Gene expression in hypothalamus and brown adipose tissue of mice divergently selected for heat loss

MARK F. ALLAN, MERLYN K. NIELSEN and DANIEL POMP

Department of Animal Science, University of Nebraska, Lincoln, Nebraska 68583-0908


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Allan, Mark F., Merlyn K. Nielsen, and Daniel Pomp. Gene expression in hypothalamus and brown adipose tissue of mice divergently selected for heat loss. Physiol Genomics 3: 149–156, 2000.—Gene expression was evaluated in mice divergently selected for 16 generations for heat loss, measured by direct calorimetry. The high (MH) heat loss line has ~50% greater heat loss, ~35% less body fat, ~20% greater feed intake, and twofold greater activity levels than the low (ML) heat loss line. At 11 wk, inbred males (developed from MH and ML) were euthanized 3 h after dark for dissection of tissues and extraction of RNA. Differential display PCR (DD-PCR) was used to evaluate transcriptional differences between lines in hypothalamus and brown adipose tissue (BAT). Evaluation was replicated within and across lines, using family pools of mRNA. Two genes were confirmed by competitive RT-PCR and/or Northern analysis to have greater levels of mRNA present in ML relative to MH mice. In both hypothalamus and BAT, the ribosomal protein L3 (RPL3) gene was expressed at higher levels in ML, whereas an unknown expressed sequence tag (EST) was also found at higher levels in the hypothalamus of ML mice. These results implicate RPL3 in regulation of energy balance and extend the genetic dissection of response to selection to the transcriptional level.

energy balance; differential display; obesity


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
ALTHOUGH MUTATIONS AT SINGLE genes can cause obesity in humans (14, 5), the number of individuals afflicted is few. The majority of genetic variation for body fatness is determined by complex combinations and interactions of multiple genes with effects on various components of energy intake and energy expenditure (20, 21, 22). A unique animal model to study the polygenic regulation of energy balance was established by Nielsen et al. (18). They divergently selected high (MH) and low (ML) lines of mice on the basis of heat loss (kcal·kg-0.75·day-1) for 16 generations, with selection criteria measured in 9 to 11-wk-old male mice by direct calorimetry. Selection resulted in a 53% difference in heat loss between the MH and ML lines. As correlated responses to selection, MH mice were much leaner despite increased feed intake relative to ML mice (15) and were also more active, with an elevated core body temperature (17). Despite changes in heat loss, feed intake, and body composition, the two lines did not differ in body weight (15, 19).

Chromosomal regions harboring genes influencing energy balance have been identified in mice (16, 21, 25, 26, 28). These quantitative trait loci (QTL) scans will continue to play a role in the genetic dissection of energy balance. However, given the difficulty in cloning QTL, the identity and direct physiological relevance of nearly all of these loci have yet to be determined.

To better understand the polygenic nature of energy balance, we employed a gene expression approach. Traditionally, gene expression studies have compared differences in amounts of mRNA between two or more samples for known candidate genes. This is quite limiting, given that the vast majority of loci have yet to be discovered and/or characterized. In contrast, analysis using gene microarrays promises extremely high throughput and comparison of expression "in parallel" (6). However, the technology is not yet feasible for widespread adoption, and the number of expressed genes represented on current arrays is relatively small. In the present study, the technique of differential display PCR (DD-PCR) was employed. Differential display PCR enables a genome-wide scan for genes that are differentially expressed in a treatment (or genotype) compared with another treatment (or genotype) (10, 30).

The objectives of this project were to advance our understanding of the polygenic regulation of obesity and maintenance energy requirements in mammals and to extend the dissection of the nature of selection response for complex characters to the transcriptional level. Specifically, we have used DD-PCR to identify specific mRNAs that differ in concentration in lines of mice that have undergone 16 generations of divergent (high and low) genetic selection for heat loss.


    MATERIAL AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

Resource population.
Mice used in this study originated from high (MH) and low (ML) lines divergently selected for heat loss (kcal·kg-0.75·day-1) for 16 generations. The selection criterion was measured in 9- to 11-wk old male mice by direct calorimetry. Selection lines originated from a four-way (outbred lines) composite population (7). Selection applied for 16 generations using three replications of selection lines resulted in heat loss means (kcal·kg-0.75·day-1) of 179.1 for MH, 107.5 for ML, and 134.9 for the randomly selected control line. Realized heritability was estimated at 0.28 based on divergence of MH and ML. Detailed results of MH and ML selection lines were presented by Nielsen et al. (18, 19), Moody et al. (15), and Mousel and Nielsen (17). Following the relaxation of selection, lines have been maintained with a minimum of inbreeding and are currently at generation 32.

At generation 21, full-sibling matings were initiated in 10 sublines per selection line (replicate 1 only) to develop inbred populations. Inbred high (IH from MH) and low (IL from ML) matings were carried out for four generations (cumulative inbreeding estimated at F = 0.60) prior to initiation of this study.

Mouse care and maintenance.
Mice were weaned at 3 wk, caged by sex with wood chip bedding, and provided ad libitum access to water and feed (Teklad 8604 rodent chow). Temperature was maintained at 22°C, with relative humidity at 35–50%, and a 12:12-h light-dark cycle starting at 0700. All procedures and protocols were approved by the University of Nebraska Institutional Animal Care and Use Committee.

Collection of phenotypic data.
Heat loss was measured (18) on male mice at ~11 wk of age using 10 direct, gradient-layer, individual-animal calorimeters (model 0601-s; Thermonetic, San Diego, CA). Briefly, male mice were weighed and placed in calorimeters with a 2-g feed pellet at ~1700. Following a 30-min acclimation period, heat loss measurements were recorded every minute for a 15-h period. Data were adjusted for metabolic body weight and projected to a 1-day basis (kcal·kg-0.75·day-1). Mice not consuming the 2-g of feed were omitted from the data set, to eliminate variation in heat loss caused by variation in feed intake during measurement. Least-squares means for heat loss were 157.6 and 98.9 kcal·kg-0.75·day-1 for IH and IL, respectively (P < 0.0001).

Tissue collection.
Three to five hours into the dark period, 13-wk old male mice were decapitated after brief exposure to CO2. Time of tissue collection was selected to represent the time period during which maximal divergence in heat loss and activity between the high and low selection lines occurs. Hypothalamus, pituitary, brown adipose (BAT), adrenal gland, liver, heart, spleen, and white adipose (subcutaneous and epididymal fat pads) tissues were dissected and snap-frozen in liquid nitrogen within 5 min of decapitation. Removal of the hypothalamus was conducted from the ventral side of the brain. The optic chiasm was dissected away from the anterior portion of the hypothalamus, followed by dissection of the mammillary nuclei from the posterior of the hypothalamus. The entire hypothalamus was removed including the arcuate, ventromedial, dorsomedial, and paraventricular nuclei.

Total RNA preparation.
Total RNA was isolated from individual tissues using TRIZOL LS reagent (GIBCO, Life Technologies). For hypothalamus, RNA was extracted in the presence of 30 µg of glycogen (Boehringer Mannheim, Indianapolis, IN) and with the help of a glass bead suspension matrix (Rnaid Matrix; Bio 101, La Jolla, CA). RNA samples were resuspended in DEPC-treated H2O and stored at -80°C until needed.

Prior to analysis, RNA samples were treated with DNase I (GIBCO) in the presence of RNasin (Promega, Madison, WI) to remove any DNA contamination. DNase I was removed from the reaction by standard phenol/chloroform extraction procedures. RNA was quantified with a fluorometer (TD-700; Turner Designs, Sunnyvale, CA), using fluorescein excitation and emission wavelengths.

cDNA synthesis.
To decrease the likelihood of false positives in DD-PCR, total RNA was pooled from two full-brothers within selected families. Two separate pools were evaluated within lines, yielding a total of four pools (2 IH and 2 IL), each consisting of two mice. The selection of high (IH) and low (IL) families was based on the extremes in calorimeter measurements. Family averages were ranked with the two highest and two lowest families selected. Family averages for heat loss were 177.0 and 175.9 for IH and 84.5 and 95.1 for IL (kcal·kg-0.75·day-1).

Ten sets of cDNA were synthesized using 10 different anchor primers with the following sequence (5'-3'): T7 promoter (dT12) NN (where NN can be GA, GC, GG, GT, CA, CC, CG, AA, AC, AG). First-strand cDNA synthesis consisted of 0.2 µg RNA mixed with 1 µg of an anchor primer and brought to 10 µl with DEPC-treated H2O. The 10 µl mixture was incubated at 65°C for 5 min, and immediately placed on ice. A 10-µl mixture containing 1x Superscript buffer, 10 mM DTT, 40 U SuperScript reverse transcriptase II (GIBCO), 5 U RNasin and 0.5 mM of each dNTP was combined with the 10 µl containing the RNA and primer. This mixture was incubated for 10 min at 25°C, 60 min at 42°C, followed by 70°C for 15 min, and held at 4°C. All cDNA samples were stored at -40°C.

DD-PCR.
To minimize variation within lines, DD-PCR was performed on cDNA originating from tissues from the partially inbred IH and IL lines. In addition to the 10 3' anchor primers used for cDNA synthesis, DD-PCR utilized 20 different 5' arbitrary primers with the following sequence 5'-M13 rev-48 (arbitrary 10-mer)-3', resulting in 200 different PCR reactions per pool. The two tissues used for DD-PCR were hypothalamus and BAT. Internal labeling was used for hypothalamus DD-PCR, whereas fluorescently (TMR; TAMRA) end-labeled forward primers were used in the BAT PCR.

Hypothalamus DD-PCR was carried out in a total volume of 20 µl containing 2 µl cDNA, 1.5 mM MgCl2, 0.25 mM each dNTP, 1 U Taq polymerase with 1x MgCl2-free supplied buffer (Sigma-Aldrich, St. Louis, MO), 0.2 µM appropriate 3' anchor primer, 0.2 µM 5' arbitrary primer, and 2.5 µCi [{alpha}-33P]dATP. BAT fluorescent DD-PCR conditions were the same except for the following: 10 µl total volume, with 1 µl of cDNA reaction, 3.75 mM MgCl2, 0.5 U Taq polymerase, 0.35 µM TMR-anchor 3' anchor primer, and 0.35 µM 5' arbitrary primer. Initial denaturation was at 95°C for 2 min followed by four cycles of denaturing at 92°C for 15 s, annealing at 46°C for 30 s (50°C, 30 s for BAT), and extension at 72°C for 2 min. This was followed by 25 cycles of denaturing at 92°C for 15 s, annealing at 60°C for 30 s, and an extension at 72°C for 2 min, with a final extension of 72°C for 7 min. PCR reactions were carried out in a Peltier thermal cycler (model PTC-220; MJ Research, Waltham, MA) with heated lid.

Hypothalamus DD-PCR products were mixed with loading dye (formamide and bromophenol blue), denatured at 95°C for 2 min, and placed immediately on ice. DD-PCR products were separated on a 4.5% polyacrylamide denaturing gel (Sequagel XR; National Diagnostics, Atlanta, GA) for 16 h at 800 V (100 W) and at 40°C in a GenomyxLR DNA sequencer (Genomyx, Foster City, CA). The same DD-PCR products were also separated on a 6% polyacrylamide denaturing gel 3.25 h at 2,700 V (100 W) and at 50°C. The low-voltage, long run generates separation of large cDNA fragments whereas the high-voltage, short run allows smaller fragments to be captured in the gel. When electrophoresis was completed, gels were dried and exposed to Kodak BioMax MR film (Eastman-Kodak, Rochester, NY). After 12–16 h of exposure at RT, films were developed and used as templates to identify differential expression of cDNA representing the IH and IL lines. Gels were marked, and bands with equal intensities within lines and different intensities between lines were excised, placed in a PCR tube, and stored at -40°C.

For electrophoresis of BAT DD-PCR, product plus loading dye (formamide and dextran blue) were electrophoresed at 3,000 V (100 W), at 50°C. Fluorescent images were analyzed on a GenomyxSC fluorescent imaging scanner using Adobe Photoshop 4.0 (Adobe Systems, San Jose, CA). Gels were scanned after 2.5 and 5 h of electrophoresis. Using a virtual grid overlay, bands of interest were excised from the gel and placed in a 1.5-ml microcentrifuge tube with 50 µl of sterile nuclease-free TE (10 mM Tris-Cl, pH 7.4). Samples were incubated for 30 min at 50°C and stored at -40°C.

To estimate sizes of DD-PCR products, a sequencing reaction using pBluescript KS as template was electrophoresed in all gels. For sequencing reactions, we used the Cyclist Taq DNA sequencing kit (Stratagene, La Jolla, CA) or the Thermo Sequenase fluorescent labeled primer cycle sequencing kit (Amersham Pharmacia, Buckinghamshire, UK) following the manufacturers’ protocols.

Template for reamplification.
PCR using the gel band as source of template yielded reamplified products to be used as template for preparation of probes for hybridization and for sequencing. Reamplification PCR was carried out in a total volume of 40 µl containing the excised DD-PCR band (hypothalamus) or 3 µl of incubated excised band solution (BAT), 1.5 mM MgCl2, 0.25 mM each dNTPs, 2 U Taq polymerase with 1x MgCl2-free supplied buffer (Sigma-Aldrich), with 0.2 µM 3' anchor primer and 0.2 µM 5' arbitrary primer. Thermal cycling included five cycles with a 50°C annealing temperature, followed by 25 cycles with a annealing at 60°C.

Evaluation of DD-PCR results.
Northern blot analysis was first performed using probes generated directly from reamplification PCR. A cDNA probe specific for glyceraldehyde-3-phosphate dehydrogenase (G3PDH) was used to standardize RNA loading on the blots. Whereas IH and IL lines were used for DD-PCR to control for false positives, tissues from replicate 1 MH and ML (generations 25–27) mice were used for Northern blots. The rationale was to test for gene expression differences in the original outbred selection lines. For collection of tissues and extraction of RNA, we followed methods previously stated. Separate pools of MH and ML hypothalamus RNA were created from the individual samples and quantified. Ten micrograms of pooled RNA was mixed with 5 µl of 10x MOPS, 8.7 µl formaldehyde, and 25 µl formamide, loaded into a 1% formaldehyde denaturing gel, and electrophoresed for 2–3 h at 110 V, with 1x MOPS buffer at 4°C. The gel was washed in DEPC-treated H2O for 10 min, followed by a wash in 1 liter of 50 mM NaOH for 15 min. RNA was transferred from the gel to a Hybond-N+ nylon membrane (Amersham Pharmacia) with a vacuum blotter (Bio-Rad, Hercules, CA), using 10x SCC for 1.75 h. Membranes were dried for 15–20 min and cross-linked using a UV light for 3–4 min. Membranes were stored at -20°C until needed.

Probes were prepared for hybridization using asymmetric PCR. A total PCR reaction volume of 50 µl consisted of 200 ng of dd-reamplified product or G3PDH cDNA, 1.5 mM MgCl2, 0.2 mM each d(G,T,C)TP, 0.006 mM dATP, 5 U Taq polymerase with 1x MgCl2-free supplied buffer (Sigma-Aldrich), 0.12 µM appropriate 3' anchor primer, 0.002 µM 5' arbitrary primer, and 1 µCi [{alpha}-32P]dATP Thermal cycling conditions were 40 cycles of denaturing at 95°C for 30 s, annealing at 60°C for 30 s, and extension at 72°C for 2 min, with a final extension period at 72°C for 7 min. Samples were stored at 4°C or -20°C.

Membranes were prehybridized at 42°C with a prehybridization buffer consisting of 0.025 M KPO4, 5x SCC, 5x Denhardt’s solution, 50 µg of sheared salmon sperm DNA, 50% formamide, and 16.5% DEPC-treated H2O for a minimum of 2 h. After the prehybridization period, the buffer was removed and hybridization buffer was added (prehybridization buffer plus 10% dextran sulphate). The probe was denatured at 95°C for 5 min and added immediately to the cylinder and allowed to hybridized for 16 h at 42°C. Following hybridization, the membranes were removed from the cylinders and washed twice with 1x SCC and 1% SDS and twice with 0.25x SCC, 1% SDS. Following washing, membranes were exposed to X-ray film (Eastman-Kodak, Rochester, NY).

Membranes were stripped of probes using a solution of 1 mM Tris·HCl (pH 8), 1 mM EDTA (pH 8), and 0.1x Denhardt’s solution. Membranes were probed with G3PDH after stripping to standardize RNA loading on blots. Quantification of Northern hybridization results was done using a phosphor Imager:SF (Molecular Dynamics).

Sequencing.
Differential display products were sequenced following gel purification (Qiagen, Santa Clarita, CA), using the Cyclist Taq DNA sequencing kit and/or Thermo Sequenase fluorescent labeled primer cycle sequencing kit. Sequencing reactions were electrophoresed on a GenomyxLR DNA sequencer (Genomyx). Products were also sequenced by automated dideoxy terminator cycle sequencing at the University of Nebraska DNA Sequencing Core Facility.

Competitive PCR assay.
Competitive RT-PCR was used to further confirm RPL3 mRNA expression differences for hypothalamus and BAT, and to test for differences in epididymal fat pad and skeletal muscle. This assay used a 415-bp portion of the RPL3 gene with an internal competitor template that was 375 bp in length, which was generated by creating a 40-bp deletion in the endogenous RPL3 target. To synthesize the competitor template using PCR, cDNA from the RPL3 gene was used as a template under the following conditions (40 µl total volume): 2 µl RPL3 cDNA, 2.0 mM MgCl2, 0.25 mM each dNTPs, 0.2 µM each primer (forward primer RPL3 P3, 5'-TCATTGCCCACACCCAGATGTCCAGGTGAACGGAGGCACT-3'; reverse primer RPL3 3R, 5'-GCATTGTTCTTGATCAGTTTGCC-3'), 1 U AmpliTaq Gold with 1x MgCl2-free supplied buffer (Perkin-Elmer, Branchburg, NJ), and 1x RediLoad. Thermal cycling conditions included an initial denaturing step at 95°C 5 min, followed by 40 cycles of denaturing at 95°C for 30 s, annealing at 60°C for 45 s, and extension at 72°C for 1 min, with a final extension period at 72°C for 5 min. Competitor samples were separated on a 2% agarose gel and gel purified.

Pools of RNA were created within MH and ML (replicate 1; generations 25–27). Decreasing concentrations of the RPL3 competitor were amplified together with a fixed amount of cDNA template (representing pooled RNA). Complementary DNA were reverse transcribed using the cDNA protocol previously described with one exception. The 3' primer used to initiate reverse transcription was RPL3 3R 5'-GCATTGTTCTTGATCAGTTTGCC-3' instead of an oligo(dT). PCR was carried out in a total volume of 10 µl containing 1 µl of RPL3 cDNA reaction, 1 µl of competitor, 1.5 mM MgCl2, 0.25 mM each dNTPs, 0.4 µM each primer (forward primer RPL3 2F, 5'-TCATTGCCCACACYCAGATG-3'; reverse primer RPL3 3R, 5'-GCATTGTTCTTGATCAGTTTGCC-3'), 0.5 U Taq polymerase with 1x MgCl2-free supplied buffer (Sigma-Aldrich), and 1x RediLoad. Thermal cycling conditions included denaturing at 95°C for 2 min, followed by 25 cycles with denaturing at 95°C for 30 s, annealing at 60°C for 45 s, and an extension at 72°C for 1 min, with a final extension period of 72°C for 5 min. PCR products were electrophoresed at 100 V for 1.5 h in a 4% metaphor (FMC Bioproducts, Rockland, ME) gel stained with ethidium bromide and visualized under ultraviolet (UV) light. Competitive RT-PCR results were interpreted by visually comparing the competitor concentration at which the competitor and the target products appear equal in intensities within the ML and MH lines. The greater the competitor concentration when the competitor and target become equal in product intensities, the greater the expression of RPL3.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

Molecular analysis of gene expression.
DD-PCR analysis of IH and IL tissues was accomplished using a pooling strategy that allowed for differences in mRNA to be accounted for within families. DD-PCR in hypothalamus yielded 98 bands that were potentially expressed differentially in the IL and IH pools. Of the 98 bands excised, 51 bands (52%) represented greater expression in IL vs. IH. Bands were prioritized for evaluation based on the strength and consistency of differences in intensity of DD-PCR products (See Fig. 1). Because of the lower yield of total RNA from mouse hypothalamus available for Northern blots, the number of bands selected for evaluation became limited. Thus only the nine strongest candidates were chosen for Northern analysis evaluation and sequence identification.



View larger version (98K):
[in this window]
[in a new window]
 
Fig. 1. Hypothalamus DD-PCR products from 2 different primer combinations (see numbered groups 1 and 2). Lanes labeled "L" represent inbred low (IL) line pools, and lanes labeled "H" represent inbred high (IH) line pools. Plus symbol (+) indicates an increase in DD-PCR products IL relative to IH for primer combination 2. This autoradiograph represents ~10% of the length of the gel.

 
DD-PCR in BAT tissue resulted in 65 bands that showed differential expression when the IH and IL pools were compared. Of these, 40 bands (61.5%) were expressed at greater levels in IL compared with IH. The best 18 candidates were chosen for Northern analysis evaluation and sequence identification.

Evaluation of differential display products.
Table 1 lists results from Northern analysis and sequencing of the nine hypothalamus DD-PCR candidates chosen for evaluation. Four differences observed in DD-PCR were confirmed by Northern analysis. Three of the four candidates yielded sequences with full homology to mouse ribosomal protein L3 (RPL3), whereas the fourth remains an unknown EST. Figure 2 demonstrates Northern analysis confirmation showing increased RPL3 mRNA expression in ML relative to MH.


View this table:
[in this window]
[in a new window]
 
Table 1. Characterization and sequence analysis of 9 bands differentially expressed as revealed by mRNA differential display of hypothalamus

 


View larger version (75K):
[in this window]
[in a new window]
 
Fig. 2. Northern blot hybridization confirmation of hypothalamus DD-PCR result (see Fig. 1). Top: autoradiograph of DD-PCR product used as a probe to confirm DD-PCR result. Bottom: autoradiograph of hypothalamus membrane standardized using a glyceraldehyde-3-phosphate dehydrogenase (G3PDH) probe and showing equal loading and standardization. RPL3, mouse ribosomal protein L3.

 
Results of Northern analysis of the 18 DD-PCR candidates from BAT are shown in Table 2. Expression change of only one DD-PCR product was confirmed, yielding sequence with full homology to mouse ribosomal protein L3. This product resulted from the same primer combination as one of the three hypothalamus DD-PCR products that yielded RPL3. Eleven of the 18 candidates (see Table 2) had 100% homology with sequences previously deposited in GenBank, three were partial matches, and four had no match, yielding new ESTs.


View this table:
[in this window]
[in a new window]
 
Table 2. Characterization and sequence analysis of 18 bands differentially expressed as revealed by mRNA differential display of BAT

 
RPL3 was expressed at greater levels in IL than in IH mice in both hypothalamus and BAT. Three DD-PCR products in the hypothalamus and one in the BAT showed 100% identity to part of the coding region and the 3' untranslated region of RPL3 (GenBank accession no. AI448022).

Further characterization of RPL3.
To verify that DD-PCR products being used as probes in the Northern analysis were RPL3 and to confirm the results, primers were designed from existing mouse RPL3 cDNA sequence (GenBank accession no. AI448022) to be used as a probe in Northern analysis. These primers were designed to amplify a 400-bp product of RPL3. Northern blots using pools of MH and ML total RNA from BAT, with the probe designed from the mouse RPL3 sequence, confirmed the previous Northern results (data not shown). Levels of RPL3 mRNA were ~18–24% greater in BAT of ML mice compared with MH mice.

A competitive RT-PCR test was developed to decrease the amount of RNA needed to compare RPL3 expression differences, thus enabling more extensive testing of individuals and populations, and additional tissues. Figure 3 shows the results of competitive RT-PCR for BAT on one of the IL and one of the IH mice used in the original DD-PCR pools.



View larger version (32K):
[in this window]
[in a new window]
 
Fig. 3. Competitive RT-PCR of RPL3 in brown adipose tissue (BAT). RPL3 was expressed greater in IL relative to IH. BAT mRNA from inbred animals IL 10101 and IH 9102, previously used for DD-PCR. Plus symbol (+) indicates when competitive RT-PCR reached equality of competitor and target PCR products.

 
Figure 4 shows competitive RT-PCR results of pooled hypothalamus from MH and ML. Five distinct pools of RNA from hypothalami of ML and MH mice were evaluated. Each pool consisted of five hypothalami selected randomly from 100 ML and 100 MH mice from replicate 1, generation 27. All conditions involving the rearing and harvesting of tissues matched those used in the differential display analysis. Three of the five competitive comparisons showed strong evidence of greater RPL3 mRNA in ML. Two of the five competitive RT-PCR comparisons resulted in no difference in expression. These may represent sampling variance, indicating a range in heat loss measurements within lines resulting from the random selection of mice with no prior knowledge to individual heat loss measurements.



View larger version (50K):
[in this window]
[in a new window]
 
Fig. 4. Competitive RT-PCR of RPL3 in pools of ML and MH hypothalamus. Plus symbol (+) indicates when competitive PCR reached equality of competitor and target PCR products.

 
Competitive RT-PCR results in epididymal fat and skeletal muscle also indicated increased expression of RPL3 mRNA in ML compared with MH (data not shown). Thus competitive RT-PCR confirmed the DD-PCR and Northern analysis results and extended the findings to epididymal fat and skeletal muscle.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Our results indicate that the RPL3 gene is expressed at higher levels in several tissues in a line of mice selected for low heat loss, relative to the divergently selected high heat loss line. The ML line consumes less food, is fatter and is less active then the MH line. Vincent et al. (24) used DD-PCR and Northern analysis to compare gene expression in skeletal muscle of the ob/ob mouse and its normal littermates (ob/+). They found 12 differentially expressed cDNA in the ob/ob compared with its normal littermates, with 8 being overexpressed in ob/ob. Among these eight messages overexpressed in ob/ob was RPL3. Similarities between the ob/ob mouse model and the low heat loss model (ML) in this study include increased body fat, decreased thermogenesis, and decreased core body temperature compared with ob/+ and MH mice, respectively. Differences between these models are that ML consumes less food than MH, whereas ob/ob have a dramatically greater food intake than +/ob, and no difference exists in body weight between the ML and MH lines, whereas body weight in ob/ob is much higher than its normal littermates. The latter dissimilarity also highlights the fact that differences in body fat levels between ob/ob and controls are much more dramatic then those observed between ML and MH. It is critical to note that ob/ob is caused by a mutation within a single gene (29), whereas the heat loss and body composition divergences in the ML and MH lines are polygenic in nature (18).

The quantitative genetic model employed in this project may be a reason why so few expression differences were found relative to other DD-PCR experiments. Under the assumption that response to selection for heat loss was due to allele frequency changes in a relatively large number of genes, each with relatively modest effect, then a lack of genes whose expression changes are detectable by this method may exist. To our knowledge, this is the first report of DD-PCR detection of gene expression changes resulting from long-term selection for a quantitative trait.

RPL3 sequence is highly conserved across mammalian species, lower eukaryotes, and archaebacteria (23, 8, 4). It is the largest protein of the 60S subunit of the ribosome in eukaryotes and acts as the center channel through which the new peptide emerges. Some mutations in the yeast RPL3 gene give rise to resistance to the antibiotic trichodermin, an inhibitor of ribosome peptidyl transferase activity (4). In yeast, the repression/activator protein (Rap1p) gene is a transcriptional factor for RPL3 and other ribosomal proteins. Truncation of Rap1p represses the transcription of RPL3. Rap1p is an interesting transcription factor not only because it promotes transcription of RPL3 and other ribosomal proteins, but because it also promotes transcription of translation factors and some genes involved in the glycolytic pathway (13).

Using what is known in yeast about RPL3, one can speculate as to the possible mechanisms that account for the differences in phenotype in the high and low heat loss lines of mice. If a transcription factor exists in mammals similar in nature to Rap1p which increases transcription of some genes involved in the glycolytic pathway, then this could lead to increased glycolysis. This would result in excess energy production, making the animal more efficient in energy substrate utilization. Therefore, the resulting positive energy balance might lead to increased fat deposition and signal for decreased energy consumption. Assuming that the increase in RPL3 mRNA leads to an increase in RPL3 protein, this may act as a transcriptional activator for a gene(s) that results in decreased heat loss or as a repressor in the transcription of a key gene(s) involved in increased heat loss.

In yeast, genes that encode ribosomal proteins are affected both by changes in the carbon source and amino acid availability (27). Thus expression of RPL3 may be influenced by differences in amino acid availability in the heat loss selection lines. Additionally, mRNA expression of RPL3 can be dramatically decreased in yeast by an upward shift in temperature (2). The MH and ML lines are known to have significantly different core body temperatures (17). The elevated core body temperature in MH mice might lead to a decrease in expression of RPL3 mRNA.

Divergent selection for heat loss has resulted in increased expression of RPL3 mRNA in the low ML line relative to the high MH line. At present we cannot conclude whether this is a direct or correlated response to selection. A direct response would indicate that RPL3 is a QTL for heat loss, containing heritable genetic variation within the locus. However, we have localized RPL3 to chromosome 15 (unpublished results), on which a genome-wide scan in a cross between MH and C57BL/6J (16) failed to find evidence of heat loss QTL. A correlated response would indicate that expression levels of RPL3 have been changed as a result of direct or indirect interaction between one or more heat loss QTL and the RPL3 locus.

The finding of increased expression of RPL3 in all ML tissues evaluated is intriguing. This supports a general hypothesis that, as a result of selection for heat loss, ML and MH mice may differ in overall translational activity in many tissue types. The ramifications of this toward understanding energy balance remain to be evaluated.

Although the advantages of DD-PCR are significant, the technique also has disadvantages. False positives are the largest problem when dealing with DD-PCR. In addition, the process of confirming the DD-PCR results with other expression methods is typically more time consuming than conducting the DD-PCR (12, 3, 9, 11).

The use of DD-PCR with the support of other gene expression analyses has been a helpful tool in elucidating differences in gene expression between these selection lines of mice. Discoveries of differences in gene expression will likely be an important tool in positional cloning of obesity QTL and loci that interact with QTL. The use of microarray technology on congenic lines in combination with fine mapping has already been used to positionally clone a potential QTL (1). The use of DD-PCR coupled with new technologies, such as microarrays, will continue to improve the understanding of the biology of energy balance.

Summary
Results from DD-PCR and confirmation with Northern analysis and competitive RT-PCR show strong evidence that RPL3 mRNA is expressed at a greater level in the ML vs. MH heat loss selection lines of mice. Expression differences were found in hypothalamus, BAT, skeletal muscle, and epididymal fat. The finding of Vincent et al. (24) that RPL3 has 60% greater expression in skeletal muscle in the ob/ob mouse compared with its normal littermates, along with the finding in this study of increased RPL3 mRNA expression in the ML selection line, provides impetus for further study of the role of RPL3 in energy balance in mammals.


    ACKNOWLEDGMENTS
 
We thank Diane Moody, Lori Messer, and Christy Gladney for assistance with data collection, and we thank Jeryl Hauptman for experimental animal care.

This work is published as paper no. 12954 of the Journal Series, Nebraska Agricultural Experiment Station.


    FOOTNOTES
 
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).

Address for reprint requests and other correspondence: D. Pomp, Dept. of Animal Science, Univ. of Nebraska ANS A218h, Lincoln, NE 68583-0908 (E-mail: dpomp{at}unl.edu).


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Aitman TJ, Glazier AM, Wallace CA, Cooper LD, Norsworhy PJ, Wahid FN, Al-Jajali KM, Trembling PM, Mann CJ, Shoulders CC, Graf D, St. Lezin E, Kurtz TW, Kren V, Pravenec M, Ibrahimi A, Abumrad NA, Stanton LW, and Scott J. Identification of Cd36 (Fat) as an insulin-resistance gene causing defective fatty acid and glucose metabolism in hypertensive rats. Nat Genet 21: 76–83, 1999.[ISI][Medline]
  2. Baojie L, Nierras RC, and Warner JR. Transcriptional elements involved in the repression of ribosomal protein synthesis. Mol Cell Biol 19: 5393–5404, 1999.[Abstract/Free Full Text]
  3. Burrows PD and Luce MJ. Minimizing false positives in differential display. Biotechniques 24: 766–768, 1998.[ISI][Medline]
  4. Chan HY, Zhang Y, Hoheisel JD, and O’Kane C. Identification and characterization of the gene for Drosophila L3 ribosomal protein. Gene 212: 119–125, 1998.[ISI][Medline]
  5. Clement K, Vaisse C, and Lahlou N. A mutation in the human leptin receptor gene causes obesity and pituitary dysfunction. Nature 392: 398–401, 1998.[ISI][Medline]
  6. Duggan DJ, Bittner M, Chen Y, Meltzer P, and Trent JM. Expression profiling cDNA microarrays. Nat Genet 21: 10–19, 1999.[ISI][Medline]
  7. Jones LD, Nielsen MK, and Britton RA. Genetic variation in liver mass, body mass, and liver: body mass in mice. J Anim Sci 70: 2999–3006, 1992.[Abstract/Free Full Text]
  8. Kuwano Y and Wool G. The primary structure of rat ribosomal protein L3. Biochem Biophys Res Commun 187: 50–64, 1992.
  9. Ledakis P, Tanimura H, and Fojo T. Limitations of differential display. Biochem Biophys Res Commun 251: 653–656, 1998.[ISI][Medline]
  10. Liang P and Pardee AB. Differential display of eukaryotic messenger RNA by means of the polymerase chain reaction. Science 257: 967–971, 1992.[ISI][Medline]
  11. Matz MV and Lukyanov SA. Different strategies of differential display: areas of application. Nucleic Acids Res 26: 5537–5543, 1998.[Abstract/Free Full Text]
  12. Miele G, MacRae L, McBride D, Manson J, and Clinton M. Elimination of false positives generated through PCR re-amplification of differential display cDNA. Biotechniques 25: 138–144, 1998.[ISI][Medline]
  13. Mizuta K, Tsujii R, Warner JR, and Nishiyama M. The C-terminal silencing domain of Rap1p is essential for the repression of ribosomal protein genes in response to defect in the secretory pathway. Nucleic Acids Res 26: 1063–1069, 1998.[Abstract/Free Full Text]
  14. Montague CT, Farooqi IS, and Whitehead JP. Congenital leptin deficiency is associated with severe early-onset obesity in humans. Nature 387: 903–908, 1997.[ISI][Medline]
  15. Moody DE, Pomp D, and Nielsen MK. Variability in metabolic rate, feed intake and fatness among selection and inbred lines of mice. Genet Res 70: 225–235, 1997.[ISI][Medline]
  16. Moody DE, Pomp D, Nielsen MK, and Van Vleck LD. Identification of quantitative trait loci influencing traits related to energy balance in selection and inbred lines of mice. Genetics 152: 699–711, 1999.[Abstract/Free Full Text]
  17. Mousel MR and Nielsen MK. Daily activity and core body temperature of mice selected for high and low heat loss. J Anim Sci 75, Suppl 1: 144, 1997.
  18. Nielsen MK, Jones LD, Freking BA, and DeShazer JA. Divergent selection for heat loss in mice. I. Selection applied and direct response through fifteen generations. J Anim Sci 75: 1461–168, 1997.[Abstract/Free Full Text]
  19. Nielsen MK, Jones LD, Freking BA, Nelson SM, Vorderstrasse TL, and Hussey BA. Divergent selection for heat loss in mice. II. Correlated responses in feed intake, body mass, body composition, and number born through fifteen generations. J Anim Sci 75: 1461–1468, 1997.[Abstract/Free Full Text]
  20. Perusse L, Chagnon YC, Weisnagel J, and Bouchard C. The human obesity gene map: the 1998 update. Obes Res 7: 111–129, 1999.[Abstract]
  21. Pomp D. Genetic dissection of obesity in polygenic animal models. Behav Genet 27: 285–305, 1997.[ISI][Medline]
  22. Pomp D and Nielsen MK. Quantitative genetics of energy balance-lessons from animal models. Obes Res 7: 107–110, 1999.
  23. Simonic T, Gaudi S, Giussani F, Malcovati M, Ronchi S, and Tenchini ML. cDNA sequence for bovine ribosomal protein L3 carrying a bipartite nuclear targeting motif, identified also in many other ribosomal proteins. Biochim Biophys Acta 1219: 706–710, 1994.[ISI][Medline]
  24. Vincent D, Piper M, Gammeltoft S, Maratos-Flier E, and Kahn CR. Alterations of skeletal muscle gene expression of ob/ob mice mRNA differential display. Diabetes 47: 1451–1458, 1998.[Abstract]
  25. Wardern CH, Fisler JS, Showmaker SM, Wen PZ, Svenson KL, Pace MJ, and Lusis AJ. Identification of four chromosomal loci determining obesity in a multifactorial mouse model. J Clin Invest 96: 1545–1552, 1994.
  26. West DB, Goudey-Lefevere J, York B, and Truett GE. Dietary obesity linked to genetic loci on chromosomes 9 and 15 in a polygenic model mouse model. J Clin Invest 94: 1410–1416, 1994.[ISI][Medline]
  27. Woudt LP, Mager WH, Nieuwint RT, Wassenaar GM, Van der Kuyl AC, Murre JJ, Foekman MF, Brockhoff PG, and Planta RJ. Analysis of upstream activation sites of yeast ribosomal protein genes. Nucleic Acids Res 11: 6037–6048, 1987.
  28. York B and West DB. Sensitivity to dietary obesity linked to a locus on chromosome 15 in a Cast/Ei x C57BL/6J cross. Mamm Genome 7: 677–681, 1996.[ISI][Medline]
  29. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, and Friedman JM. Positional cloning of the mouse obese gene and its human homologue. Nature 372: 425–431, 1994.[ISI][Medline]
  30. Zhang JS, Ducan EL, Chang ACM, and Reddel RR. Differential display of mRNA. Mol Biotechnol 10: 155–165, 1998.[ISI][Medline]