Return to Cover of Wave to the Future

Address for correspondence:
Lisa A. Cassis, Ph.D.
Associate Professor
Division of Pharmacology and Experimental Therapeutics
Room 417, College of Pharmacy
Lexington, KY 40536-0082

Running Title: Near-infrared and infrared imaging

Acknowledgements: This research was supported by NIH HL52934

Abstract

Near-infrared (1000 - 3000 nm) spectrometry, which employs an external light source for determination of chemical composition, has been previously utilized for industrial determination of the fat content of commercial meat products, for in vivo determination of body fat, and in our laboratories for determination of lipoprotein composition in carotid artery atherosclerotic plaques. Infrared (3000 - 5000 nm) imaging, which measures the surface temperature of objects, was previously examined for quantitation of energy expenditure. However, no previous studies have examined combined near-IR/IR imaging as a tool for the determination of energy expenditure. The purpose of this study was to utilize near-IR and IR spectrometry under a variety of experimental conditions for determination of superficial lipid composition and surface heat emission in rats. Results demonstrate that near-IR spectrometry was capable of predicting the genotype of Zucker obese rats before the development of obesity (3 and 17 days of age). Pharmacologic analysis of the dose-dependent effects of norepinephrine (NE) and angiotensin II (AII) administration on superficial interscapular lipid composition by near-IR spectrometry demonstrated that both agents mobilized lipids albeit with different lipid spectral profiles. Using a tunable-range video camera with different band pass filters, near-IR imaging of adult obese rats demonstrated increases in superficial lipid composition. Moreover, IR imaging with the tunable-range video camera allowed for regional-specific measurements of surface heat emission in conscious, freely moving rats. Following treatment with isoproterenol (ISO) and AII, tail heat emission increased, but dorsal body heat emission differed between the two pharmacologic agents. Chronic administration of low doses of AII resulted in an increase in dorsal surface heat emission. Collectively, these results demonstrate that with a single-instrument, indexes of superficial lipid composition and surface heat emission can be obtained quickly and noninvasively for the analysis of energy expenditure.

Introduction

Near-infrared (IR) spectrometry has been used industrially for several years to determine saturation of unsaturated fatty acid esters (1). Near-IR spectrometry uses an tunable light source external to the experimental subject to determine its chemical composition. Industrial utilization of near-IR will allow for the in vivo measurement of the tissue-specific rate of oxygen utilization as an indirect estimate of energy expenditure. However, assessment of regional oxygen consumption by these methods is complex, requiring a high level of surgical skill for implantation of indwelling catheters to isolate the organ under study.

A major advantage of near-IR spectral analysis is its chemical imaging ability. Additionally, near-IR spectral imaging provides information on details of various internal structures including muscle, bone, and arteries (14). Using a specially modified infrared (IR) video camera, different band pass filters permit the measurement of temperature (+ 0.03 C) by black body heat emission (IR analysis) or by collection of image spectra over the range of 1000 - 4000 nm (near-IR and IR spectral range). In contrast to near-IR imaging spectrometry, which requires an external light source, infrared imaging is based on internally generated heat radiated as light. Measurements can be made quickly on conscious, freely moving animals placed in a thermoneutral environment. Theoretically, the use of an IR video camera modified to work in the near-IR range should allow for spatially resolved noninvasive measurements of surface temperature (as an index of energy expenditure) and superficial lipid composition (as an index of lipid metabolism) (6). The purpose of this study was to determine if near-IR spectrometry and infrared imaging of rats could be used to noninvasively examine regional subcutaneous lipid composition and surface energy expenditure, respectively. In this study, prototype pharmacologic agents with known effects on thermogenesis were administered to rats to demonstrate the effectiveness of near-infrared and infrared techniques. Results from this study demonstrate that combined near-IR and IR imaging with a tunable-range video camera detected drug-induced alterations in superficial lipid composition and heat emission.

Animals. Adult (250-300 g) male Sprague Dawley rats (Harlan Laboratories, Indianapolis) were housed in an AAALAC approved animal facility under normal light/dark conditions with free access to food and water. For studies examining acute drug administration, two separate routes of drug administration were examined, namely injection (s.c.) of anesthetized rats (ketamine, 120 mg/kg; acepromazine, 12 mg/kg) with drug followed by imaging as described below. Alternatively, rats were anesthetized and femoral artery and venous catheters implanted for measurement of mean arterial pressure and drug administration, respectively.

For studies imaging obese rats, litters from matings of female Zucker obese (fa/fa) rats with male (Fa/?) rats were purchased and received when rat pups were 14 days of age. Rat pups from 4 litters (average of 14 rat pups/litter) were imaged at 17 - 19 days of age. After obtaining body weights, rat pups were euthanized and the right inguinal fat pad removed for phenotypic identification of genotype (15). In a separate study, rat pups from 2 litters of female obese (fa/fa) rats were imaged at 3 days of age, and allowed to grow until 16 weeks of age. Adult rats were weighed, and imaged as described below.

Chronic infusion studies. Rats were anesthetized with ether, a small incision was made in the upper back, and an Alzet minipump (model 2004) filled with an AII (4 mg/ml) solution in sterile saline was inserted subcutaneously. The final rate of AII delivery was approximately 100 ng/kg/min. Preliminary studies, using HPLC separation of angiotensin peptide fragments coupled with AII radioimmunoassay, demonstrated that following 2 weeks at 37C, AII was stable (less than 5% degradation of parent AII) in the osmotic minipump.

Near-IR spectroscopic analysis of lipid metabolism. Initial experiments examined the near-IR lipid profile in the interscapular region of anesthetized adult rats (N = 2). Rats were shaved in the interscapular region, and 3 near-IR scans were obtained with a stationary fiber-optic probe placed immediately over the dorsal interscapular region. Scanning was then performed after removal of the skin overlaying the interscapular region, followed by removal of subcutaneous white adipose tissue, until scanning of the exposed brown adipose tissue was performed.

Near-IR spectroscopic analysis of agonist-induced lipid metabolism. For all studies using acute administration of drug (AII, 200 µg/kg; NE 400 µg/kg, N = 8/drug) to adult rats, anesthetized rats were shaved in the interscapular region, and near-IR spectra of the interscapular region obtained with a stationary fiber-optic probe. The interscapular region was examined due to the localization of the largest source of brown adipose tissue, the interscapular brown fat pad (ISBAT) in this region, and the concomitant localization of superficial white adipose tissue. Three control near-IR scans of the interscapular region were obtained before drug injection. Rats were then injected with either AII, NE or saline vehicle and scans were performed every 4 minutes over a 50 minute period. For studies examining acute infusions of increasing concentrations of drugs (AII, 2 - 200 ng/min; NE, 1 - 100 µg/min; N = 4/drug), mean arterial pressure was continously monitored in the femoral artery catheter with an in-line pressure transducer connected to a recording polygraph. Each dose of drug was infused over a 20 min period, and scans were taken every 4 min. Mean arterial pressure was continuously monitored. Spectral data was analyzed with the BEST (Bootstrap Error-adjusted Single-sample Theory) metric (16) and a Convex Exemplar parallel supercomputer.

For studies in obese rats, unanesthetized rat pups (3 days of age, 17 - 19 days of age) were placed in a small box to eliminate movement, and spectrometric measurements made over the dorsal surface of rat pups in the interscapular region with the stationary fiber-optic near-IR probe. For studies in adult obese and lean rats, rats were imaged using a liquid-nitrogen-cooled InSb focal plane array camera with sound annotation capability.

Infrared Thermal Imaging. Thermal IR imaging was performed using a liquid-nitrogen-cooled InSb focal plane array camera (temperature precision = 0.03^o C; 3000 - 5000 nm) with sound annotation capability. Initial studies examined the effect of acute peripheral administration of either AII (200 µg/kg, s.c.; N = 3), isoproterenol (ISO, 50 µg/kg, s.c.; N = 1) or saline vehicle (N = 3). Before injection, rats were shaved over the dorsal interscapular surface, placed in individual plastic chambers, and 256 x 256 pixel IR emission images were taken every 3 minutes by opening a port in the chamber for the camera lens. A thermometer was used to calibrate the A/D in the camera from 35 - 40 ^oC. Because the rats were shaved to increase the spatial resolution of the camera over organs, the interscapular temperature (> 40^oC) was offscale in all rats used in these initial experiments.

In a separate group of rats, thermal IR black body heat emission was monitored following 10 days of chronic AII or saline infusion (N = 6/group). Using the thermal IR video camera, conscious, unrestrained rats were imaged for 10 - 15 seconds each (25 - 50 C temperature range).

Near-IR spectroscopic analysis of lipid composition in normal rats using a stationary fiber-optic probe. Using a stationary near-IR fiber optic probe, spectra were obtained from tissue in the interscapular region of anesthetized adult male Sprague Dawley rats. In initial experiments, dissected interscapular white and brown adipose tissue were scanned with the fiber optic probe after removal of tissue from the animal. Both white and brown adipose tissue show absorbance in three regions (1670 - 1720 nm, 1720 - 1800 nm, 1800 - 1850 nm) characteristic of lipidFigure 1. The ratio of the absorbances in these three regions differed between white and brown adipose tissue. Specifically, the near-IR spectrum of brown adipose tissue showed a broad band at 1820 nm that corresponded to saturated fatty acids (4). This band (1820 nm) was greatly diminished in dissected white adipose tissue.

To determine the tissue region analyzed by the stationary fiber optic probe, subsequent experiments examined the interscapular region of shaved rats through the skin. Scanning was then performed after removal of skin, followed by removal of subcutaneous white adipose tissue, until scanning of the exposed interscapular brown adipose tissue was performed. Skin spectra differed from exposed brown and white adipose tissue. Near-IR spectra of brown adipose tissue through skin still showed absorbance at 1820 nm, indicating that light from the near-IR fiber optic probe penetrated from white superficial adipose tissue to interscapular brown adipose tissue.

Near-infrared spectroscopic analysis of lipid composition in Zucker obese rats. Rat pups from 4 litters of female obese (fa/fa) rats mated with male lean (Fa/?) rats were scanned with the stationary fiber-optic near-IR probe at 17 - 19 days of age. For these studies, unanesthetized rats were placed in a small box to eliminate movement, and scanned over the dorsal interscapular region. Figure 2 illustrates the complete near-IR spectra of individual 17 day old rats from two litters (N = 8) as the standard deviation from the mean of all of the rats spectra combined. A positive standard deviation represents greater absorbance at an individual wavelength, indicative of an increase in concentration(s) of constituent(s). Validation of the genotype of rat pups was made by measurement of the weight of right inguinal fat pads normalized to body weight Figure 3, N = 2 litters; body weight: lean, 31.9 + 1.3 g, N = 31; obese, 32.8 + 0.9 g, N = 14). The major differences in near-IR spectra of 17 day old preobese and lean rats were (1) the lipid region (1700, 1740 nm), (2) bound/unbound water region (1940, 2050 nm), (3) carbohydrate-bound water region (1600 nm). (The peak at 1400 nm is Wood's anomaly, which is an artifact produced by the physics of diffraction gratings.) These spectral differences correlate to the following changes in the interscapular zone of obese rats (1) a decrease in lipid (1700, 1740 nm) concentration in the interscapular zone, (2) a decrease in the spectral region representing bound/unbound water (1940, 2050 nm) and (3) an increase in the spectral region for carbohydrate-bound water (1600 nm). These differences in near-IR spectra were used for phenotypic prediction of genotype in 17 day old preobese and lean rats (Fig. 3, N = 2 litters). Principal component (PC) analysis of 701-wavelength spectra produced a single point for each spectrum for each rat, with the spectra from obese and lean rats clustering in different regions of hyperspace.

In a separate group of rats, near-IR spectral analysis was performed with the stationary fiber-optic probe placed over the interscapular region of unanesthetized three day old rats from two litters. Validation of genotype was made by allowing rat pups to grow to adulthood (16 weeks of age; obese body weight = 607 + 5.4 g; lean body weight = 399 + 6.6 g; N = 6/group). PC analysis demonstrated that distances between near-IR spectra from lean and obese rats were smaller than at 17 days of age, but existed sufficiently to allow for prediction of genotype Figure 4. From the PC plot the centers of the spectra from obese and lean rats was calculated. The vector connecting these two points illustrates the spectrum of the differences between obese and lean rats (Fig. 4). Subtraction of lean from obese absorbances gave negative numbers, and demonstrated an increase in the lipid peak absorbance in obese rats compared to lean. At 16 weeks of age, these rats were then imaged with a tunable filter wheel and near-IR video camera. Video images were obtained at lipid and protein wavelengths. A representative scan from an obese and lean rat is illustrated in Figure 5. The inset box denotes the interscapular region previously analyzed with the stationary fiber-optic near-IR probe. Absorbances from pixels in the inset box were used to compare the lipid region, demonstrating an increase in lipid absorbance in the obese rat. Moreover, the ratio of the lipid absorbance image to the protein absorbance image was greater in obese rats (0.700 + 0.008, lean; 0.792 + 0.010, obese; P < 0.01).

Near-IR spectroscopic analysis of agonist-induced lipid metabolism using a stationary fiber-optic probe. Initial experiments examined the near-IR spectrum of interscapular adipose tissue over time (50 min) in adult rats injected s.c. with normal saline (0.2 ml) in the mid-dorsal region. The full spectral region from 1100 - 2500 nm was correlated to time using PC analysis to identify spectral regions showing a linear time-dependent change. Principal components two and four (accounting for only 16.7% of total spectral variations) showed a significant change with time. These two components were used with inverse principal axis transformation to calculate spectra of each rat at each time point. Spectra calculated in this manner contained only the spectral variations correlated with time, and contained minimal random variations from movement of the rat, detector noise, etc. In saline-treated rats, the correlation (r) between the full spectrum calculated for the two components that correlated to time was 0.82 (SE of estimate = 6.4 min). Examination of the loadings (elements of the transformation matrix) of PC2 (12% of total spectral variation) revealed that PC2 arose from variations in water; PC4 (4.7% of total spectral variation) showed a slight variation in the lipid region but a larger variation in the protein-amide region. Overall, the spectral absorbances of the saline-treated rats oscillated and showed no monotonic relationship with time.

Norepinephrine (NE) was used as a prototype agent to alter lipid profiles in rat adipose tissue and to determine the effect on near-IR spectra. Administration of NE (400 µg/kg, s.c.) to adult rats resulted in a 10 mm Hg (10%) increase in mean arterial pressure. PC analysis of the near-IR spectra obtained from rats administered NE revealed five PCs with significant correlation to time. Using these five components, spectra obtained from each rat were subjected to inverse principal axis transformation. To compare the shapes of the peaks in the lipid region, multiplicative scatter correction was applied to the spectra. The lipid region of the calculated spectra from a typical rat is illustrated in Figure 6. The dominant feature observed in all rats and depicted in Figure 6 is a constant decrease in the size of lipid peaks (1700 and 1740 nm). The shape of the two peaks in the lipid region did not change with time.

Administration of AII (200 µg/kg, s.c.) to adult rats resulted in a 87 mm Hg (98%) increase in mean arterial pressure. PC analysis of the near-IR spectra obtained from adult rats administered AII revealed two PCs with signficant correlation to time. Inverse principal axis transformation of these two components revealed a decrease in lipid peak height (1700 and 1738 nm) and a shift in the observed peak areas to longer wavelengths (1738 nm) (Figure 7). The multiplicative scatter corrected peaks showed a decrease in the peak observed at 1700 nm with time (5 - 25 min elapsed) after administration of AII. After 25 min, the 1700 nm peak height returned towards control (pre-AII) values; in addition, the increase in blood pressure was no longer evident 30 min post-AII injection (data not shown).

Continuous intravenous infusion of increasing doses of AII on lipid spectral profiles in the interscapular region was performed. When given intravenously, AII resulted in a dose-dependent decrease in lipid absorbance (1700 nm) at an 1000-fold lower dose than NE (Figure 8). However, both agents resulted in dose-dependent increases in mean arterial pressure (Fig. 8). In addition to lipid spectroscopic profiles, the water absorbance peak (1940 nm) increased in a dose-dependent manner following administration of AII and NE (Fig. 8).

Thermal Infrared Imaging of AII-induced Alterations in Heat Emission. Thermal IR imaging of conscious, unrestrained rats was used to examine the effect of a single, acute s.c. injection of AII (200 µg/kg, s.c.; N = 3), isoproterenol (ISO, 50 µg/kg, s.c.; N = 1) or saline vehicle (N = 3) on regional heat emission. The camera was calibrated for a surface temperature tange of 35 - 40 C. Because the rats were shaved to increase the spatial resolution of the camera over organs, the interscapular temperature (> 40^oC) was offscale. Administration of ISO resulted in an increase in tail skin temperature that began at 9 min, reached 40^oC at 15 min, and remained elevated for the rest of the experiment (data not shown). Increases in dorsal body temperature occurred simultaneously with those in tail skin temperature in ISO-treated rats (Figure 9). Figure 10 illustrates the spatial resolution for black body thermal IR heat emission in an AII-treated rat. AII administration resulted in an increase in the tail surface temperature. In contrast to ISO, AII administration resulted in a transient drop (1^oC) in dorsal (posterior to ISBAT) surface temperature which occurred subsequent (3 - 6 min) to an increase in tail surface temperature (Figure 10). Changes in surface temperature in the tail and dorsal region were specific to AII, in contrast to a gradual nonspecific increase in head temperature in the AII, saline and ISO groups (data not shown).

Following chronic AII or saline infusion for 10 days, thermal IR black body heat emission was performed using a tunable range video camera in conscious, unrestrained rats Figure 11. Images were obtained for 10 - 15 seconds each (25 - 50 C temperature range) in each rat. In contrast to results from acutely AII-injected rats, chronic AII-infusion resulted in an increase in thorax and abdomen surface temperatures compared to saline-infused controls. However, tail surface temperatures were decreased in AII-infused rats.

In order for near-IR and IR imaging to become a viable noninvasive method for determination of subcutaneous lipid composition and surface energy expenditure, respectively, quantitative biochemical information must be obtained from spectra. In this study, near-IR spectrometry was used successfully to qualitatively and quantitatively determine the lipid composition of adipose tissue in the interscapular region of anesthetized rats, and to phenotypically predict the genotype of preobese Zucker rats. Moreover, near-IR imaging of adult obese rats with a tunable range video camera demonstrated quantitative increases in interscapular lipid composition. Using the tunable range video camera in the infrared spectral region, administration of agents known to increase energy expenditure resulted in an increase in surface heat emission from freely moving, conscious rats. Collectively, these results suggest this single-instrument noninvasive methodology can provide quantitative information on subcutaneous lipid composition and surface temperature as indexes of peripheral energy expenditure.

Methods for quantifying energy expenditure and rates of substrate oxidation have been available since the early part of this century (17, 18). Energy expenditure can be defined on the basis of the first law of thermodynamics (19). The first law of thermodynamics can be written in terms of process rates, _i, process enthalpies i, the rate of production of heat Q and the rate of performance of external work W by the body:

_i i = -(Q + W)

The summation encompasses all processes in the body. Using direct calorimetry, measurements of the amount of heat generated and dissipated by the body within an insulated environment can be made as an index of energy expenditure. Direct calorimeters detect heat production by either measuring heat transferred to a water jacket surrounding the chamber housing the subject or by changes in thermal gradients of specially designed suits worn by the subject (18,19). A disadvantage of direct calorimetry is the apparatus is cumbersome and requires that the subject remain in a physically confined environment for long periods. Moreover, direct calorimeters measure conductive, convective, and radiant heat loss but do not measure energy expended during the performance of external work (12). Thus, the experimental subject has to remain at rest during the measure. Most importantly, direct calorimetry does not provide information on the nature of the substrates oxidized to generate the measured heat (10). Results from the present study demonstrate that infrared imaging of freely moving rats allowed for very quick (3 - 5 seconds) measurement of surface temperature of different body areas. Spontaneous infrared photon emission was used to establish the temperature of the rat, so heat emission could be calculated from rat temperature measurements. However, in the present study, exact heat emission measurements were not calculated because the experimental subjects were not in a closed system, and heat loss to the environment was not measured. Regardless, temperature measurements obtained in the present study provided useful information since heat loss to the environment was similar for rats housed under the controlled conditions of the animal quarters environment.

Theoretically, under controlled environmental conditions during infrared imaging, energy expenditure as reflected by heat emission can be calculated according to the following equation: Q=m_Ti^Tf c dT, where Q is the heat energy transferred in Joules, c is the specific heat of the rat (nomally ~3.7 x 10³ J/kg C), m is the mass of the rat, T is the temperature, taking into the account both the initial (Ti) and final (Tf) temperature. Thus, within a closed system, infrared imaging is capable of providing measurements of heat emission in a similar manner as direct calorimetry. However, in contrast to direct calorimetry, in the present study use of the near-IR video camera over a tunable range allowed not only for regional-specific measurements of heat emission, but also provided information on the nature of the substrates that were oxidized to generate the measured absorbances of the external light source. Moreover, measurements were made very quickly, with the camera collecting 51.44 frames/second and a megabyte of spectral information collected every 6.6 seconds.

Indirect calorimetry is the most commonly used methodology for examining energy expenditure in experimental animals and in clinical studies (10). Measurement of the consumption of oxygen and expiration of carbon dioxide can be performed in an open or a closed system, each having its respective advantages and disadvantages. In the closed-circuit system for indirect calorimetry, the experimental subject is located in an air-proof chamber, where carbon-dioxide and water produced by respiration are trapped, and only oxygen consumption modifies the composition of the inside air (11). Alterations in the subject respiratory gas exchange result in changes in the pressure of the chamber, detected with various analytical methods. In the open system for indirect calorimetry, both ends of the respiration chamber are open to the environment, air is pulled through the chamber at a constant flow rate, and the respiratory exchange of the experimental subject results in a reduction in oxygen content and an increase in carbon dioxide content of the air (10). Heat production is determined by using a ratio of carbon dioxide expired to oxygen inhaled, the respiratory quotient. Respiratory quotients have been demonstrated to vary according to the substrate oxidized (20, 21); thus, in contrast to direct calorimetry, indirect calorimetry provides information on the type of substrate oxidized in the energy expenditure measurement. However, without the difficult surgical methodologies of isolating individual organs for assessment of organ-specific oxygen consumption, no information is provided on regional-specific substrate oxidation using indirect calorimetry. Results from the present study demonstrate that, in contrast to indirect calorimetry, the combined methodologies of near-IR and IR imaging provide quantitative measurement of regional-specific substrate utilization and heat emission. Thus, using a single-instrument video camera over a tunable range, measurements analogous to direct (heat emission from an internal light source) and indirect (substrate utilization probed by an external light source) calorimetry were made.

Infrared imaging was used in this study to determine regional-specific increases in surface temperature following drug administration. Previous investigators have examined the ability of infrared thermography to detect changes in mean body surface temperature in postsurgical patients receiving total parental nutrition, or in healthy subjects in the fasting state or following meal ingestion (22). Measurements for infrared thermography in previous studies were comparable to those obtained using indirect calorimetry and to the literature values for direct calorimetry measurements. In the present study, regionally specific measurements of heat emission were performed in response to various pharmacologic agents using the tunable range video camera. Previous investigators have demonstrated that administration of the nonselective beta-adrenergic receptor agonist, isoproterenol, resulted in an increase in tail skin temperature that is secondary to an increase in oxygen consumption and heat production (23). In agreement with previous studies, infrared imaging of isoproterenol-treated rats in the present study resulted in an increase in tail and dorsal surface body temperature (23). The time frame for alterations in tail and body temperature obtained from infrared imaging in the present study was in agreement with that obtained previously using conventional methods (23). Previous investigators have demonstrated that acute administration of the vasoactive peptide, angiotensin II (AII), resulted in an increase in tail skin temperature associated with a subsequent drop in body core temperature and oxygen consumption (23). In agreement with previous investigators, in the present study administration of AII resulted in an immediate increase in tail skin temperature which preceded a transient drop in dorsal surface temperature (23). These results demonstrate that noninvasive infrared imaging predicted quantitative changes in heat emission in different surface body regions following drug administration. However, in contrast to previous studies which utilized mean tail and colonic temperature measurements made with inserted thermocouples and indirect calorimetry to assess energy expenditure (23), IR imaging allowed for noninvasive, rapid, quantitative measurement of regional-specific changes in surface temperature with a single instrument.

Interestingly, results from the present study demonstrate that chronic infusion of a low subpressor dose of AII to rats resulted in an increase in dorsal surface temperature concomitant with a decrease in tail surface temperature. These results are in direct contrast to results obtained in the present study following acute administration of single pressor dose of AII. Future studies will examine the dose-dependent effects of AII on IR measurements of surface temperature, to determine mechanisms contributing to AII-mediated alterations in energy expenditure.

The Zucker obese rat is a widely employed rodent model for the study of obesity (24).

Genomic analysis of the fa/fa obese rat demonstrated that the autosomal, recessive defect lies in the gene encoding the leptin receptor (25). Perpetuation of the Zucker obese rat colony requires either breeding of obese (fa/fa) females to lean (Fa/?) males or heterozygous breeding, resulting in heterogenous litter populations. Before the fa/fa gene was cloned, identification of the genotype of rat pups from litters by these breeding methods could not be easily made until the onset of obesity at approximately 35 days of age (26). Thus, genotype prediction was dependent on various phenotypic measures, including sizing of fat cells from inguinal fat pad biopsies (5-7 days of age) (27,28), GDP binding in brown fat mitochondria (2 - 14 days of age) (29), rate of oxygen consumption (2 - 3 days of age) (30), and measurement of the percentage body weight of the right inguinal fat pad (17 days of age) (31). Many of these measures are invasive, precluding chronic study of factors contributing to the development of obesity. Results from the present study demonstrate that noninvasive near-IR spectral analysis predicted the genotype of preobese Zucker rats as early as 3 days of age. Moreover, near-IR spectral analysis provided information on the chemical constituents altered in preobese Zucker rats compared to lean controls. Increases in the amount of subcutaneous and internal lipid are well documented in the adult Zucker obese rat (32, 33). In the present study, noninvasive near-IR imaging of adult obese rats demonstrated a regional-specific increase in subcutaneous lipid content. Moreover, differences in lipid spectral signatures (saturated vs. unsaturated fatty acids) in skin and subcutaneous tissue can be identified with near-IR imaging.

The lipid region of biological tissue in near-IR spectra is considered to center at approximately 1700 nm. Absorbance peaks in this region arise from carbon-hydrogen (C-H) stretches from carbon atoms connected in chains, the backbone of fatty acids. The energy of C-H stretches is reflected in the wavelengths of the peaks; cis hydrogens have the highest energy and the shortest wavelength (1680 - 1730 nm) (4). Trans and/or saturated hydrogens have a lower energy and longer wavelength (1760 - 1860 nm). Therefore hydrogens on saturated fatty acids appear at longer wavelengths than those on unsaturated fatty acids (4). Changes in these spectral bands correspond to alterations in lipid profiles. In the present study, near-IR spectra were taken from sequential dissection of tissues in the interscapular region. Brown and white adipose tissue exposed sequentially in the interscapular region show absorbance in three regions characteristic of lipid. However, the ratio of absorbances in these three lipid regions differed between brown and white adipose tissue. Specifically, the near-IR spectra of brown adipose tissue showed a broad band at 1820 nm that corresponded to saturated fatty acids (4). This band (1820 nm) was greatly diminished in white adipose tissue. Moreover, skin spectra differed from brown and white adipose tissue; however, scanning of the interscapular region still showed absorbance at the characteristic brown adipose tissue wavelength of 1820 nm. These results demonstrate penetration of the light beam through skin to the level of interscapular brown adipose tissue. Thus, near-IR spectra obtained in this study encompass the combined spectral profiles of skin, subcutaneous white adipose tissue and brown adipose tissue in the interscapular region.

Previous investigators have demonstrated that fatty acids derived from dietary origin have a higher unsaturation index (% unsaturation) as compared to fatty acids inherent to brown adipose tissue (higher percentage of saturated fatty acids) (34). In agreement, results from this study demonstrated that near-IR spectra from exposed interscapular brown adipose tissue contained a high percentage of saturated fatty acids (longer wavelength), and demonstrated that light from the stationary fiber-optic probe penetrated sufficiently to allow for detection of this spectral signature.

A stationary fiber-optic near-IR probe was used in the present study to examine lipid constituents in the interscapular region following either a single injection or continous infusion of NE or AII. NE was used as a prototype agent known to result in lipid mobilization from white and brown adipose tissue. The lipolytic response of brown adipose tissue to NE is well documented, providing liberated fatty acids as a fuel for thermogenesis (35). Additionally, NE results in lipolysis of white adipose tissue (36). Previous investigators have demonstrated that the maximum rate of lipolysis in response to NE was similar for white and brown adipose tissue. However, there are conflicting reports on the potency of NE-induced lipolysis in white versus brown adipose tissue (36, 37). Results from the present study demonstrate that administration of NE resulted in a decrease in saturated and unsaturated fatty acid lipid spectral profiles. Moreover, the effect of NE on lipid spectral profiles was dose-dependent. Concomitant with decreases in lipid absorbance, administration of NE resulted in an increase in water content, indicative of either an increase in blood flow or extracellular fluid volume. In agreement with results from the present study, previous investigators demonstrated that systemic administration of NE resulted in an increase in adipose blood flow (38). Thus, NE-induced changes in lipid spectra as measured by the near-IR fiber-optic probe most likely represent lipolysis-derived mobilization of fatty acids, rather than decreases in lipid absorbance resulting from diminished blood flow through interscapular adipose tissue.

Interestingly, results from the present study demonstrate that a single pressor dose of AII resulted in a decrease in lipid absorbance and a shift in observed peak areas to longer wavelengths. In contrast to alterations in lipid peaks of adipose tissue observed after NE administration, in response to AII the observed peak at 1738 nm (saturated fatty acids) increased in height relative to the 1700 nm (unsaturated fatty acid) peak. Therefore, results from the present study suggest that administration of AII resulted in a decrease in unsaturated (1700 nm) fatty acids primarily from white adipose tissue.

In conclusion, results from the present study demonstrate that near-IR and IR imaging of freely moving, conscious rats allows for the noninvasive assessment of lipid composition and surface body temperature. Moreover, quantitative measurements are obtained quickly, providing information on specific substrates utilized coupled to measurement of regional energy expenditure. Collectively, these results suggest that near-IR and IR imaging will serve as a valuable tool for the evaluation of novel pharmacologic agents for the treatment of obesity.

Figure legends

Figure 1. Difference in near-infrared spectra from dissected interscapular brown and white adipose tissue. Interscapular brown and white adipose tissue were dissected from adult rats, and near-IR spectrometry was performed on isolated, dissected tissues. Multiplicative scatter correction was applied to the spectrum of each adipose tissue before subtraction of white adipose spectra from brown adipose spectra. Both white and brown adipose tissue show absorbance in three regions (1670 - 1720 nm, 1720 - 1800 nm, 1800 - 1850 nm) characteristic of lipid. The ratio of the absorbances differed between white and brown adipose tissue, with brown adipose tissue showing a broad band at 1820 nm that corresponds to saturated fatty acids. This band (1820 nm) was greatly diminished in dissected white adipose tissue.

Figure 2. Near-infrared spectra from 17 day old Zucker obese and lean rats. Rat pups from 2 litters of female obese (fa/fa) rats mated with lean males (Fa/?) were placed in a small, open top box at 17 - 19 days of age, and scanned with a stationary near-IR fiber optic probe placed over the interscapular region. Absorbance differences in standard deviations from the mean of all rats spectra combined for individual obese (----) and lean (--) rats (N = 7 obese, N = 6 lean) are illustrated across the complete near-IR spectral range.

Figure 3. Phenotypic prediction of genotype of 17 - 19 day old obese (0) and lean (+) rats by measurement of right inguinal fat pad weight (top) and by near-IR principal component analysis (bottom). For phenotypic prediction of genotype, 17 - 19 day old rats were scanned using the stationary near-IR fiber-optic probe. Principal component analysis of 701-wavelength near-IR spectra produced a single point for each spectrum from each rat (bottom). The spectra from obese and lean rats cluster in different regions of spectral hyperspace. Validation of the phenotypic prediction of genotype was made by measurement of the weight of the right inguinal fat pad (y-axis) plotted against the body weight (x-axis) (top) of rats.

Figure 4. Near-IR spectra (top) and principal component analysis (bottom) of obese (0) and lean (+) rat pups at 3 days of age. Rat pups from one litter were scanned with a stationary near-IR fiber-optic probe placed over the interscapular region. The spectrum of the differences between obese and lean rats (top) was obtained by subtraction of lean from obese absorbances at each wavelength. Positive deflections, representing an increase in the absorbance, were observed in the lipid spectral region (1700 - 1840). Principal component analysis (bottom) of the 701-wavelength spectra produced a single point representing the spectrum from each rat in PC space. The obese and lean rats clustered in different regions of hyperspace; however, differences in near-IR spectra were not as significant as those observed at 17 days of age.

Figure 5. A representative near-IR scan from an adult lean (a) and obese (b) rat. Rats from one litter, used for phenotypic prediction of genotype at 3 days of age by near-IR spectral scanning, were imaged with a tunable near-IR video camera at 16 weeks of age at protein and lipid wavelengths. The lipid wavelength is shown, with the intensity of lipid absorbance denoted by an increase in the lightness of the image. The inset box denotes the interscapular region, which demonstrates an increase in lipid absorbance from the adult obese rat compared to the lean control.

Figure 6. Near-IR spectra from the interscapular region of an adult norepinephrine-treated rat. Male Sprague-Dawley rats were anesthetized, shaved over the interscapular region, and near-IR spectra obtained with a stationary fiber-optic probe placed over the interscapular region. Three scans were taken of each rat for pre-drug control comparisons. Rats were injected (s.c.) with NE (400 µg/kg), and near-IR spectra obtained every 5 min for 50 min (times of scanning indicated on graph). Inverse principal axis transformation of spectra with multiplicative scatter correction demonstrates a time-dependent decrease in lipid absorbance at 1700 and 1740 nm following NE administration.

Figure 7. Near-IR spectra from the interscapular region of an adult angiotensin-treated rat. Male, Sprague-Dawley rats were anesthetized, shaved over the interscapular region, and near-IR spectra obtained with a stationary fiber-optic probe placed over the interscapular region. Three scans were taken of each rat for pre-drug control comparisons. Rats were injected (s.c.) with AII (200 µg/kg), and near-IR spectra obtained every 5 min for 50 min (times of scanning indicated on graph). Inverse principal axis transformation of spectra with multiplicative scatter correction demonstrates a decrease in lipid peak height over a shorter time period (35 min) following AII administration. Moreover, a shift in lipid peak height to longer wavelengths occurred following AII administration.

Figure 8. Dose-response curves for the effect of continous i.v. infusions of AII (left) or NE (right) on mean arterial blood pressure (top), near-IR absorbance at 1700 nm (middle), and near-IR absorbance at 1940 nm (bottom). Male, Sprague-Dawley adult rats were anesthetized, and femoral artery and vein catheters implanted for measurement of mean arterial pressure and drug administration, respectively. Near-IR spectra were obtained every 4 min over 20 min with each concentration of drug infused for 20 min. Mean arterial pressure (top) increased to a similar maximal extent following AII (ng/min) and NE (µg/min) infusion. In addition, near-IR absorbance in the lipid region at 1700 nm (middle) decreased to a similar extent following AII or NE administration. Near-IR absorbance in the water region at 1940 nm (bottom) increased following AII and NE administration.

Figure 9. Spatially resolved thermal IR emission in an isoproterenol-treated rat. Male, Sprague-Dawley adult rats were shaved over the dorsal surface before ISO (50 µg/kg, s.c.) or saline administration. Rats were imaged with the video camera every 3 min for 20 min. Thermal IR imaging of a saline-treated (left) and an ISO-treated (right) rat is illustrated at 9 min post-injection. The tail and body of the ISO-treated rat exhibited an increase in surface temperature compared to the control rat (tail not visible due to its low temperature).

Figure 10. Regional temperature measurements by infrared imaging following acute AII administration. Male, Sprague-Dawley adult rats were shaved over the dorsal surface before AII (200 µg/kg, s.c.) or saline administration. Rats were imaged with the video camera every 3 min for 20 min. Temperature measurements over time are illustrated for individual rats in the tail (top) region and in the dorsal (bottom) body region. Administration of AII resulted in an increase in tail temperature (at 3 min) that occurred prior to a transient decrease (at 5 min) in body temperature. Spatially resolved temperature in an AII-treated rat 9 min post-injection is illustrated. The white appearance indicates a higher temperature; in the saline-injected control rat, the rat's tail was not visible (data not shown).

Figure 11. Regional temperature measurements by infrared imaging following chronic AII administration. Male, Sprague-Dawley adult rats were implanted with osmotic minipumps containing either saline or AII (100 ng/kg/min) for 10 days. Thermal IR imaging was performed with the tunable range video camera in conscious, unrestrained rats. A representative saline control (top, left) and AII-infused (top, right) rat is illustrated. Differences between saline and AII-infused rats include loss of visibility of the tail and an increase in interscapular visibility in AII-infused rats compared to saline control. Temperature data (bottom) was Z-scored (body temperature - mean body temperature for individual body part/standard deviation) to prevent tail variations in temperature from masking temperature differences in other organs. Tail temperature was decreased in AII-infused rats; thorax and abdomen temperature were increased in AII-infused rats.

References

1. Wetzel D: Near-infrared reflectance analysis: sleeper among spectroscopic techniques. Anal. Chem. 55:1165A-1176A, 1983.

2. Nagao A, Uozumi, J, Iwamoto M, Yamazaki M: Determination of fat content in meats by near-infrared reflectance spectroscopy. Chem. Abstr 102:219697r, 1985.

3. Conway J, Norris K, Bodwell C: A new approach for the estimation of body composition: infrared interactance. Amer. J. Clin. Nutr. 40:1123, 1984.

4. Carney JM, Landrum W, Mayes L, Zou Y, Lodder RA: Near-IR spectrophotometric monitoring of stroke-related changes in the protein and lipid composition of whole gerbil brains. Anal. Chem. 65:1305-1313, 1993.

5. Cassis LA, Lodder RA: Near-IR imaging of atheromas in living arterial tissue. Anal. Chem. 65:1247-1256, 1993.

6. Dempsey RJ, Davis DG, Buice RG, Lodder RA: Biological and medical applications of near-infrared spectrometry. Appl. Spectrosc. 50:18A-34A, 1996.

7. Dempsey RJ, Cassis LA, Davis DG, Lodder RA: Near infrared imaging and spectroscopy in stroke research: lipoprotein distributions and disease. Ann. New. York Acad. Sci., in press, 1997.

8. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM: Positional cloning of the mouse obese gene and its human homologue. Nature 372:425-432, 1994.

9. Himms-Hagen J: Role of thermogenesis in the regulation of energy balance in relation to obesity. Can. J. Physiol. Pharmacol. 67:394-401, 1989.

10. Simonson DC, DeFronzo RA: Indirect calorimetry: methodological and interpretative problems. Am. J. Physiol. 258 (Endocrinol. Metab.21):E399-E412, 1990.

11. Even PC, Mokhtarian A, Pele A: Practical aspects of indirect calorimetry in laboratory animals. Neurosci. Biobehav. Rev. 18(3):435-447, 1994.

12. Schutz Y: The basis of direct and indirect calorimetry and their potentials. Diabetes/Metabolism Reviews, Vol. 11, No. 4, 838-403, 1995.

13. Lindmark L, Rudholm L, Lundholm K: A method for measuring energy metabolism and whole body oxidation in small animals with special reference to experimental metabolism and nutrition. Scand. J. Clin. Lab. Invest. 46:273-281, 1986.

14. VandeVan M, French T, Fishkin J, Gratton F: Near-infrared imaging spectroscopy of mammalian tissue in the frequency domain. Biophys. J. 59:167a, 1991.

15. Lavau M, Bazin R: Inguinal fat pad weight plotted versus body weight as a method of genotype identification in 16-day-old Zucker rats. J. Lipid Res. 23:941-943, 1982.

16. Lodder RA, Heiftje G: Quantile BEAST attacks the false-sample problem in near-infrared reflectance analysis. Appl. Spectroscop. 42(8):1351-1365, 1988.

17. Atwater WO, Rosa EB: A new respiration calorimeter. Wasington, DC: US Dept. Agriculture, 1899.

18. Benedict FG, Carpenter TM: Respiration calorimeters for studying respiratory exchange and energy transformation of man. Washington, DC: Carnegie Inst. Washington, 1910.

19. Garby L: Measurement of whole-body energy expenditure. Int. J. Obes. Relat. Metab. Disord. 17:S7-S9, 1993.

20. Lusk G: The elements of the science of nutrition. Philadelphia, PA: Suanders, 1928.

21. Hirsch J: Fatty acid patterns in human adipose tissue. In: Handbook of Physiology. Adipose tissue. Washington, DC: Am. Physiol. Soc., 1965, sect. 5, chapt. 17, p. 181-189.

22. Shuran M, Nelson R: Quantitation of energy expenditure by infrared thermography. Am. J. Clin. Nutr. 53:1361-1367, 1991.

23. Wilson K, Fregly M: Angiotensin II-induced hypothermia in rats. J. Appl. Physiol. 58(2):534-543, 1985.

24. Zucker L, Zucker F: Fatty, a new mutation in the rat. J. Heredity 52:275-278, 1961.

25. Chua SC, Chung WK, Wu-Peng S, Zhang Y, Liu S, Tartaglia L, Leibel R: Phenotypes of mouse diabetes and rat fatty due to mutations in the OB (Leptin) receptor. Science 271:994-996, 1996.

26. Bell G, Stern J: Evaluation of body composition of young obese and lean Zucker rats. Growth 41:63-80, 1977.

27. Johnson P, Zucker L, Cruce J, Hirsch J: Cellularity of adipose depots in the genetically obese Zucker rat. J. Lipid Res. 12:706-714, 1971.

28. Given R, Hietanen E, Greenwood M: Increase adipose tissue lipoprotein lipase activity during development of the genetically obese rat (fafa). Metab. 27:1955-1965, 1978.

29. Bazin R, Eteve D, Lavau M: Evidence for decreased GDP binding to brown adipose tissue mitochondria of obese Zucker (fa/fa) rats in the very first days of life. Biochem. J. 221:241-245, 1984.

30. Planche E, Joliff M, Bazin R: Energy expenditure and adipose tissue development in 2- to 8- day old Zucker rats. Inter. J. of Obesity 12:353-360, 1987.

31. Boulange A, Planche E, Gasquet P: Onset of genetic obesity in the absence of hyperphagia during the first week of life in the Zucker rat (fa/fa). J. Lipid Res. 20:857-864, 1979.

32. Hausman G, Kasser TR, Shapira JF, Martin RJ: Techniques for identification of the young obese zucker rat and some observations on brown adipose tissue morphology and histochemistry in the young obese zucker rat. Inter. J. of Obesity 7:487-492, 1983.

33. Meiergrankenfeld B, Abelenda M, Jauker H, Klingenspor M, Kershaw EE, Streamson CC, Leibel RL, Schmidt I: Perinatal energy stores and excessive fat deposition in genetically obese (fa/fa) rats. Am. J. Physiol. 270:E700-E708, 1996.

34. Carneheim C, Cannon B, Nedergaard J. Rare fatty acids in brown fat are substrates for thermogenesis during arousal from hibernation. Amer. J. Physiol. 256:R146-R154, 1989.

35. Bukowiecki L. Lipid metabolim in brown adipose tissue. In Trayhurn, P., Nicholls D (eds): Brown adipose tissue, p105-121, 1986.

36. Heerden J, Oelofsen W: A comparison of norepinephrine and ACTH-stimulated lipolysis in white and brown adipocytes of female rats. Comp. Biochem. Physiol. 93(2):275-279, 1989.

37. Saggersen E: Sensitivity of adipocyte lipolysis to stimulatory and inhibitory agents in hypothyroidism and starvation. Biochem. J. 238:387-394, 1984.

38. Rosell S, Belfrage E: Blood circulation in adipose tissue. Physiol. Rev. 59:1078-1104, 1979.

Return to Wave of the Future Table of Contents

Return to Cover of Wave to the Future