High expression of naked plasmid DNA in muscles of young rodents
High expression of naked plasmid DNA in muscles of young rodentsI. Danko, P. Williams, H. Herweijer, G. Zhang, J.S. Latendresse, I. Bock+ and J.A. Wolff*
University of Wisconsin-Madison, Department of Pediatrics, Waisman Center, Madison, WI 53705, USA
Received February 18, 1997;Revised and Accepted June 23, 1997
There is a time window at 2 weeks of age for achieving very high levels of foreign gene expression from the intramuscular injection of naked plasmid DNA in mice and rats. The highest expression, over 1 [mu]g of luciferase protein/muscle, was obtained in Balb/C mice using constructs containing the CMV promoter, a chimeric intron and the luc+ luciferase gene. Approximately 50% of the myofibers were intensely blue following the intramuscular injection of a [beta]-galactosidase expression vector in 2 week old Balb/C mice. The effects of age, mouse strain and construct were multiplicative, resulting in >1000-fold greater luciferase and ~20-fold more [beta]-galactosidase-positive cells. These high levels of expression were unstable and were not observed in larger animals (dog, rhesus monkey). These results indicate that enormous levels of foreign gene expression can be obtained in muscle with naked DNA in vivo and will enable the temporary effects of gene function and expression in rodent muscle to be expeditiously studied.
Several methods of gene transfer into muscle are under development for laboratory studies and gene therapy (1 -3 ). These include direct methods in which the gene is delivered to muscle in vivo and in situ and indirect methods in which the gene is delivered to myoblasts in culture that are then transplanted (4 -7 ). Direct methods are favored because of their inherent simplicity and the inefficiency of myoblast transplantation (8 -11 ).
Both viral and non-viral vectors have been used for direct gene transfer into muscle. The viral vectors have been derived from adenoviruses (12 ), herpes simplex virus (13 ), retroviruses (14 -16 ) and more recently adeno-associated viruses (17 ,18 ). While molecular conjugates (e.g., polylysine complexes) and cationic lipids have been used to deliver genes directly into other tissues (19 ,20 ), they have not been generally useful in muscle tissue in vivo. Naked plasmid DNA (pDNA) remains the most efficacious non-viral method in skeletal, diaphragmatic, and cardiac muscle to date (21 -24 ). The mechanism of DNA uptake is unknown, although we and others have postulated an active cellular uptake process (25 -27 ).
A variety of factors have been shown to affect the levels of foreign gene expression from intramuscularly injected pDNA. Muscle regeneration induced by myotoxic agents enabled higher levels of expression (28 ,29 ). Expression can also be aided by improved distribution of the pDNA which has been accomplished by preinjection of muscles with large volumes of hypertonic solutions, polymers, or by improved injection technique (26 ,30 ,31 ). Several studies have shown that newborn animals can express naked plasmid DNA (32 -35 ). In carp, plasmid expression of chloramphenicol acetyl transferase (CAT) in muscle was greater in young than old fish (33 ). Plasmid expression was deemed efficient after injection into Xenopus tadpole muscle (34 ). Of particular note was the increased CAT expression from the SV40 promoter in 4-6 week old mice as compared with older mice (35 ). Other studies have shown the importance of the promoter and construct. Generally, the viral promoters RSV and CMV enable higher levels of expression than mammalian promoters (32 ,36 ). Luciferase expression from a CMV immediate-early promoter with a chimeric intron was ~3-fold greater than from a RSV promoter (36 ). Further studies by the same group showed that replacement of the kanamycin gene for the ampicillin gene, removal of SV40 origin of replication, modification of the plasmid backbone and improved transcription terminators brought about very large increases in luciferase expression (37 ).
The present report extends these previous studies and demonstrates exceptionally high levels of foreign gene expression from refined plasmid vectors in 2 week old mouse and rat muscles. Preliminary studies are presented in dogs and non-human primates. The implications for laboratory studies and gene therapy are discussed.
Ten [mu]g of pCILuc in 50 [mu]l of normal saline were injected into the quadriceps of mice and rats at different ages and the luciferase activities were assayed in the entire muscle after 1 week (Fig. 1 ). In all three mouse strains, the greatest luciferase levels were obtained at 2 weeks of age. The mouse 2 week levels were 30- to 120-times higher than the 8 week levels. Injections into 20 week old Balb/C mice yielded 1200-times less luciferase expression on average than in 2 week old mice. In the newborn mice, the lower levels of expression are possibly due to the injection fluid not being well-retained in the very small muscles.
The [beta]-galactosidase reporter system was also used to determine the effect of age on expression. One week after 10 [mu]g of pCILacZ in 50 [mu]l of normal saline was injected into the quadriceps of 2 week old ICR mice, the muscles were examined histologically for [beta]-galactosidase expression. Approximately 50% of the myofibers were intensely blue (Fig. 3 A), which is at least one order of magnitude greater than the percent of positive myofibers our laboratory has previously obtained in older mice. A large number of [beta]-galactosidase-positive myofibers were also obtained in 2 week old rats (Fig. 3 B, C). No blue myofibers are found following injection with pCMVLux.
The luciferase expression of various constructs containing different promoters and luciferase genes were compared in 2 week old ICR mice (Table 1 ). The two pCI-based constructs, pCILuc and pCILuc2 yielded the highest levels of expression. A comparison of luciferase expression from pCILuc and pBS.CMVLuc indicates that the pCI elements enabled 8-fold more expression. A comparison of luciferase expression from pBS.CMVLuc and pBS.CMVLux indicates that the Luc gene enabled 2.5-times more expression than the Lux gene. These differences are similar to those obtained following transient transfection with cationic lipids into 3T3 cells in culture (data not shown). Finally, a comparison of pBS.CMVLux and pBS.RSVLux indicates that the RSV promoter enabled approximately two times more luciferase expression.
The luciferase expression of the various constructs in 2 week old ICR mice was compared with their expression in older ICR mice (Table 2 ). For all the constructs, the greatest expression was obtained at 2 weeks of age; the least expression was obtained at 8 weeks of age. The ratio of 2-8 week old expression was greatest for pBS.RSVLux and smallest for SV40. In C57/Bl mice, injections of pCILuc also yielded higher levels of expression in 2 week old mice (Table 3 ). However, injections of pSV40Lux DNA yielded similar levels in 2 and 4 week old mice but less in 8 week old mice (Table 3 ).
The muscle creatine kinase promoter was evaluated in young and old Balb/c mice. One week after injection of pBS.MCKLuc, the mean luciferase level in 2 week old mice was 12.4 ng/muscle (+-4.9, n = 8), ~90 times greater than the mean of 0.14 ng (+-0.13, n = 8) in 8 week old mice.
The stability of luciferase expression was determined by assaying the levels in excised quadriceps muscles at various times after the muscles in 2 week old ICR mice were injected with different expression constructs (Fig. 5 A). The most time points were analyzed for the pCILuc construct for which the luciferase levels decreased ~250-fold from 1-2 weeks after injection (Fig. 5 A). Thereafter, luciferase expression persisted for at least 22 weeks. Luciferase levels also decreased from 1 to 8 weeks after injection of the pBS.CMVLux and pBS.RSVLux constructs (Fig. 5 A).ABCD
Luciferase activity was assayed 1 week after various muscles in puppies of different ages were injected with 200 or 1000 [mu]g of pCILux in one ml of normal saline. Overall the luciferase levels were substantially less than those in rats (Fig. 6 ). The maximum levels were obtained in 4 week old puppies and in the TA and biceps muscles which are smaller than the quadriceps. Whereas the highest levels of total muscle luciferase in the dogs were 2-6-fold less than those in C57/Bl and ICR mice and rats, the efficiencies of expression in terms of ng luciferase/[mu]g pCILux injected were ~62- and ~2.7-fold less than that in the mice and rats, respectively.
Similar injections and analyses were done in 2 week old monkeys except 40-180 mg of muscle were obtained on open biopsy at the site of injection (Table 4 ). One week after injection with 1 mg of pCILux in 0.5 ml of normal saline, the muscles contained substantially less luciferase activity than did the rodents. In terms of luciferase activity per mg of muscle tissue, the primate results were similar to that in 2 week old puppies but ~10-fold less than that in 4 week old puppies.
These results indicate that there is a time window at 2 weeks of age for achieving very high levels of foreign gene expression from intramuscularly injected pDNA. This time window was observed with several different constructs, which contained either the CMV, RSV, SV40, or MCK promoter. It also occurred in three different mouse strains and in rats. The highest expression, over 1 [mu]g of luciferase protein/muscle, was obtained in Balb/C mice using constructs containing the CMV promoter, chimeric intron, and the luc+ luciferase gene. Of note was that the effects of age and construct were multiplicative. We have recently shown that very high levels of expression in hepatocytes can be obtained when pDNA in hypertonic solutions is injected into the portal vein (39 ). The highest efficiencies of expression in terms of ng of luciferase protein/[mu]g of pCILuc delivered were comparable in 2 week old muscle (~100 ng/[mu]g) and liver (~170 ng/[mu]g) (Zhang et al., in preparation). These muscle and liver results indicate that enormous levels of foreign gene expression can be obtained with naked DNA in vivo.
The only deviation to this time window at 2 weeks of age was the expression of pSV40Lux in C57/Bl mice (Table 3 ). Luciferase expression in 2 and 4 week old mice were similar but substantially less in 8 week old animals. The similar levels in 2 and 4 week old mice were specific to the injection of pSV40Lux in C57/Bl mice. It was observed neither with the pCILuc vector in C57/Bl mice (Table 3 ) nor with pSV40Lux in ICR mice (Table 2 ). Our results with pSV40Lux in C57/Bl mice were similar to the previously published results of Wells and Goldspink who observed higher expression of CAT from the SV40 promoter in 4 week old than from 2 week old C57/Bl mice (35 ).
The reason for greater expression in 2 week old rodents is open to speculation. Since the mechanism of pDNA uptake in muscle is not known, definitive conclusions cannot be reached. The larger number of [beta]-galactosidase-positive cells is consistent with increased cellular pDNA uptake and a greater number of transfected myofibers. The diffuse pattern of intramyofiber staining observed after intramuscular injection into 2 week old (but not adult) mice may be due to a greater ability for cellular DNA uptake in the younger muscle. Also, Southern blot analysis indicated that the amount of pDNA 1 week after injection was greater in 2 week old than adult mice (data not shown). The greater amount of pDNA associated with the muscle may also result from a better delivery of the pDNA throughout the young muscle because it contains less extracellular matrix to limit its distribution (25 ,40 ). The greater expression in dy/dy mice which have a disrupted extracellular matrix is consistent with this hypothesis. The greater infectivity of adenoviral or herpes virus vectors in younger mouse muscles has been explained by a decreased limitation by the extracellular matrix (13 ,41 ,42 ).
On the other hand, the increased expression could be the result of increased transcriptional activity of the promoters. The cellular milieu of myofibers in 2 week old animals may contain increased amounts of rate-limiting transcriptional factors that are common to the four promoters studied. Increased transcriptional activity may not only explain the increased luciferase levels but also the increased number of [beta]-galactosidase-positive cells. Previous studies using flow cytometry have suggested that enhancers increase the likelihood of expression from a transfected gene and not necessarily the amount of expression per cell (43 ). Similarly, the greater number of [beta]-galactosidase-positive cells could result from an increased probability that the transfected gene is transcriptionally active in young muscle. Also, the small myofibers in the younger muscle may concentrate the [beta]-galactosidase and lower the threshold for detecting X-gal staining. However, greater expression in young mice was also observed when the MCK promoter was used. The fact that this promoter is more transcriptionally active in differentiated muscle cells (44 ) argues against increased transcriptional activity solely accounting for the greater expression in the 2 week old mice.
The high levels of expression in 2 week old animals were unstable. It occurred in 2 week old ICR and Balb/C mice and with all three promoters (Fig. 5 ). In older animals, expression was not unstable as we have previously shown (45 ). In fact, it appeared that total luciferase gravitated to 1 ng/muscle regardless of the initial levels. High levels of luciferase in 2 week old animals decreased to ~1 ng/total muscle and then persisted for at least 22 weeks. Low levels of luciferase also persisted after an initial period of high expression in another study of optimized plasmid constructs in muscle (37 ). As we had previously shown for older animals (45 ), luciferase levels rose to ~1 ng/total muscle over the first 8 weeks after injection. Perhaps the immune system is responsible for limiting luciferase expression above the threshold of 1 ng/total muscle. Myofibers expressing high amounts of luciferase could be destroyed by the immune system while myofibers expressing less luciferase escape immune detection. The preliminary studies in which luciferase expression persists in 2 week old animals immunosuppressed by FK506 is consistent with this hypothesis.
The high but transient levels of luciferase expression in 2 week old animals is reminiscent of a similar course following plasmid injections into bupivacaine treated muscle (29 ). Muscles of both 2 week old mice and mice following bupivacaine treatment contain an increased number of proliferating and fusing myoblasts (46 ). This state could enable either increased plasmid uptake or increased expression at the transcriptional or translational level as discussed above.
Muscle injections were done in rats, dogs, and non-human primates to determine its possible clinical utility in human infants. For example, newborn screening could identify newborns with Duchenne muscular dystrophy who could be injected as early as 2 weeks of age. It was encouraging that the mice data extended to rats. However, the results in dogs and non-human primates were less promising. It appears that this technique cannot be used to treat intrinsic disorders of muscle, such as Duchenne muscular dystrophy, which require >10% of the muscles to be transfected (47 ). Nonetheless, the levels of expression in these young animals were substantially greater than in older animals (40 ). The use of DNA intramuscular injections for immunization may be more effective in infants.
In the laboratory, the ability to obtain such high levels of foreign gene expression in young muscle from pDNA will enable the mouse to be used as cell culture is now used. The transfer of genes into cells in culture has been a critically important tool for deciphering the function of genes. Typically, the gene under study is placed within a plasmid vector and transiently transfected into the appropriate cell in culture. Isoforms and mutant forms of the gene under study can be quickly placed into plasmid expression vectors and studied. The use of pDNA avoids the laborious steps necessary for the production of viral vectors or generation of transgenic mice and thereby enables many different genes and their related mutated forms to be quickly studied. The technique described in this report will permit temporary effects of gene function in muscle to be expeditiously probed within a mammalian organism.
C57/Bl, Balb/C and ICR mice, Sprague-Dawley rats and Beagle dogs were purchased from Harlan-Sprague Dawley (Indianapolis, IN). The mdx-4Cv dystrophic mice (48 ) and 129/ReJ-Lama2dy mice were obtained from the Jackson Laboratory (Bar Harbor, ME). Rhesus monkeys were bred and maintained at the Wisconsin Regional Primate Center. Animals were anesthetized with metofane (mice), a mixture of Ketamine and Xylazine (rats) and isoflurane (dogs and monkeys). The DNA was injected over a 1 min period using a 1 ml syringe and 27 gauge needle. Muscles were directly exposed to facilitate the injections into specific muscle groups. Incisions were closed using silicone treated silk sutures. All animal protocols were approved by the institutional animal care committee.
Some of the mice received daily intramuscular injections of 0.8 [mu]g/g body weight of FK506 starting 1 day prior to pDNA injection.
Plasmid DNA was purified using QIAGEN plasmid mega kits (Qiagen, Santa Clarita, CA) or by double cesium chloride banding as described previously (40 ). The construction of pBS.CMVLux, pBS.RSVLux, and pSV40Lux has been described (49 ). The plasmid pCILuc contains the CMV promoter and a chimeric intron and was constructed by inserting the luc+ gene (NheI-EcoRI fragment) from pSP-luc+ (Promega, Madison, WI) into NheI and EcoRI sites in the multiple cloning site of pCI (Promega). To generate pBS.CMVLuc, the luc+ gene (KpnI-EcoRI fragment from pSP-luc+, Promega) was cloned into KpnI and EcoRI sites in the multiple cloning site of the pBlueCMV expression vector (49 ). pCILuc2 was constructed by removing the f1 origin and ampicillin resistance gene from pCR3 (Invitrogen, Carlsbad, CA) by StuI and SspI digestion followed by religation. The 5' chimeric intron and multiple cloning site sequences (HindIII-NotI fragment) from pCI (Promega) were inserted into HindIII-Bsp120I sites to replace the original pCR3 multiple cloning site. The luciferase gene (NheI-EcoRI fragment from pSP-luc+) was inserted into NheI and EcoRI sites.
pRSVBecker contains a Becker-like dystrophin gene expressed from a RSV promoter (50 ). pCIBecker contains a Becker-like dystrophin gene (SalI fragment from pUC18-Becker) inserted into the SalI site of pCI. The pBS.MCKLuc construct contains the mouse muscle creatine kinase gene promoter [3300 bp, (51 )] in pBluescript. The luc+ gene and SV40 polyadenylation signal (HindIII-ClaI fragment from pCILuc) were placed downstream of the MCK promoter. pCILux was constructed by inserting the lux gene (HindIII-BamI fragment pBS.CMVLux) into the SmaI site of pCI. Plasmid DNA was fluorescently labeled with rhodamine as described (M. Sebestyen et al., manuscript submitted).
Luciferase activity in muscle was analyzed as previously reported (21 ) except muscle extracts were prepared in larger volumes of luciferase lysis buffer (0.1% Triton X-100, 0.1 M potassium phosphate, 1 mM DTT, pH 7.8). The entire muscles of mice and rats were homogenized in 500 [mu]l of lysis buffer using a PRO 200 homogenizer (PRO Scientific Inc., Monroe, CT). The whole muscles of dogs were weighed (0.38-6.7 g) and divided into equal parts of 95-837 mg and then homogenized with 1 ml of lysis buffer. The muscle biopsies from the monkeys (40-180 mg) were homogenized with 500 [mu]l lysis buffer. Supernatants were collected after centrifugation at 4000 r.p.m. at 4oC for 10 min. One to 20 [mu]l of the supernatant were analyzed for luciferase activity. Relative light units (RLU) were converted to pg of luciferase using standards from Analytical Luminescence Laboratories (San Diego, CA). Luciferase protein (pg) = 5.1 * 10-5 * RLU.
Eight [mu]m thick tissue sections were stained for [beta]-galactosidase expression with X-Gal for 8-16 h (21 ,49 ). Hematoxylin was used for the counterstain. Dystrophin immunostaining was done using rabbit antibody 6-10 as previously reported (49 ,50 ). DNA labeled with rhodamine fluorescence was analyzed on 10 [mu]m sections, fixed in 70% ethanol (4 min) using a Nikon microscope and Photometrics SenSys CCD camera.
We thank T. Lockie for assistance in preparing this manuscript and E. Langer for technical assistance. This research was supported in part by the Muscular Dystrophy Association (USA) and by NIH grant number RR00167 to the Wisconsin Regional Primate Research Center.
25 Wolff, J., Dowty, M., Jiao, S., Repetto, G., Berg, R., Ludtke, J., Williams, P. and Slautterback, D. (1992) J. Cell Sci., 103, 1249-1259.MEDLINE Abstract
26 Levy, M.Y., Barron, L.G., Meyer, K.B. and Szoka, F.C., Jr. (1996) Gene Ther., 3, 201-211.MEDLINE Abstract
27 Hagstrom, J.E., Rybakova, I.N., Staeva, T., Wolff, J.A. and Ervasti, J.M. (1996) Biochem. Mol. Med., 58, 113-121.
28 Davis, H.L., Demeneix, B.A., Quantin, B., Coulombe, J. and Whalen, R.G. (1993) Hum. Gene Ther., 4, 733-740.MEDLINE Abstract
29 Danko, I., Fritz, J.D., Jiao, S., Hogan, K., Latendresse, J.S. and Wolff, J.A. (1994) Gene Ther., 1, 114-121.MEDLINE Abstract
31 Mumper, R.J., Duguid, J.G., Anwer, K., Barron, M.K., Nitta, H. and Rolland, A.P. (1996) Pharmaceut. Res., 13, 701-709.
32 Wolff, J., Williams, P., Acsadi, G., Jiao, S., Jani, A. and Chong, W. (1991) BioTechniques, 4, 474-485.
33 Hansen, E., Fernandes, K., Goldspink, G., Butterworth, P., Umeda, P.K. and Chang, K.-C. (1991) FEBS J., 290, 73-76.
34 de Luze, A., Sachs, L. and Demeneix, B. (1993) Proc. Natl. Acad. Sci. USA, 90, 7322-7326.MEDLINE Abstract
35 Wells, D.J. and Goldspink, G. (1992) FEBS Letts, 306, 203-205.
36 Manthorpe, M., Cornefert-Jensen, F., Hartikka, J., Felgner, J., Rundell, A., Margalith, M. and Dwarki, V. (1993) Hum. Gene Ther., 4, 419-431.MEDLINE Abstract
37 Hartikka, J., Sawdey, M., Cornefert-Jensen, F., Margalith, M., Barnhart, K., Nolasco, M., Vahlsing, H.L., Meek, J., Marquet, M., Hobart, P., Norman, J. and Manthorpe, M. (1996) Hum. Gene Ther., 7, 1205-1217.MEDLINE Abstract
38 Xu, H., Christmas, P., Wu, X.R., Wewer, U.M. and Engvall, E. (1994) Proc. Natl. Acad. Sci. USA, 91, 5572-5576.MEDLINE Abstract
39 Budker, V., Zhang, G., Knechtle, S. and Wolff, J.A. (1996) Gene Ther., 3, 593-598.MEDLINE Abstract
40 Jiao, S., Williams, P., Berg, R.K., Hodgeman, B.A., Liu, L., Repetto, G. and Wolff, J.A. (1992) Hum. Gene Ther., 3, 21-33.MEDLINE Abstract
41 Acsadi, G., Jani, A., Massie, B., Simoneau, M., Holland, P., Blaschuk, K. and Karpati, G. (1994) Hum. Mol. Genet., 3, 579-584.MEDLINE Abstract
42 Acsadi, G., Massie, B. and Jani, A. (1995) J. Mol. Med., 73, 165-180.MEDLINE Abstract
43 Walters, M.C., Fiering, S., Eidemiller, J., Magis, W., Groudine, M. and Martin, D.I.K. (1995) Proc. Natl. Acad. Sci. USA, 92, 7125-7129.MEDLINE Abstract
*To whom correspondence should be addressed. Tel: +1 608 263 5993; Fax: +1 608 263 0530; Email: wolffj@macc.wisc.edu
+Visiting scholar: Pecs University Medical School, Department of Pediatrics, Pecs, Hungary
-->
This page is maintained by OUP admin. Last updated Wed Aug 13 15:52:16 BST 1997. Part of the OUP Journals World Wide Web service.
Copyright
Oxford University Press, 1996
