc M, Mahfoudi A, Scherman D, Miklavcic D et al. Mechanisms of in vivo DNA electrotransfer: respective contributions of cell electropermeabilisation and DNA electrophoresis. Mol Ther 2002; 5: 133–140.

    CAS  Article  Google Scholar 

  4. 4

    Satkauskas S, Andre F, Bureau M, Scherman D, Miklavcic D, Mir LM . Electrophoretic component of electric pulses determines the efficacy of in vivo DNA electrotransfer. Hum Gene Ther 2005; 16: 1194–1201.

    CAS  Article  Google Scholar 

  5. 5

    Marty M, Sersa G, Garbay JR, Gehl J, Collins CG, Snoj M et al. Electrochemotherapy – an easy, highly effective and safe treatment of cutaneous and subcutaneous metastases: results of ESOPE (European Standard Operating Procedures of Electrochemotherapy) study. Eur J Cancer Suppl 2006; 4: 3–13.

    CAS  Article  Google Scholar 

  6. 6

    Gehl J, Geertsen PF . Palliation of haemorrhaging and ulcerated cutaneous tumours using electrochemotherapy. Eur J Cancer Suppl 2006; 4: 35–37.

    Article  Google Scholar 

  7. 7

    Mir LM . Bases and rationale of the electrochemotherapy. Eur J Cancer Suppl 2006; 4: 38–44.

    Article  Google Scholar 

  8. 8

    Heller LC, Heller R . In vivo electroporation for gene therapy. Hum Gene Ther 2006; 17: 890–897.

    CAS  Article  Google Scholar 

  9. 9

    Lu QL, Bou-Gharios G, Partridge TA . Non-viral gene delivery in skeletal muscle: a protein factory. Gene Therapy 2003; 10: 131–142.

    CAS  Article  Google Scholar 

  10. 10

    Kreiss P, Bettan M, Crouzet J, Scherman D . Erythropoietin secretion and physiological effect in mouse after intramuscular plasmid DNA electrotransfer. J Gene Med 1999; 1: 245–250.

    CAS  Article  Google Scholar 

  11. 11

    Rizzuto G, Cappelletti M, Maione D, Savino R, Lazzaro D, Costa P et al. Efficient and regulated erythropoietin production by naked DNA injection and muscle electroporation. Proc Natl Acad Sci USA 1999; 96: 6417–6422.

    CAS  Article  Google Scholar 

  12. 12

    Payen E, Bettan M, Rouyer-Fessard P, Beuzard Y, Scherman D . Improvement of mouse [beta]-thalassemia by electrotransfer of erythropoietin cDNA. Exp Hematol 2001; 29: 295–300.

    CAS  Article  Google Scholar 

  13. 13

    Maruyama H, Sugawa M, Moriguchi Y, Imazeki I, Ishikawa Y, Ataka K et al. Continuous erythropoietin delivery by muscle-targeted gene transfer using in vivo electroporation. Hum Gene Ther 2000; 11: 429–437.

    CAS  Article  Google Scholar 

  14. 14

    Terada Y, Tanaka H, Okado T, Inoshita S, Kuwahara M, Akiba T et al. Efficient and ligand-dependent regulated erythropoietin production by naked dna injection and in vivo electroporation. Am J Kidney Dis 2001; 38: S50–S53.

    CAS  Article  Google Scholar 

  15. 15

    Muramatsu T, Arakawa S, Fukazawa K, Fujiwara Y, Yishida T, Sasaki R et al. In vivo gene electroporation in skeletal muscle with special reference to the duration of gene expression. Int J Mol Med 2001; 7: 37–42.

    CAS  PubMed  Google Scholar 

  16. 16

    Maruyama H, Ataka K, Gejyo F, Higuchi N, Ito Y, Hirahara H et al. Long-term production of erythropoietin after electroporation-mediated transfer of plasmid DNA into the muscles of normal and uremic rats. Gene Therapy 2001; 8: 461–468.

    CAS  Article  Google Scholar 

  17. 17

    Ataka K, Maruyama H, Neichi T, Miyazaki J, Gejko F . Effects of erythropoietin-gene electrotransfer in rats with adenine-induced renal failure. Am J Nephrol 2003; 23: 315–323.

    CAS  Article  Google Scholar 

  18. 18

    Rizzuto G, Cappelletti M, Mennuni C, Wiznerowicz M, DeMartis A, Maione D et al. Gene electrotransfer results in a high-level transduction of rat skeletal muscle and corrects anemia of renal failure. Hum Gene Ther 2000; 11: 1891–1900.

    CAS  Article  Google Scholar 

  19. 19

    Fattori E, Cappelletti M, Zampaglione I, Mennuni C, Calvaruso F, Arcuri M et al. Gene electro-transfer of an improved erythropoietin plasmid in mice and non-human primates. J Gene Med 2005; 7: 228–236.

    CAS  Article  Google Scholar 

  20. 20

    Magnani M, Rossi L, Stocchi V, Cucchiarini L, Piacentini G, Fornaini G . Effect of age on some properties of mice erythrocytes. Mech Ageing Develop 1988; 42: 37–47.

    CAS  Article  Google Scholar 

  21. 21

    Toniatti C, Bujard H, Cortese R, Ciliberto G . Gene therapy progress and prospects: transcription regulatory systems. Gene Therapy 2004; 11: 649–657.

    CAS  Article  Google Scholar 

  22. 22

    Gossen M, Freundlieb S, Bender G, Muller G, Hillen W, Bujard H . Transcriptional activation by tetracyclines in mammalian cells. Science 1995; 268: 1766–1769.

    CAS  Article  Google Scholar 

  23. 23

    Perez N, Plence P, Millet V, Greuet D, Minot C, Noel D et al. Tetracycline transcriptional silencer tightly controls transgene expression after in vivo intramuscular electrotransfer: application to interleukin 10 therapy in experimental arthritis. Hum Gene Ther 2002; 13: 2161–2172.

    CAS  Article  Google Scholar 

  24. 24

    Lamartina S, Roscilli G, Rinaudo CD, Sporeno E, Silvi L, Hillen W et al. Stringent control of gene expression in vivo by using novel doxycycline-dependent trans-activators. Hum Gene Ther 2002; 13: 199–210.

    CAS  Article  Google Scholar 

  25. 25

    Chenuaud P, Larcher T, Rabinowitz JE, Provost N, Joussemet B, Bujard H et al. Optimal design of a single recombinant adeno-associated virus derived from serotypes 1 and 2 to achieve more tightly regulated transgene expression from nonhuman primate muscle. Mol Ther 2004; 9: 410–418.

    CAS  Article  Google Scholar 

  26. 26

    Miklavcic D, Semrov D, Mekid H, Mir LM . A validated model of in vivo electric field distribution in tissues for electrochemotherapy and for DNA electrotransfer for gene therapy. Biochim Biophys Acta 2000; 1523: 73–83.

    CAS  Article  Google Scholar 

  27. 27

    Mir LM, Bureau MF, Gehl J, Rangara R, Rouy D, Caillaud JM et al. High-efficiency gene transfer into skeletal muscle mediated by electric pulses. Proc Natl Acad Sci USA 1999; 96: 4262–4267.

    CAS  Article  Google Scholar 

  28. 28

    Bohl D, Naffakh N, Heard JM . Long-term control of erythropoietin secretion by doxycycline in mice transplanted with engineered primary myoblasts. Nat Med 1997; 3: 299–305.

    CAS  Article  Google Scholar 

  29. 29

    Moller PH, Eriksen J, Gehl J . Tet-On induction with doxycycline after gene transfer in mice: sweetening of drinking water is not a good idea. Anim Biotechnol. In press.

  30. 30

    Gothelf A, Mir LM, Gehl J . Electrochemotherapy: results of cancer treatment using enhanced delivery of bleomycin by electroporation. Cancer Treat Rev 2003; 29: 371–387.

    CAS  Article  Google Scholar 

  31. 31

    Mir LM, Gehl J, Sersa G, Collins CG, Garbay J, Billard V et al. Standard operating procedures of the electrochemotherapy: instructions for the use of bleomycin or cisplatin administered either systemically or locally and electric pulses delivered by the CliniporatorTM by means of invasive or non-invasive electrodes. Eur J Cancer Suppl 2006; 4: 14–25.

    CAS  Article  Google Scholar 

  32. 32

    Aihara H, Miyazaki J . Gene transfer into muscle by electroporation in vivo. Nat Biotechnol 1998; 16: 867–870.

    CAS  Article  Google Scholar 

  33. 33

    Gehl J, Sorensen TH, Nielsen K, Raskmark P, Nielsen SL, Skovsgaard T et al. In vivo electroporation of skeletal muscle: threshold, efficacy and relation to electric field distribution. Biochim Biophys Acta 1999; 1428: 233–240.

    CAS  Article  Google Scholar 

  34. 34

    Samakoglu S, Fattori E, Lamartina S, Toniatti C, Stockholm D, Heard JM et al. betaMinor-globin messenger RNA accumulation in reticulocytes governs improved erythropoiesis in beta thalassemic mice after erythropoietin complementary DNA electrotransfer in muscles. Blood 2001; 97: 2213–2220.

    CAS  Article  Google Scholar 

  35. 35

    Durieux AC, Bonnefoy R, Freyssenet D . Kinetic of transgene expression after electrotransfer into skeletal muscle: Importance of promoter origin/strength. Biochim Biophys Acta 2005; 1725: 403–409.

    CAS  Article  Google Scholar 

  36. 36

    Fabre E, Bigey P, Orsini C, Scherman D . Comparison of promoter region constructs for in vivo intramuscular expression. J Gene Med 2006; 8: 636–645.

    CAS  Article  Google Scholar 

  37. 37

    Fisher JW . Erythropoietin: physiology and pharmacology update. Exp Biol Med 2003; 228: 1–14.

    CAS  Article  Google Scholar 

  38. 38

    Zhou HS, Liu DP, Liang CC . Challenges and strategies: the immune responses in gene therapy. Med Res Rev 2004; 24: 748–761.

    CAS  Article  Google Scholar 

  39. 39

    Mathiesen I . Electropermeabilisation of skeletal muscle enhances gene transfer in vivo. Gene Therapy 1999; 6: 508–514.

    CAS  Article  Google Scholar 

  40. 40

    Donà M, Sandri M, Rossini K, Dell'Aica I, Podhorska-Okolow V, Carraro U . Functional in vivo gene transfer into the myofibers of adult skeletal muscle. Biochem Biophys Res Com 2003; 312: 1132–1138.

    Article  Google Scholar 

  41. 41

    Wang XD, Tang JG, Xie XL, Yang JC, Li S, Ji JG et al. A comprehensive study of optimal conditions for naked plasmid DNA transfer into skeletal muscle by electroporation. J Gene Med 2005; 7: 1235–1245.

    CAS  Article  Google Scholar 

  42. 42

    Molnar MJ, Gilbert R, Lu Y, Liu AB, Guo A, Larochelle N et al. Factors influencing the efficacy, longevity, and safety of electroporation-assisted plasmid-based gene transfer into mouse muscles. Mol Ther 2004; 10: 447–455.

    CAS  Article  Google Scholar 

  43. 43

    Hartikka J, Sukhu L, Buchner C, Hazard D, Bozoukova V, Margalith M et al. Electroporation-facilitated delivery of plasmid DNA in skeletal muscle: plasmid dependence of muscle damage and effect of poloxamer 188. Mol Ther 2001; 4: 407–415.

    CAS  Article  Google Scholar 

  44. 44

    Durieux A, Bonnefoy R, Busso T, Freyssenet D . In vivo gene electrotransfer into skeletal muscle: effects of plasmid DNA on the occurrence and extent of muscle damage. J Gene Med 2004; 6: 809–816.

    CAS  Article  Google Scholar 

  45. 45

    Wang M, Orsini C, Casanova D, Millan JL, Mahfoudi A, Thuillier V . MUSEAP, a novel reporter gene for the study of long-term gene expression in immunocompetent mice. Gene 2001; 279: 99–108.

    CAS  Article  Google Scholar 

  46. 46

    Trochon-Joseph V, Martel-Renoir D, Mir LM, Thomaidis A, Opolon P, Connault E et al. Evidence of antiangiogenic and antimetastatic activities of the recombinant disintegrin domain of metargidin. Cancer Res 2004; 64: 2062–2069.

    CAS  Article  Google Scholar 

  47. 47

    Perez N, Bigey P, Scherman D, Danos O, Piechaczyk M, Pelegrin M . Regulatable systemic production of monoclonal antibodies by in vivo muscle electroporation. Genet Vaccines Ther 2004; 2: 2.

    Article  Google Scholar 

  48. 48

    Bettan M, Emmanuel F, Darteil R, Caillaud JM, Soubrier F, Delaère P et al. High-level protein secretion into blood circulation after electric pulse-mediated gene transfer into skeletal muscle. Mol Ther 2000; 2: 204–210.

    CAS  Article  Google Scholar 

  49. 49

    Gossen M, Bujard H . Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. Proc Natl Acad Sci USA 1992; 89: 5547–5551.

    CAS  Article  Google Scholar 

  50. 50

    Kistner A, Gossen M, Zimmermann F, Jerecic J, Ullmer C, Lübbert H et al. Doxycycline-mediated quantitative and tissue-specific control of gene expression in transgenic mice. Proc Natl Acad Sci USA 1996; 93: 10933–10938.

    CAS  Article  Google Scholar 

Download references


We thank Vibeke Uhre and Marianne Fregil for excellent technical assistance. The study was supported by the Danish Research Agency (22-03-0367) and (22-02-0523).

Author information



Corresponding authors

Correspondence to H Gissel or J Gehl.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Hojman, P., Gissel, H. & Gehl, J. Sensitive and precise regulation of haemoglobin after gene transfer of erythropoietin to muscle tissue using electroporation. Gene Ther 14, 950–959 (2007).

Download citation


  • electro gene transfer
  • erythropoietin
  • skeletal muscle
  • tet-on system

Further reading


Quick links