Optical Imaging of Cardiac Reporter Gene Expression in Living Rats
Background— Studies of cardiac gene transfer rely on postmortem analysis using histologic staining or enzyme assays. Noninvasive imaging of the temporal and spatial characteristics of cardiac gene expression in the same subject offers significant advantages.
Methods and Results— Rats underwent direct myocardial injection via left thoracotomy with adenovirus-expressing firefly luciferase (Ad-CMV-Fluc; n=30). The reporter substrate d-luciferin was injected intraperitoneally. Serial images were acquired by use of a cooled charged couple detector (CCD) camera. Results are expressed as relative light unit per minute (RLU/min). Rats transduced with 1×109 plaque-forming units show decremental cardiac luciferase activity over time: 152 070±21 170 (day 2), 195 806±62 630 (day 5), 7250±2941 (day 8), and 2040±971 RLU/min (day 14). To assess the detection sensitivity, serially diluted titers of Ad-CMV-Fluc were injected: 1×109 (195 393±14 896), 1×108 (33 777±18 179), 1×107 (417±91), 1×106 (185±64), 1×105 (53±1), and control (54±1) (P<0.05 for 1×109, 1×108, and 1×107 plaque-forming units versus control adenovirus-expressing mutant thymidine kinase [Ad-CMV-HSV1-sr39tk]; n=3). Finally, rats were euthanized, and in vitro luciferase activity correlated with in vivo CCD signals (r2=0.92).
Conclusions— This study demonstrates for the first time the feasibility of imaging the location, magnitude, and time course of cardiac reporter gene expression in living rats. Cardiac gene therapy studies could be aided with wider application of this approach.
Received January 29, 2002; revision received February 15, 2002; accepted February 15, 2002.
Efficient and long-term reporter gene expressions in the myocardium using viral and nonviral vectors have been reported since the early 1990s.1–3⇓⇓ However, single “snapshot” assessment of reporter gene activity by postmortem tissue sampling precludes longitudinal analysis within the same animal. Therefore, the development of a simple and reproducible method for in vivo imaging of gene expression in response to physiological, hormonal, or promoter influences would be important for gene therapy studies.
Advances in imaging technologies have made visualization of in vivo reporter gene expression possible. Current approaches include single-photon emission computed tomography (SPECT), positron emission tomography (PET), magnetic resonance imaging (MRI), and optical fluorescent and bioluminescent imaging.4 Among these modalities, optical bioluminescent imaging using a cooled charged couple detector (CCD) camera provides a rapid, sensitive, and cost-effective option for small animal studies. We have recently shown that CCD imaging of adenoviral-mediated firefly luciferase (Ad-CMV-Fluc) reporter gene expression in skeletal muscles of living mice is feasible, reproducible, and quantitative.5
To date, there has been no published report on noninvasive imaging of cardiac gene expression. We therefore explored the possibility of using a cooled CCD camera for imaging the location, magnitude, and duration of reporter gene expression in the myocardium of living rats.
Construction of Recombinant Adenoviruses
Construction of replication-defective adenovirus carrying firefly luciferase (Fluc) driven by a constitutive cytomegalovirus (CMV) promoter (Ad-CMV-Fluc) and of the control adenovirus carrying the mutant herpes simplex virus type 1 thymidine kinase (Ad-CMV-HSV1-sr39tk) has been described.5,6⇓ The firefly luciferase gene and enzyme are subsequently referred to as Fluc and FL, respectively.
Thirty-four Sprague-Dawley rats (weighing 250 to 350 g; Charles River Laboratories, Wilmington, Mass) were studied under protocols approved by the University Committee on the Use and Care of Animals. Rats received 4:1 mixture of ketamine (80 mg/kg) and xylazine (10 mg/kg) intraperitoneally for anesthesia; banamine (2.5 mg/kg) for pain relief; atropine (40 μg/kg) for prevention of salivation and bradycardia; and normal saline (3 mL) for volume replacement. One animal died from respiratory distress after extubation.
Myocardial Injection of Adenovirus
The anesthetized rats were ventilated with a Harvard respirator. A left thoracotomy was performed to expose the beating heart. Varying titers of Ad-CMV-Fluc were injected into the anterolateral wall of the left ventricle using a 30-gauge needle with 30 to 50 μL of injectate volume. Control animals were injected with 1×109 plaque-forming units (pfu) of Ad-CMV-HSV1-sr39tk.
Quantification of Bioluminescence From Cooled CCD Camera and Biochemical Analysis
The Xenogen In Vivo Imaging System (IVIS) consists of a cooled CCD camera mounted on a light-tight specimen chamber, a cryogenic refrigeration unit, a camera controller, and a computer system for data analysis. Both the IVIS and its imaging analysis software are commercially available (Xenogen Corp). d-Luciferin was injected intraperitoneally at a dose of 125 mg/kg body weight (Xenogen Corp). Serial images were acquired every 2 to 6 days. The bioluminescence was quantified as described previously.5 Rats were euthanized at various time points (days 2 to 14) after viral transduction. Biochemical assays were performed as reported previously by a luminometer (Turner Design-20/20).5
All data are given as mean±SEM. For statistical analysis, the 2-tailed Student’s t test was used. Differences were considered significant at P<0.05. Luciferase activities from in vivo CCD imaging and in vitro luminometer assays were correlated by least-squares linear regression.
FL Activity in Myocardium
To assess the time course of cardiac FL activity, a cooled CCD camera was used to image the same animal for 2 weeks (Figure 1). On average, cardiac FL peaks between days 2 to 5 and declines rapidly afterward: 152 070±21 170 (day 2), 195 806±62 630 (day 5), 7250±2941 (day 8), and 2040±971 relative light unit (RLU)/min (day 14) (n=5). All values are statistically significant (P<0.05) compared with rats injected with control Ad-CMV-HSV1-sr39tk, which yield background signals of 54±1 RLU/min in the thoracic region of interest (n=3). Because of leakage of adenovirus from the myocardium, the liver FL activity predominates after day 8: 7780±3601 (day 2), 70 187±41 169 (day 5), 180 257±159 924 (day 8), and 227 504±201 760 RLU/min (day 14).
To determine the minimal viral titer of Ad-CMV-Fluc that can be imaged in the myocardium of living rats, serially diluted titers were injected into the heart (1×109 to 1×105 pfu; n=3 each). At day 2, significant signals of cardiac FL transduced with 1×107 pfu of Ad-CMV-Fluc are detected: 195 393±14 896 (1×109), 33 777±18 179 (1×108), 417±91 (1×107), 185±64 (1×106), 53±1 (1×105), and 54±1 RLU/min (control) (P<0.05 for 1×109, 1×108, and 1×107 pfu versus control) (Figure 2A).
Correlation of In Vivo CCD Signal Versus In Vitro Luminometer Activity
To verify that bioluminescence obtained by noninvasive imaging reflects actual FL enzyme activity in the heart, animals were euthanized, tissues were homogenized, and in vitro cardiac FL activity was determined. Additional rats were transduced with 1×109 pfu Ad-CMV-Fluc, scanned, and euthanized at other time points: day 1 (n=2), day 3 (n=3), day 4 (n=2), day 6 (n=1), and day 9 (n=1). Regression analysis shows a high degree of correlation between results obtained by in vivo CCD imaging and by in vitro biochemical assays of myocardium (r2=0.92) (Figure 2B).
This study demonstrates for the first time that gene expression in the myocardium of living subjects can be monitored noninvasively. The location, magnitude, and persistence of Fluc reporter gene expression in the same animal were followed for 2 weeks. Cardiac FL activity peaks within the first 3 to 5 days and declines rapidly thereafter because of mounting of host inflammatory and cellular responses against adenoviral vectors.7 Consistent with earlier observations, Fluc expression via direct myocardial injection was localized to the area surrounding the site of injection.2 The detection threshold of CCD camera is remarkable, down to 1×107 pfu of Ad-CMV-Fluc. Most importantly, results from in vivo cardiac CCD imaging correlate well with in vitro enzyme assays. These findings suggest that in vivo cardiac CCD imaging can be used to monitor gene expression in lieu of or in parallel with biochemical analysis.
We believe optical imaging has many applications for cardiac gene therapy studies. Because of its noninvasive nature, ideal conditions needed for cardiac gene expression can be determined quickly and accurately. These include analysis of vectors, delivery techniques, promoter specificity, and transgenic models. Likewise, pharmaceutical agents that prolong or enhance therapeutic cardiac gene expression can be screened. In our study, the gradual increase of FL activity in the liver was likely affected by the quality of injection, with egress of adenovirus from the myocardium into systemic circulation and eventual binding to Coxsackie-adenovirus receptors on hepatocytes.2,3,8⇓⇓ Substituting a cardiac tissue-specific promoter (eg, myosin light chain kinase) for the constitutive CMV promoter may diminish extra-cardiac activity.8 Furthermore, with its high detection sensitivity (≈1×107 pfu for adenovirus), CCD imaging may allow monitoring of gene delivery by vectors with low transduction efficiency (eg, plasmid DNA). Finally, the observed variability of FL activity among different rats implies considerable individual variations in response to gene transfer and thus emphasizes the need for noninvasive monitoring.
Several features make optical gene imaging unique. The CCD camera operates by converting photons striking a CCD pixel into electrons, allowing it to register and detect visible through near-infrared light signals.5 Because photons are generated only when the reporter probe (d-luciferin) interacts with the reporter protein (FL) intracellularly, this “smart” reporter probe yields minimal background signal, as shown by control thymidine kinase rats injected with d-luciferin. In contrast, radiotracers used with SPECT and PET emit radioactivity constantly, irrespective of reporter probe/protein interaction, therefore generating significant background signal.4 In addition, FL enzyme turnover in the presence of d-luciferin has a fast half-life (T1/2=3 hours), and FL levels are related linearly with light emissions up to 7 orders of magnitude.5 These properties allow for real time and quantitative measurements of Fluc gene expression.
The main drawbacks of bioluminescent reporter gene imaging are 2-fold. First, CCD imaging is limited to small animals because the efficiency of light transmission is dependent on both tissue type and tissue scattering. Estimates from our prior in vivo studies in mice show that attenuation from skin and muscle is 2- to 3-fold.5 Second, the optical images are not tomographic and cannot localize the exact site of gene expression within the heart from nearby organs. In contrast, the micro PET used in our laboratory has a 23-mm3 (or 8-mm3) resolution and yields 3D tomographic resolution.9 To further address the relative merits of each approach, we are currently exploring vectors in which both optical (Fluc) and PET reporter gene (HSV1-sr39tk) are co-expressed. Eventually, the goal is to indirectly study the expression of any therapeutic gene by linking it to a reporter gene that can be repetitively imaged.10
In summary, noninvasive monitoring of cardiac gene transfer offers significant advantages. Optical imaging of reporter gene expression in animal studies provides an accurate, practical, and cost-effective method to assess the efficiency of gene delivery and expression. Clinical catheter-based detection of vascular or myocardial reporter gene expression may also eventually be possible. Continued refinement of catheters that are shielded from extraneous sources of light, transmission of light via optical fibers, and portable highly sensitive cooled CCD cameras should help to translate these approaches into the clinical setting.
This work was supported by funding from Department of Energy contract DE-FC03-87ER60615 (Dr Gambhir), National Institutes of Health RO1 CA82214-01 (Dr Gambhir), Small Animal Imaging Resource Program R24 CA92865 (Dr Gambhir), and an American Heart Association Western Affiliate postdoctoral grant (Dr Wu).
- ↵Wolff JA, Malone RW, Williams P, et al. Direct gene transfer into mouse muscle in vivo. Science. 1990; 247: 1465–1468.
- ↵Guzman RJ, Lemarchand P, Crystal RG, et al. Efficient gene transfer into myocardium by direct injection of adenovirus vectors. Circ Res. 1993; 73: 1202–1207.
- ↵2 Huard J, Lochmuller H, Acsadi G, et al. The route of administration is a major determinant of the transduction efficiency of rat tissues by adenoviral recombinants. Gene Ther. 1995; 2: 107–115.
- ↵Ray P, Bauer E, Iyer M, et al. Monitoring gene therapy with reporter gene expression. Semin Nuclear Med. 2001; 31: 312–320.
- ↵Wu JC, Sundaresan G, Iyer M, et al. Noninvasive optical imaging of firefly luciferase reporter gene expression in skeletal muscles of living mice. Mol Ther. 2001; 4: 297–306.
- ↵Gambhir SS, Bauer E, Black ME, et al. A mutant herpes simplex virus type 1 thymidine kinase reporter gene shows improved sensitivity for imaging reporter gene expression with positron emission tomography. Proc Natl Acad Sci U S A. 2000; 97: 2785–2790.
- ↵Yang Y, Li Q, Ertl HC, et al. Cellular and humoral immune responses to viral antigens create barriers to lung-directed gene therapy with recombinant adenoviruses. J Virol. 1995; 69: 2004–2015.
- ↵Franz WM, Rothmann T, Frey N, et al. Analysis of tissue-specific gene delivery by recombinant adenoviruses containing cardiac-specific promoters. Cardiovasc Res. 1997; 35: 560–566.
- ↵Cherry SR, Gambhir SS. Use of positron emission tomography in animal research. ILAR J. 2001; 42: 219–232.
- ↵Yu Y, Annala AJ, Barrio JR, et al. Quantification of target gene expression by imaging reporter gene expression in living animals. Nat Med. 2000; 6: 933–937.