In Vivo Imaging of Far-3
To determine the minimum dose of Katushka plasmid needed to give detectable fluorescent intensity, we decreased the amount of pTurboFP635 to 0.5 and 1 µg, respectively. Electrotransfer with 1.0 µg of plasmid resulted in detectable fluorescent signal with an intensity of 1,090 NC, proving that as little as 1.0 µg of Katushka plasmid is detectable by in vivo imaging (data not shown).
Lifetime analysis of Katushka expression
After excitation, fluorescent proteins are characterized by a specific decay time, known as lifetime. Determination of the lifetime enables recognition of a specific protein by time domain analysis. Lifetime analysis of the transgenic Katushka signal obtained within the first 2 weeks after DNA electrotransfer showed a lifetime of 2.1 ns. This corresponds to the expected lifetime of Katushka (Fig. 4). In line with the decrease in fluorescent intensity, the lifetime also decreased at 4 weeks after DNA electrotransfer (Fig. 2). The temporal point spread function (TPSF) at 4 weeks showed a dual display, indicating that a real yet weak Katushka signal was mixed with the background signal (data not shown).
Fig. 4 Time course of the lifetime of Katushka expression, showing the same muscles as in (Fig. 1).
Comparison of Katushka versus GFP expression
To compare the efficacy of Katushka with GFP, which has been used extensively for imaging, a scanning series comparing the two was performed (Fig. 5). Again, the fluorescent intensity of Katushka peaked at 1 week after DNA electrotransfer and returned to background level within 4 weeks. The same pattern of GFP intensity was present with peak intensity obtained 1 week after DNA electrotransfer. GFP, however, did not show the same degree of decrease in fluorescent intensity and the signal remained detectable for at least 8 weeks. Looking at the lifetime analyses, Katushka lifetime decreased at 3 weeks after treatment, while GFP lifetime remained stable for at least 6 weeks (data not shown).
Fig. 5 Comparison of Katushka and GFP expression in muscles after DNA electrotransfer. Intensity of Katushka or GFP followed over time and the color scale is set to the same range for both Katushka and GFP. The left leg was transfected, while the right leg served as control. The pictures are representative of four mice for each gene.
3D distribution of Katushka expression
The time-of-flight imaging acquisition enabled us to determine the spatial distribution of the fluorescent signal. Through 3D analysis (Fig. 6) we determined the spatial location of both the Katushka and GFP signal in muscles 1 week after DNA electrotransfer. For Katushka, the fluorescent signal was located 0.1 mm from the top of the leg, reaching 5.6 mm down. From the side of the leg, the fluorescent signal ranged from 1 mm under the skin to 5 mm inside the leg. The longitude of the fluorescent signal was 5 mm. GFP expression was approximately located in the same area. This volume is equivalent to the location of the tibialis cranialis muscle, which was the intended target for the DNA electrotransfer.
Fig. 6 3D analysis of Katushka (left panel) and GFP (right panel) expression in muscles 1 week after DNA electrotransfer. The 3D analysis allows us to determine the spatial distribution of the Katushka expression, which in this case coincides with the location of the tibialis cranialis muscle.
Discussion
Bio-imaging shows great advantages for detection of gene expression in vivo (1;14). In this study, we report highly efficient Katushka expression in muscles after DNA electrotransfer with little background expression. As little as 1.0 µg Katushka plasmid is sufficient for in vivo detection. Due to the favorable light penetration in the infrared region, precise determination of the spatial expression in the tissue was possible, and we found the Katushka signal to be located in the region corresponding to the tibialis cranialis muscle. Compared to GFP, the Katushka signal had significantly higher peak intensity in the first weeks; however, the signal was of shorter duration and in vivo detection was lost after 4 weeks.
DNA electrotransfer is a highly efficient method for in vivo gene delivery with high level and long-term expression of the transferred genes (15;16). For in vivo imaging of muscular transgenic expression, penetrating and stable signals are needed. In vivo imaging with Katushka offered high sensitivity and precision of the signal, but the signal decayed over a few weeks. This shows that even though the brightness of Katushka is superior to other far-red fluorescent proteins (4), Katushka is not the optimal marker for long-term expression as is obtained after DNA electrotransfer to muscle tissue. Katushka may prove useful as a marker of weak gene expression due to the superior brightness. In this study, the intensity of Katushka was 3-fold higher than GFP at 1 week after DNA electrotransfer.
A particular advantage of time domain optical imaging is the possibility to determine the spatial distribution of the fluorescent signal. We found that both Katushka and GFP expression was good tools in 3D analysis as little background and divergence of the signal occurred. Based on the geometric positions generated by the 3D analysis, we were able to give an estimate of the location of the fluorescent signal. In the transfected legs, the fluorescent signal corresponded to the location of tibialis cranialis muscle, which was the intended target of our transfections. Since we were imaging tissues, which were only 5 mm in depth, we did not find any significant differences between the two fluorochromes regarding the estimation of the spatial distribution. Far-red fluorescent proteins should in theory be superior in deep tissue imaging; however, in our case, the tibialis cranialis muscle was not located deep enough for this effect to be important.
The initial study describing Katushka proved it highly photostable with fast recovery after bleaching (4). Thus, the weekly scans, which we performed, should not be bleaching the signal to a degree that could explain the decrease in intensity, which we observe in our experiments. Katushka has a fast maturation time (20 min); however, the half-life in vivo is unknown. Our studies showed that Katushka has a lower stability in vivo than GFP. Katushka is a newly described fluorescent molecule and few studies on the biochemical properties of Katushka exists (4), thus whether the low stability is due to intrinsic protein features or whether Katushka might be immunogenic in vivo remains to be investigated.
In conclusion, time domain optical imaging with intensity, lifetime, and 3D analyses ensured extensive characterization of Katushka expression; proving that Katushka offered excellent evaluation of the transfection efficacy with a signal much brighter than GFP. The maintenance of the fluorescent signal was, however, lost within a few weeks, showing that Katushka is not a good candidate for long-term detection of in vivo transgenic expression. Both Katushka and GFP were good markers for the spatial distribution of gene expression in the tibialis cranialis muscle.
Acknowledgements We thank Marianne Fregil, Anne Boye, and Lone Christensen for excellent technical assistance. This project was supported by funds from the Danish National Advanced Technology Foundation (Højteknologifonden) and the Danish National Research Foundation (#02-512-55).
