Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print June 21, 2002, 10.1148/radiol.2242011589
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2242011589v1
224/2/446    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Mahmood, U.
Right arrow Articles by Weissleder, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mahmood, U.
Right arrow Articles by Weissleder, R.
(Radiology 2002;224:446-451.)
© RSNA, 2002

Experimental Studies

Feasibility of in Vivo Multichannel Optical Imaging of Gene Expression: Experimental Study in Mice1

Umar Mahmood, MD, PhD, Ching-Hsuan Tung, PhD, Yi Tang, MD and Ralph Weissleder, MD, PhD

1 From the Center for Molecular Imaging Research, CNY149-5403, Department of Radiology, Massachusetts General Hospital, Bldg 149, 13th St, Charlestown, MA 02129. Received September 26, 2001; revision requested December 3; revision received December 17; accepted January 7, 2002. U.M. supported in part by a research grant from the RSNA Research and Education Foundation. Address correspondence to U.M. (e-mail: mahmood@helix.mgh.harvard.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
PURPOSE: To develop and test a multichannel reflectance imaging system for small animals on the basis of a previously developed single-channel setup.

MATERIALS AND METHODS: The imaging system was composed of modular parts, including a light source, excitation filters, emission filters, and a charged-coupled device for recording images. On the basis of generated excitation and absorption spectra of green fluorescent protein (GFP), tricarbocyanine 5.5 (Cy5.5), and indocyanine green (ICG), filters were selected to allow spectral separation and optimize resultant recorded signal. The system was tested by using a combination of the fluorochromes to confirm spectral separation. In vivo tests were performed in nude mice with tumors that expressed cathepsin B, which could be evaluated by using a Cy5.5-based activatable probe and GFP. For each in vivo tumor type and channel, statistical analysis was performed on the basis of signal intensity in the region of interest.

RESULTS: The different fluorochromes were readily distinguished with the system; characteristics such as power were determined for all wavelengths. The system demonstrated a linear response for GFP, a monotonic response for Cy5.5 over a range of more than three orders of magnitude of concentration, and a more complex response for ICG. In vivo analysis demonstrated the ability to image GFP expression and cathepsin B expression separately in tumors: As expected, marked differences were observed in GFP-expression imaging between tumor types (1,363 arbitrary units [AU] ± 236 [SD] vs 110 AU ± 11 for GFP-positive and GFP-negative tumors, respectively; P < .001), whereas similar cathepsin B expression (1,070 AU ± 285 vs 1,168 AU ± 367; P > .5) was observed. Histologic analysis confirmed in vivo findings.

CONCLUSION: Imaging multiple gene expressions simultaneously in vivo by using optical imaging is feasible.

© RSNA, 2002

Index terms: Experimental study • Genes and genetics • Neoplasms, experimental studies


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Currently, much of disease imaging is based on changes in anatomy; these changes typically occur over a relatively long time, often measured in months. Noninvasive imaging with light photons in the near-infrared region of the electromagnetic spectrum represents an intriguing new avenue for extracting biologic parameters that often cannot be measured with conventional imaging, which potentially allows earlier detection of disease and earlier assessment of the effectiveness of therapy. Light photons interact with tissues and may be used to determine physical characteristics of the tissue, analogous to computed tomography, magnetic resonance imaging (1–4), or ultrasonography (5,6). The addition of targeted (7–10) and activatable (11–13) near-infrared fluorochromes, red-shifted fluorescent proteins, or bioluminescent probes (14,15) aids in extracting additional molecular parameters in vivo.

The goal of this study was to develop and test a multichannel reflectance imaging system for small animals on the basis of a previously developed setup (16). We hypothesized that, analogous to multichannel fluorescence microscopy (which typically uses fluorochromes in the visible region), near-infrared multichannel imaging should be feasible in experimental in vivo settings and ultimately in clinical settings, as well. The major advantages of imaging different biologic targets simultaneously and independently include the ability to (a) colocalize multiple targets, (b) probe for differential expression levels of multiple targets, (c) analyze the combination of expression levels of particular importance in cancer, where one target alone is rarely overexpressed, (d) develop miniarrays for in vivo target assessment, (e) image the temporal and spatial correlation of distinct enzymatic pathways in disease, (f) image the effects on a pathway of therapy targeting another pathway, and (g) provide reference channels for activatable "smart" sensors.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Imaging System Design
The whole-body mouse imaging system on which the current system was based has been previously described as a single-channel imaging system (16). Multichannel imaging places additional requirements on the system; the major differences are the additional multiple excitation and emission bandpasses that have additional optimization requirements (and different wavelengths) and the resultant changes in acquisition parameters.

Overall, the imaging system consists of multiple separate functional components encased in a black light-tight box:(a) broadband white-light (halogen) source, (b) selective excitation filters, (c) imaging chamber for the animal, (d) emission filters to select fluorescence, (e) charged-coupled device for recording emission photons, and (f) computer for image acquisition control and storage. The excitation and emission filters are readily exchangeable, with switching times between channels currently approximately a few seconds. To maximize spectral purity, all filters are five-cavity interference filters (Omega Optical, Brattleboro, Vt). All excitation filters are 25 mm in diameter, and all emission filters are 50 mm in diameter.

The absorption and emission spectra for all the fluorochromes of interest were obtained: intracellular enhanced green fluorescent protein (GFP), tricarbocyanine 5.5 (Cy5.5), and indocyanine green (ICG) spectra were determined with the use of a fluorometer (U-4500; Hitachi, Tokyo, Japan) (Fig 1); fluorescein absorption and emission spectra were also obtained (not shown for clarity of Figure). The choice of bandpass filter systems was based on the observed wavelengths of the fluorochromes. The filter sets (especially for GFP and Cy5.5) were chosen to maximize signal received from the fluorescent molecules—that is, emission filter bandpasses were matched to the emission spectra, whereas excitation bandpasses were shifted to shorter wavelengths than peak excitation frequency to decrease cross-talk between excitation and emission channels. The exact center and bandpass wavelengths for the filter sets are detailed in the Table and are also depicted graphically in Figure 1.



View larger version (60K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Graph shows fluorochrome spectra and selected bandpass filters. Absorption (gray) and emission (black) spectra for (from left to right) GFP, Cy5.5, and ICG. Bandpass cutoff for the excitation filters (blue shading) and emission filters (red shading) for each fluorochrome are also shown.

 

View this table:
[in this window]
[in a new window]

  		Filter Wavelengths, Power at Source, and Photon Flux at Imaging Surface for the Three Imaging Channels

 
Peak fluorescein excitation and emission of 490 nm and 525 nm, respectively, are similar enough to the excitation and emission spectra of GFP to allow the use of the chosen GFP filter set (Table) for fluorescein imaging. The longest wavelength channel was designed for possible use with multiple near-infrared dyes with excitation and emission wavelengths longer than those of Cy5.5, with fluorescence wavelengths approximately in the 800-nm range; ICG is the prototype near-infrared dye that may be used in this third channel. Power measurements were obtained with a LaserCheck instrument (Coherent, Auburn, Calif) at the appropriate wavelengths immediately distal to the excitation filter and at the imaging surface on which the animal was placed.

Fluorochromes
Four fluorochromes were used in this study as primary reporters: fluorescein (Sigma, St Louis, Mo), enhanced GFP, Cy5.5 (Amersham-Pharmacia, Piscataway, NJ), and ICG (Akorn, Buffalo Grove, Ill). ICG and fluorescein were used as examples of nontargeted reporter probes, while GFP was expressed in a gliosarcoma cell line as a fluorescent optical marker gene in tumor cells. Cy5.5 was used as an enzyme-responsive reporter to measure cathepsin B enzyme activity (12). This probe belongs to a group of smart probes in which the fluorochrome is activated (ie, markedly increases in fluorescence) with enzymatic cleavage (17).

GFP-expressing Cell Line
The 9L rodent glioma cell line and a stably transfected clone that overexpresses GFP (18) were cultured in Dulbecco modified Eagle medium supplemented with 10% fetal calf serum and 1% penicillin and streptomycin (all products from Cellgro Mediatech, Washington, DC). The GFP-positive clone was selected by using the antibiotic G418 (Cellgro Mediatech), since the plasmid encoding for GFP also contained a gene for resistance to G418. For in vitro studies, cells were resuspended in Hanks buffered saline without phenol red.

In Vitro and Phantom Evaluation
The imaging system was tested by using phantoms consisting of 96-well cell culture plates (Corning, Corning, NY). For evaluating the effectiveness of spectral separation, wells containing 6 x 106 9L GFP-positive glioma cells, 100 µL of 100 µmol/L (10 nmol) free Cy5.5 dye, or 100 µL of 100 µmol/L (10 nmol) free ICG were used. White-light images were obtained at 0.075 second (minimum system acquisition time). Fluorescence imaging times of 30 seconds per channel were used, although an excellent signal-to-noise ratio can be achieved in a fraction of this time (3–5 seconds) at these concentrations. Samples were evaluated by using a dilution series from 100 µmol (100 µL x 1 mmol/L) through 0.01 nmol (100 µL x 0.1 µmol/L) for all three fluorochromes (fluorescein, Cy5.5, and ICG) by using acquisition times (for all dilutions) of 15, 24, and 90 seconds for the GFP and/or fluorescein, Cy5.5, and ICG channels, respectively. Fluorescein was used to evaluate the GFP channel for the serial dilution experiments, because unlike intracellular GFP, molar concentration is readily determined. To additionally evaluate the GFP channel more directly, 9L GFP-positive and 9L GFP-negative cells were placed in a 96-well plate with a 16x concentration range evaluated at twofold serial increments in cell number from 62,500 to 1,000,000 cells per well.

Animal Experiments
For in vivo evaluation of the imaging system and demonstration of multichannel imaging of different gene expressions, 105 tumor cells were implanted into the subcutaneous tissues of the lower anterior abdominal walls of four nude mice (nu/nu Taconic, Germantown, NY), with 9L GFP-positive glioma cells injected on the animals’ right side and 9L GFP-negative glioma cells injected on the animals’ left side. Animals were imaged 24 hours after intravenous injection of 2 nmol of cathepsin B enzyme–activated probe (containing Cy5.5; see "Fluorochromes," earlier in article) when the tumors were approximately 4 mm in diameter; cathepsin B was expressed in both tumors. In vivo imaging times were 0.075 second for the white-light anatomic image, 2 minutes for the GFP image, and 30 seconds for the cathepsin B enzyme activity image (measured by means of Cy5.5 fluorescence).

Approval was received for this study from our institution’s animal care committee. Ketamine (subcutaneous administration of 9 mg per 100 g) and xylazine (subcutaneous administration of 0.9 mg per 100 g) were used as an anesthetic, and pentobarbital (intraperitoneal administration of 20 mg per 100 g) was used for sacrifice. For histologic evaluation, animals were euthanized, and tumors were removed and snap frozen in liquid nitrogen. Sections remained unstained for fluorescence microscopic evaluation. By using standard fluorescent microscopy filter sets (Omega Optical) for GFP and Cy5.5, the same sections were evaluated by using white-light, GFP, and Cy5.5 channels.

Image and Statistical Analysis
All regions of interest (ROIs), both in vitro and in vivo, were placed by one author (U.M.). For GFP-positive and GFP- negative cell phantom experiments and for free fluorochrome experiments, a circular ROI was placed that encompassed the entire well for each cell concentration. Single mean values for signal intensity were obtained for each well, and standard linear regression was performed to correlate signal intensity with cell count per well.

For in vivo evaluation, mean signal intensity values were obtained by using ROI measurements obtained immediately over the region of GFP-positive and GFP-negative tumors for both GFP-channel and Cy5.5-channel images for each tumor. Because of the scattering of light through tissue that resulted in a penumbra surrounding the tumors, a same-sized circular ROI was placed over the area of the tumor with the highest signal intensity. The smaller size of the ROI compared with the tumor size was chosen to result in less variability between measurements of the same tumor and of different tumors. ROI size was 177 pixels. Mean values and SDs of signal intensity for each tumor type and for each channel were then calculated for the four animals collectively on the basis of the mean ROI intensities calculated for each tumor type and channel. Significance was determined by using the Student t test and reported as P values.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Fluorochrome Spectra and Choice of Filter Systems
Figure 2, a series of images from a 96-well plate containing a pellet of 9L cells expressing GFP, free Cy5.5 dye, and free ICG dye in different wells, demonstrates that the designed in vivo system allows very clear separation between channels. The normalized mean signal intensities for the three targets in the different channels were as follows: for the GFP channel, GFP = 100 arbitrary units (AU), Cy5.5 = 1.9 AU, and ICG = 0.7 AU; for the Cy5.5 channel, GFP = 1.1 AU, Cy5.5 = 100 AU, and ICG = -0.3 AU; and for the ICG channel, GFP = 0.6 AU, Cy5.5 = 5.7 AU, and ICG = 100 AU. Note that excellent discrimination is seen despite high concentrations of the other fluorochromes.



View larger version (145K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a. Spectral separation of imaging channels. (a) White-light, (b) GFP-channel, (c) Cy5.5-channel, and (d) ICG-channel images of a 96-well plate containing GFP-expressing cells (upper left well, all four images), free Cy5.5 (middle well, all four images), and free ICG (lower right well, all four images). Images demonstrate clear channel separation despite high concentration of fluorochromes in the three wells.

 


View larger version (90K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b. Spectral separation of imaging channels. (a) White-light, (b) GFP-channel, (c) Cy5.5-channel, and (d) ICG-channel images of a 96-well plate containing GFP-expressing cells (upper left well, all four images), free Cy5.5 (middle well, all four images), and free ICG (lower right well, all four images). Images demonstrate clear channel separation despite high concentration of fluorochromes in the three wells.

 


View larger version (89K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2c. Spectral separation of imaging channels. (a) White-light, (b) GFP-channel, (c) Cy5.5-channel, and (d) ICG-channel images of a 96-well plate containing GFP-expressing cells (upper left well, all four images), free Cy5.5 (middle well, all four images), and free ICG (lower right well, all four images). Images demonstrate clear channel separation despite high concentration of fluorochromes in the three wells.

 


View larger version (88K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2d. Spectral separation of imaging channels. (a) White-light, (b) GFP-channel, (c) Cy5.5-channel, and (d) ICG-channel images of a 96-well plate containing GFP-expressing cells (upper left well, all four images), free Cy5.5 (middle well, all four images), and free ICG (lower right well, all four images). Images demonstrate clear channel separation despite high concentration of fluorochromes in the three wells.

 
The excitation source in the current system is a 150-W halogen lamp. The Table summarizes the power output for the current configuration of the different bandpass filters. Measurements obtained immediately distal to the excitation filter and proximal to the fiberoptic light guides show the efficiency of the filter, including the bandwidth chosen. Measurements obtained at the surface of the platen show the excitation photon flux seen at the surface of the animal. The imaging platen size is 450 cm2; homogeneity is achieved over the surface by means of light diffusers, with additional postprocessing "masks" further correcting any field inhomogeneity.

Correlation between Fluorochrome Concentration and Signal Intensity
Figure 3 summarizes the correlation between fluorochrome concentration and normalized fluorescence signal intensity obtained with use of the imaging system. As is evident from the data, there is a monotonic increase in signal intensity with fluorescein and Cy5.5 over a range of more than three orders of magnitude of concentration. At the highest concentration tested (1 mmol/L), quenching of Cy5.5 is observed, as expected. The graph also demonstrates some of the idiosyncrasies of ICG, which is known to demonstrate considerable concentration-dependent shifts and changes in the shape of absorption and emission spectra (19). Hence, dyes other than ICG may be more useful in multichannel optical imaging in combination with Cy5.5.



View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Graph shows normalized signal intensity versus concentration of free fluorochrome. Calibration curves for fluorescein (FITC), Cy5.5, and ICG from 0.1 µmol/L to 1 mmol/L demonstrate monotonically increased signal intensity through 300-µmol/L concentration for fluorescein and Cy5.5 (ie, a clinically relevant range). The system response for ICG is more complex and reflects the marked changes in spectral properties of ICG with variations in concentration.

 
To further demonstrate the linearity of response of the system and the adequacy of the filter combinations for the GFP channel, 9L GFP-expressing and 9L control cells were evaluated in a 96-well plate over a 16x concentration range. Figure 4a shows the relative lack of signal from the control cells and the graded signal intensity from the GFP-expressing cells. Figure 4b demonstrates the linearity of response over the range evaluated and shows that "bleed-through" of excitation light has been avoided. Linear regression yields r2 values of 0.997 and 0.966 for GFP-positive and GFP-negative cells, respectively. The slopes of the regression are 24E-4 and 1.8E-4, a 13-fold difference. The lowest concentration tested, approximately 60,000 cells, demonstrates a difference in mean signal intensity (minus background) of 6.5-fold between GFP-positive and GFP-negative cells.



View larger version (71K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a. Linearity of system response to GFP concentration. (a) GFP-channel image of 96-well plate with 9L glioma cells expressing (top row) or not expressing (bottom row) GFP (middle row blank). Serial twofold decreases in cell number from left to right, from 1 million cells to 62,500 cells. (b) Graph shows signal intensity versus cell number from image a. Note the linear increase in signal intensity from GFP-positive cells (solid line) and (as expected) relative lack of signal intensity from GFP-negative cells (dotted line).

 


View larger version (18K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b. Linearity of system response to GFP concentration. (a) GFP-channel image of 96-well plate with 9L glioma cells expressing (top row) or not expressing (bottom row) GFP (middle row blank). Serial twofold decreases in cell number from left to right, from 1 million cells to 62,500 cells. (b) Graph shows signal intensity versus cell number from image a. Note the linear increase in signal intensity from GFP-positive cells (solid line) and (as expected) relative lack of signal intensity from GFP-negative cells (dotted line).

 
In Vivo Imaging
A typical and representative imaging experiment is summarized in Figure 5a: the first image was obtained with nonselective white light, which provides an anatomic map. The second image is a map of cathepsin B expression; since both tumors express cathepsin B, increased signal intensity is seen in this channel with both tumors. The third image is a GFP expression map; as expected, only the GFP-positive tumor is visualized. The fourth image is a computed image obtained by dividing GFP expression by cathepsin B expression and represents a map of the two gene expression ratios across the animal. In all four mice, strong signal intensity was observed in the GFP channel in the GFP-positive tumors (1,363 AU ± 236), whereas minimal (comparable to background) signal intensity (110 AU ± 11) was seen in the GFP-negative tumors (P < .001). As expected, all four mice demonstrated strong signal intensity in the cathepsin B channel for both tumors (1,070 AU ± 285 and 1,168 AU ± 367 for GFP-positive and GFP-negative tumors, respectively; P > .5). Correlative multichannel fluorescence microscopic images of the tumor samples are shown in Figure 5b, confirming the above observations.



View larger version (76K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a. In vivo images of GFP and cathepsin B expression. (a) Representative images from a nude mouse obtained 24 hours after intravenous administration of cathepsin B-sensitive imaging probe. Genetically identical 9L glioma tumors, except for GFP expression (GFP-positive tumor on animal’s right flank, GFP-negative tumor on animal’s left flank), were implanted in the lower abdominal wall; both tumors express cathepsin B. From left to right: white-light, cathepsin B (CaB), GFP, and computed image (GFP/CaB) of the ratio of GFP expression to cathepsin B expression. The different wavelength images clearly depict separate gene expressions in vivo. (b) Histologic evaluation of GFP-positive tumor by using white light (first image), GFP fluorescence microscopy (second image, GFP), Cy5.5 fluorescence microscopy (third image, CaB), and false-color combination (fourth image), with GFP signal in green and Cy5.5 signal in red confirming both gene expressions and demonstrating the partially extracellular nature of cathepsin B expression. T = location of tumor.

 


View larger version (54K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b. In vivo images of GFP and cathepsin B expression. (a) Representative images from a nude mouse obtained 24 hours after intravenous administration of cathepsin B-sensitive imaging probe. Genetically identical 9L glioma tumors, except for GFP expression (GFP-positive tumor on animal’s right flank, GFP-negative tumor on animal’s left flank), were implanted in the lower abdominal wall; both tumors express cathepsin B. From left to right: white-light, cathepsin B (CaB), GFP, and computed image (GFP/CaB) of the ratio of GFP expression to cathepsin B expression. The different wavelength images clearly depict separate gene expressions in vivo. (b) Histologic evaluation of GFP-positive tumor by using white light (first image), GFP fluorescence microscopy (second image, GFP), Cy5.5 fluorescence microscopy (third image, CaB), and false-color combination (fourth image), with GFP signal in green and Cy5.5 signal in red confirming both gene expressions and demonstrating the partially extracellular nature of cathepsin B expression. T = location of tumor.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
Signal observed with optical-based imaging techniques may reflect intrinsic physical properties of tissues and may be based on scatter, absorption, impedance, or fluorescence; the addition of extrinsic fluorochromes as optical "contrast agents" potentially allows for molecular imaging by coupling the fluorescent signal with a molecular process in vivo. There are two main techniques that are used for in vivo fluorescence imaging: reflectance (surface-weighted) imaging and tomographic imaging (20), which can be divided into techniques such as those that rely on early arriving photons (21) to avoid more scattered photons, those that use continuous light and take advantage of the scattered photons in their reconstructive algorithms (3), and those that modulate light to localize fluorescence on the basis of interference mechanisms (22). Both reflectance and tomographic systems are relevant to the future use of optical imaging in humans. Reflectance systems are strongly surface weighted; however, many diseases in humans are near a surface, including diseases of the colon, esophagus, bronchi, peritoneum (during endoscopy or laparoscopy), and axilla. Tomographic systems, which have been used to penetrate more than 10 cm of soft tissue (20), lend themselves to imaging deeper structures such as the breast (20), prostate, and brain (1,2).

The multichannel paradigm places additional constraints on the system when compared with single-channel systems. In the current study, the filter sets (especially for GFP and Cy5.5) were chosen to maximize signal received from the fluorescent molecules—that is, emission filter bandpasses were matched with the emission spectra. Because of the relatively small Stokes shifts between excitation and emission (typically on the order of 20 nm), a conscious decision was made to excite the fluorochromes at slightly lower wavelengths than their maximum absorption to decrease the likelihood of excitation photon bleed-through into the final image. The decreased fluorescence yield from exciting at a wavelength that is lower than the maximum can be overcome by increasing excitation photon flux or by increasing the sensitivity of the receiving charged-coupled device. The former method is easier and less costly than the latter and was used in the current configuration. Moreover, relatively wide bandpasses (30–40 nm) were chosen to maximize both excitation photon flux and the number of emission photons that were seen with the charged-coupled device. Despite the five-cavity interference mechanism and relatively sharp cutoffs, the bandpass filters have relatively long, albeit small, tails. This is much less of an issue in histologic fluorescence settings, given the markedly smaller volume of background tissue providing a reflective surface for bleed-through when compared with the in vivo setting. Future improvements include the addition of polarizing filters to further separate excitation and emission bands, thus allowing closer frequency spacing of excitation and emission wavelengths. More imaging channels may also be added, including Cy3.5 between GFP and Cy5.5 and channels for dyes further in the near-infrared region of the spectrum beyond ICG.

The current multichannel configuration can be readily applied to tomographic systems. An additional optical switch in series with those typically present to change excitation location can be implemented to change source wavelength. Mechanical, manual, or optical switching of emission filters would also be required. Thus, it is expected that for many applications, more than one biologic parameter will be interrogated. Rational probe design for the multichannel paradigm will be disease specific: Breast cancer, for example, may have activatable probes for extracellular proteases combined with activatable probes at a second wavelength for intracellular kinases. Such combinations, which are essentially targeted in vivo miniarrays, will increase the sensitivity and specificity of disease detection and will allow more specific follow-up of antineoplastic therapy.

Practical applications: A number of the most common diseases that affect people today, including many cancers, cardiovascular diseases, inflammatory diseases, neurologic disorders, and some degenerative disorders, may potentially be better assessed at an earlier stage in the future when optical imaging is incorporated into the clinical evaluation–imaging paradigm. Multichannel optical imaging further expands the utility of this modality by allowing the spatial distribution of biologic parameters, including different gene expressions, to be determined nearly simultaneously. Since many pathologic conditions, especially predisease states, fall outside of simple one-gene Mendelian genetics, a multiple-gene imaging approach may help in our quest for earlier detection.


   ACKNOWLEDGMENTS
 
The authors thank Anna Moore, PhD, for the cell lines, and Wellington Pham, PhD, for assistance with chemistry.


   FOOTNOTES
 
Abbreviations: AU = arbitrary unit, Cy5.5 = tricarbocyanine 5.5, GFP = green fluorescent protein, ICG = indocyanine green, ROI = region of interest

Author contributions: Guarantor of integrity of entire study, U.M.; study concepts and design, U.M., R.W.; literature research, U.M.; experimental studies, all authors; data acquisition, U.M., Y.T.; data analysis/interpretation, U.M.; statistical analysis, U.M.; manuscript preparation, U.M.; manuscript definition of intellectual content and editing, U.M., R.W.; manuscript revision/review, all authors; manuscript final version approval, U.M., R.W.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 

  1. Boas DA, Gaudette T, Strangman G, Cheng X, Marota JJ, Mandeville JB. The accuracy of near infrared spectroscopy and imaging during focal changes in cerebral hemodynamics. Neuroimage 2001; 13:76-90.[Medline]
  2. Benaron DA, Hintz SR, Villringer A, et al. Noninvasive functional imaging of human brain using light. J Cereb Blood Flow Metab 2000; 20:469-477.[Medline]
  3. Ntziachristos V, Hielscher AH, Yodh AG, Chance B. Diffuse optical tomography of highly heterogeneous media. IEEE Trans Med Imaging 2001; 20:470-478.[CrossRef][Medline]
  4. Ramanujam N, Vishnoi G, Hielscher A, Rode M, Forouzan I, Chance B. Photon migration through fetal head in utero using continuous wave, near infrared spectroscopy: clinical and experimental model studies. J Biomed Opt 2000; 5:173-184.[CrossRef][Medline]
  5. Fujimoto JG, Boppart SA, Tearney GJ, Bouma BE, Pitris C, Brezinski ME. High resolution in vivo intra-arterial imaging with optical coherence tomography. Heart 1999; 82:128-133.[Abstract/Free Full Text]
  6. Fujimoto JG, Pitris C, Boppart SA, Brezinski ME. Optical coherence tomography: an emerging technology for biomedical imaging and optical biopsy. Neoplasia 2000; 2:9-25.[CrossRef][Medline]
  7. Licha K, Hessenius C, Becker A, et al. Synthesis, characterization, and biological properties of cyanine-labeled somatostatin analogues as receptor-targeted fluorescent probes. Bioconjug Chem 2001; 12:44-50.[CrossRef][Medline]
  8. Becker A, Hessenius C, Licha K, et al. Receptor-targeted optical imaging of tumors with near-infrared fluorescent ligands. Nat Biotechnol 2001; 19:327-331.[CrossRef][Medline]
  9. Bugaj JE, Achilefu S, Dorshow RB, Rajagopalan R. Novel fluorescent contrast agents for optical imaging of in vivo tumors based on a receptor-targeted dye-peptide conjugate platform. J Biomed Opt 2001; 6:122-133.[CrossRef][Medline]
  10. Achilefu S, Dorshow RB, Bugaj JE, Rajagopalan R. Novel receptor-targeted fluorescent contrast agents for in vivo tumor imaging. Invest Radiol 2000; 35:479-485.[CrossRef][Medline]
  11. Bremer C, Tung CH, Weissleder R. In vivo molecular target assessment of matrix metalloproteinase inhibition. Nat Med 2001; 7:743-748.[CrossRef][Medline]
  12. Weissleder R, Tung CH, Mahmood U, Bogdanov A, Jr. In vivo imaging of tumors with protease-activated near-infrared fluorescent probes. Nat Biotechnol 1999; 17:375-378.[CrossRef][Medline]
  13. Tung CH, Mahmood U, Bredow S, Weissleder R. In vivo imaging of proteolytic enzyme activity using a novel molecular reporter. Cancer Res 2000; 60:4953-4958.[Abstract/Free Full Text]
  14. Contag CH, Jenkins D, Contag PR, Negrin RS. Use of reporter genes for optical measurements of neoplastic disease in vivo. Neoplasia 2000; 2:41-52.[CrossRef][Medline]
  15. Contag PR, Olomu IN, Stevenson DK, Contag CH. Bioluminescent indicators in living mammals. Nat Med 1998; 4:245-247.[CrossRef][Medline]
  16. Mahmood U, Tung CH, Bogdanov A, Jr, Weissleder R. Near-infrared optical imaging of protease activity for tumor detection. Radiology 1999; 213:866-870.[Abstract/Free Full Text]
  17. Tung CH, Bredow S, Mahmood U, Weissleder R. Preparation of a cathepsin D sensitive near-infrared fluorescence probe for imaging. Bioconjug Chem 1999; 10:892-896.[CrossRef][Medline]
  18. Moore A, Marecos E, Simonova M, Weissleder R, Bogdanov A, Jr. Novel gliosarcoma cell line expressing green fluorescent protein: a model for quantitative assessment of angiogenesis. Microvasc Res 1998; 56:145-153.[CrossRef][Medline]
  19. Landsman ML, Kwant G, Mook GA, Zijlstra WG. Light-absorbing properties, stability, spectral stabilization of indocyanine green. J Appl Physiol 1976; 40:575-583.[Abstract/Free Full Text]
  20. Ntziachristos V, Yodh AG, Schnall M, Chance B. Concurrent MRI and diffuse optical tomography of breast after indocyanine green enhancement. Proc Natl Acad Sci U S A 2000; 97:2767-2772.[Abstract/Free Full Text]
  21. Wu J, Perelman L, Dasari RR, Feld MS. Fluorescence tomographic imaging in turbid media using early-arriving photons and Laplace transforms. Proc Natl Acad Sci U S A 1997; 94:8783-8788.[Abstract/Free Full Text]
  22. Chance B. Near-infrared images using continuous, phase-modulated, and pulsed light with quantitation of blood and blood oxygenation. Ann N Y Acad Sci 1998; 838:29-45.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 RadiologyHome page
M. S. Gee, R. Upadhyay, H. Bergquist, H. Alencar, F. Reynolds, M. Maricevich, R. Weissleder, L. Josephson, and U. Mahmood
Human Breast Cancer Tumor Models: Molecular Imaging of Drug Susceptibility and Dosing during HER2/neu-targeted Therapy
Radiology, September 1, 2008; 248(3): 925 - 935.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
R. Upadhyay, R. A. Sheth, R. Weissleder, and U. Mahmood
Quantitative Real-time Catheter-based Fluorescence Molecular Imaging in Mice
Radiology, November 1, 2007; 245(2): 523 - 531.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
H. Alencar, M. A. Funovics, J. Figueiredo, H. Sawaya, R. Weissleder, and U. Mahmood
Colonic Adenocarcinomas: Near-Infrared Microcatheter Imaging of Smart Probes for Early Detection--Study in Mice
Radiology, July 1, 2007; 244(1): 232 - 238.
[Abstract] [Full Text] [PDF]

Home page
 Physiol. GenomicsHome page
Y. Wang, M. Iyer, A. Annala, L. Wu, M. Carey, and S. S. Gambhir
Noninvasive indirect imaging of vascular endothelial growth factor gene expression using bioluminescence imaging in living transgenic mice
Physiol Genomics, January 12, 2006; 24(2): 173 - 180.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
M. A. Funovics, R. Weissleder, and U. Mahmood
Catheter-based in Vivo Imaging of Enzyme Activity and Gene Expression: Feasibility Study in Mice
Radiology, June 1, 2004; 231(3): 659 - 666.
[Abstract] [Full Text] [PDF]

Home page
 CirculationHome page
R. M. Botnar, A. S. Perez, S. Witte, A. J. Wiethoff, J. Laredo, J. Hamilton, W. Quist, E. C. Parsons Jr, A. Vaidya, A. Kolodziej, et al.
In Vivo Molecular Imaging of Acute and Subacute Thrombosis Using a Fibrin-Binding Magnetic Resonance Imaging Contrast Agent
Circulation, April 27, 2004; 109(16): 2023 - 2029.
[Abstract] [Full Text] [PDF]

Home page
 RadiologyHome page
J. H. Thrall
Nanotechnology and Medicine
Radiology, February 1, 2004; 230(2): 315 - 318.
[Full Text] [PDF]

Home page
 Cancer Res.Home page
M. F. Kircher, U. Mahmood, R. S. King, R. Weissleder, and L. Josephson
A Multimodal Nanoparticle for Preoperative Magnetic Resonance Imaging and Intraoperative Optical Brain Tumor Delineation
Cancer Res., December 1, 2003; 63(23): 8122 - 8125.
[Abstract] [Full Text] [PDF]

Home page
 ScienceHome page
H. R. Herschman
Molecular Imaging: Looking at Problems, Seeing Solutions
Science, October 24, 2003; 302(5645): 605 - 608.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
L. M. Matrisian, G. W. Sledge Jr., and S. Mohla
Extracellular Proteolysis and Cancer: Meeting Summary and Future Directions
Cancer Res., October 1, 2003; 63(19): 6105 - 6109.
[Abstract] [Full Text] [PDF]

Home page
 Molecular Cancer TherapeuticsHome page
U. Mahmood and R. Weissleder
Near-Infrared Optical Imaging of Proteases in Cancer
Mol. Cancer Ther., May 1, 2003; 2(5): 489 - 496.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2242011589v1
224/2/446    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Mahmood, U.
Right arrow Articles by Weissleder, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mahmood, U.
Right arrow Articles by Weissleder, R.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE