Differential Expression of Genes from Nitrate-Tolerant Rat AortaPi X.1 · Yan C.1 · Kim D.1 · Chen J.1 · Berk B.C.1
Center for Cardiovascular Research, Medical Center, University of Rochester, N.Y., USA Corresponding Author
Organic nitrates act as vasodilators and have long been used for treatment of cardiovascular diseases. However, the therapeutic effect of nitrates is limited by induction of nitrate tolerance which is associated with endothelial dysfunction and enhanced vasoconstriction. Multiple mechanisms cause nitrate tolerance including alterations in gene expression. To identify genes whose expression is altered due to chronic treatment with nitroglycerin (NTG), nitrate-tolerant rats were made by infusion of NTG for 3 days (10 μg/kg/min) using an osmotic minipump. We constructed a PCR-selected cDNA subtracted library from NTG-treated and vehicle-treated rat aortas. Screening of ∼500 clones in the subtracted library showed that four genes were regulated by NTG treatment. Specifically, mRNA levels of β-globin, tropoelastin, gelsolin and a small G protein were confirmed to be upregulated consistently by NTG treatment. These identified genes may play important roles in the development of nitrate tolerance and represent promising candidates to understand the mechanisms of nitrate tolerance and endothelial dysfunction in the vasculature.
Copyright © 2002 S. Karger AG, Basel
Organic nitrates have been widely used as vasodilators for the treatment of angina pectoris and heart failure. The major action of organic nitrates is to decrease vascular smooth muscle cell (VSMC) tone which leads to dilation of peripheral and coronary arteries, as well as peripheral veins . Within VSMC, it is recognized that organic nitrates (RONO2) are converted to short-lived S-nitrosothiols (RSNO), which then activate guanylate-cyclase-stimulating cGMP formation [2, 3]. cGMP leads to vasodilation by decreasing intracellular Ca2+ via reuptake of Ca2+ by the sarcoplasmic reticulum and reducing permeability to extracellular Ca2+ . However, the efficacy of nitrates is limited by tolerance that develops shortly after treatment begins. Tolerance to the effects of organic nitrates in humans was first described by Laws in 1898 and has been shown repeatedly since then in both animals and humans. Renewed interest and intensive scientific effort in the last decade have helped to illuminate this issue and have clearly shown that tolerance, if ignored, may severely limit the benefit of nitrate therapy.
Because nitrates utilize many of the same pathways that the endogenous vasodilator nitric oxide uses, understanding mechanisms of nitrate tolerance should provide insight into fundamental aspects of vessel function and endothelial dysfunction. Mechanisms proposed for nitrate tolerance include impaired nitrate biotransformation , intracellular sulfhydryl group depletion , neurohumoral counterregulation , overproduction of reactive oxygen species , and alteration in the activities of the key enzymes which regulate cGMP levels, such as guanylate cyclase and cyclic nucleotide phosphodiesterase (PDE) .
Because nitrate tolerance is associated with chronic drug treatment, changes in gene expression are likely involved in its development. Several genes important for the regulation of vascular tone have been reported. For example, decreased expression of protein kinase G mRNA and protein has been demonstrated in VSMC in vitro and in vivo in response to continuous exposure to nitroglycerin (NTG) . NTG treatment for 3 days increased endothelial nitric oxide synthase (eNOS) mRNA and protein expression in rat aortas . In the setting of nitrate tolerance, eNOS generates excess superoxide which may decrease vascular NO bioavailability . Neurohumoral counterregulatory systems have also been found to be activated rapidly in response to NTG treatment and play important roles in the development of nitrate tolerance  13]. In particular, increases in angiotensin II stimulate many other events including increases in angiotensin II type 1 receptor , endothelin-1 (ET-1) , ET-1 receptor , and NAD(P)H oxidase . We previously found that expression of PDE1A1, a Ca2+/calmodulin-stimulated cGMP-hydrolyzing PDE, was upregulated by chronic exposure to NTG . Therefore, we propose that the levels of several genes, in addition to the known ones, are regulated by exposure to NTG.
To identify candidate genes, we used normalized subtractive hybridization. We found 50 candidate genes whose mRNA levels were increased in the 3-day NTG-treated rat aorta, compared with the vehicle-treated aorta. After analysis by Northern blotting of single aortic RNA samples, four genes were confirmed to show consistent upregulation – β-globin, tropoelastin, gelsolin and a small GTPase (Ras-like Ray).
material and methods
All animal procedures were conducted in accordance with the guidelines recommended by the University Committee on Animal Resources (University of Rochester).
Male Sprague-Dawley rats (Charles River Laboratories) were housed and maintained in a temperature-controlled room regulated on a 12-hour light/dark cycle. Rats (250–300 g) were anesthetized with ketamine 40 mg/kg, xylazine 0.5 mg/kg, acepromazine 5 mg/kg i.p. plus 1/3 loading dose as needed. An osmotic minipump (model 2ML1, Alza) filled with either NTG (Zeneca, Wilmington, Del., USA) or vehicle (propylene glycol) was implanted subcutaneously at the dorsum of the neck. NTG was infused at an average rate of 10 μg/kg/min for 3 days. Thoracic aortas were snap-frozen in liquid nitrogen after removal of adventitia and stored at –80°C until use.
RNA from rat aortas was prepared with the ToTally RNA kit (Ambion) based on the manufacturer’s protocol.
The preparation of cDNA from NTG (10 μg/kg/min infusion for 3 days)- and placebo-treated (propylene glycol infusion for 3 days) aorta RNAs was performed with the Clontech SMARTTM PCR cDNA synthesis kit.
Subtractive hybridization was performed on RNA pools of 6 NTG-treated or placebo-treated aortas. Subtraction was performed using the Clontech PCR-select™ cDNA subtraction kit. Through two steps of hybridization between NTG-treated aorta cDNA and vehicle-treated aorta cDNA, and suppression PCR, the hybridized cDNAs were removed and the remaining unhybridized cDNAs represented genes that are upregulated in NTG-treated aorta. The subtracted cDNA libraries were constructed with the T/A cloning vector.
Differential screening was performed with the Clontech PCR-select differential screening kit. The dot blots (containing clones of the subtracted library) were hybridized with first-strand cDNA probes from NTG-treated aorta and vehicle-treated aorta, and also the forward- and reverse-subtracted cDNA probes. Probes were radiolabeled with the T7 PrimeQuick labeling kit (Amersham Pharmacia Biotech). The counts for labeling were 1–2 × 106 cpm/ml hybridization solution. The positive clones were chosen based upon the comparison of the difference of hybridization signals between these different probes.
The candidate clones obtained from subtractive hybridization were sequenced with the BigDye Terminator Cycle Sequencing Kit and T7 or M13 reverse primer, and then analyzed on the ALF-automatic sequencer by the core lab of the University of Rochester.
For Northern blotting, 10 μg of total RNA per lane was run on a formaldehyde-agarose gel and transferred to Hybond nylon membranes (Amersham Pharmacia Biotech) or Zeta-Probe™ GT genomic-tested blotting membrane (Bio-Rad). For slot blotting, 2 μg total RNA was mobilized onto the membrane. The blots were hybridized in formamide-containing buffers according to standard procedures . Probes were radiolabeled with the T7 PrimeQuick labeling kit (Amersham Pharmacia Biotech), and used at 1–2 × 106 cpm/ml hybridization solution. GAPDH cDNA served as a control for equal RNA loading on the multiple tissue blots.
Student’s t test for unpaired observation was used to determine the significance of differences between NTG- and vehicle-treated rat aorta samples. All values are given as means ± SE, and statistical significance was set at p< 0.05.
We have previously characterized a rat model of nitrate tolerance as assayed by the hemodynamic responses to NTG infusion . No significant difference in the baseline of mean arterial pressure between the control and the NTG-pretreated groups was found, but the change in blood pressure to NTG was significantly blunted in the NTG-pretreated groups compared to the controls (data not shown). NO-independent vasodilation (assessed by the decrease in pressure in response to hydralazine) was not different in NTG-pretreated groups .
Subtractive hybridization is a very powerful technique to identify differentially expressed cDNAs. In general, the PCR-selected method greatly enriches for differentially expressed sequences. The hybridization step after the construction of suppression subtractive cDNA library enables us to screen more than one hundred genes at one time. After that, we used Northern blotting to confirm the changes of genes in subtractive hybridization.
We synthesized cDNA from NTG (6 rats)- and vehicle-treated (6 rats) rat aorta RNA with reverse transcription-PCR cDNA synthesis method and constructed the forward subtractive cDNA library. Then we performed differential screening of more than 300 clones in the forward library, as shown in figure 1. We identified 50 candidate clones with potentially differential expression between 3-day NTG vs. vehicle-treated rat aortas.
Fig. 1. Differential screening of cDNA clones generated by subtractive hybridization. Two identical blots were made by transfer of PCR products to nylon membranes. The dot blots were hybridized to a 32P-labeled NTG (a)- or control (b)-subtracted cDNA probe. Open arrows indicated an upregulated clone; filled arrow indicated an unchanged clone.
Using slot blotting analysis, we identified 13 clones (26%) showing differential expression in NTG-treated aortas compared to vehicle-treated aortas. Figure 2 shows representative slot blotting results.
Fig. 2. Representative slot blot analysis with candidate clones found by differential screening. a F100, mRNA level increased significantly. b F114, mRNA level did not change.
The total RNAs of NTG- and vehicle-treated rat aortas for subtractive hybridization were prepared from pooled samples of 6 aortas. To confirm the validity of the 13 clones identified with slot blotting, we performed analysis of each mRNA isolated from individual aortas from at least 3 pairs of rats. As shown in table 1, we classified our findings into two groups. Group 1 mRNAs showed significant increases in each set of paired (vehicle- and nitrate-treated) aorta. These mRNAs include tropoelastin, β-globin, gelsolin and a small GTPase (Ras-like Ray). Group 2 represented mRNAs that did not have significant changes (p > 0.05) when analyzed individually by Northern blot. These included connexin 40, osteoprotegerin, hsp90, ER transmembrane protein, hsp40, GS28, UDP-N-GlcNAc pyrophosphorylase, α-NAC transcription coactivator, and U5 SnRNP-specific protein. The last 5 genes mentioned above exhibited individual variation in the mRNA levels (table 1).
Table 1. Grouping of genes identified from subtractive hybridization and differential screening
Among the group 1 genes, the β-globin gene has several interesting properties. First, two hybridization bands appeared by Northern blotting analysis (table 2, fig. 3); one of 3.5 kb and the other was small (0.3 kb). There appeared to be differential regulation of these two mRNA species with one (0.5 kb) increasing 2.31 ± 0.82 times and the other (3.5 kb) increasing 1.93 ± 0.56 times in NTG-treated aorta, compared to vehicle-treated aorta. These two RNAs with different sizes could represent two isoforms of the β-globin gene produced through alternative splicing. Gelsolin, tropoelastin and ras-like Ray mRNA levels were significantly upregulated by NTG treatment (1.50 ± 0.42, 1.58 ± 0.47, 2.14 ± 0.66 times, respectively) compared to vehicle treatment. The small GTPase ras-like Ray has been shown to be expressed in aorta, bladder, brain, heart, kidney, lung, spleen and testis. In the intestine, liver and stomach, the expression was too low to be identified by Northern blotting (fig. 4).
Table 2. Differentially regulated genes in NTG-treated aorta confirmed by Northern blotting
Fig. 3. mRNA levels of genes significantly increased in NTG-treated aorta, compared to vehicle-treated aorta. N = NTG-treated aorta; C = vehicle-treated aorta. a Northern blotting analysis. b Statistical analysis of mRNA level changes, p < 0.05.
Fig. 4. Multiple Northern blot analysis of the Ras-like Ray. Intest. = Intestine; Sk.Mus. = skeletal muscle.
The major finding of the present study is that nitrate tolerance is associated with specific upregulation of several mRNAs in the rat aorta. Specifically, we found that four mRNAs from the forward subtractive library showed consistent and significant increases in nitrate tolerance (tropoelastin, β-globin, gelsolin and a small GTPase, Ras-like Ray). In addition, another group of five mRNAs showed highly regulated expression that varied significantly among individual aorta (Hsp40, GS28, UDP-N-GlcNAc pyrophosphorylase, α-NAC transcription coactivator, and U5 SnRNP-specific protein). Of greatest significance, none of these genes would have been predicted to be regulated during nitrate tolerance on the basis of hemodynamic or neurohormonal mechanisms. The reverse subtractive library was not screened. In the discussion below we will focus on the four mRNAs that were consistently increased.
The hemoglobin family of proteins is well known to act as oxygen carriers in red blood cells. However, the function of the globins in cells other than red blood cells is poorly defined. Recently, several studies have suggested that globins may play roles in the regulation of free radical species, such as NO. Poole and Hughes  reviewed recent studies regarding the microbial hemoglobins and suggested that a major role for microbial flavohemoglobins (Hmp) is detoxification of NO. Hmp levels are low in unstressed cells, but could act as an NO trap, sequestration of NO by haem group or nitrosylation of hmp by NO+, when induced by NO or nitrosative stress . This suggests that it is possible that the β-globin gene, the homologue of hmp in mammalian cells, may also play an important role in the regulation of NO bioavailability. This hypothesis is supported by the phenotype observed in a transgenic mouse model of sickle cell disease . This transgenic mouse model is homozygous for the deletion of mouse β-globin and contains transgenes for human βS and βS-antilles globins linked to the transgene for human α-globin. Physiological and biochemical studies of aorta rings isolated from these mice demonstrate a phenotype which is quite similar to nitrate tolerance. In these mice, relaxation to the endothelium-dependent vasodilator A-23187 and to a nitric oxide donor DEA-NONOate were impaired. These results in the sickle mouse are consistent with endothelial dysfunction and suggest that β-globin is involved in the regulation of vascular tone by nitric oxide. In combination with our findings that β-globin was upregulated in nitrate-tolerant aorta, we propose an important role for β-globin in vessel function.
The tropoelastin gene generates the protein elastin which is an extracellular matrix component responsible for tissue elasticity. Previously tropoelastin gene expression was shown to be increased during vascular development and after arterial injury [17, 20]. Soluble elastin peptides are generated by elastin degradation and may be released into the surrounding extracellular environment. These elastin peptides have diverse biological functions including stimulating fibroblast adhesion to elastin fibers , promoting cell proliferation , and stimulating chemotaxis of monocytes and fibroblasts . It has been suggested that the elasticity of the extracellular matrix is important in the regulation of arterial tone . However, how the upregulation of tropoelastin during nitrate tolerance is related to nitrate tolerance development is unknown and needs further investigation.
Gelsolin is a member of an actin filament severing and capping protein family. Gelsolin may play a role in vascular tone since it has been shown that actin polymerization contributes significantly to the development of myogenic tone after forced dilation induced by an increase in intravascular pressure . It is possible that, similar to the regulation of myogenic response, the actin cytoskeleton of vascular smooth muscle also plays an important role in maintaining vascular tone. Gelsolin may also regulate other cellular functions related to nitric oxide and cGMP, such as cell proliferation and migration. Numerous papers have implicated gelsolin in cell functions including neurite filopodial retraction , modulation of neuronal ion channels , and regulation of apoptosis .
The fourth gene shown to be upregulated in nitrate tolerance was a small GTPase termed ras-like Ray. Homology analysis indicates that this protein is a member of the Ray family. The function of ras-like Ray is largely unknown. It has not previously been described in blood vessels and there are no published data regarding tissue distribution. Our data show ubiquitous expression in aorta, bladder, brain, heart, kidney, lung, spleen and testis.
Previously, we have found PDE1A1 expression was increased by chronic exposure to NTG, which at least partially contributes to nitrate tolerance . This study demonstrates more genes are regulated by long-term NTG treatment, suggesting the regulation of gene expression is one mechanism of nitrate tolerance development. However, other kinds of regulation might also be involved. For example, in nitrate-tolerant mice with eNOS overexpression, soluble guanylate cyclase activity was reduced more than two fold without change in protein level . These results suggest that a posttranslational modification is also important in the development of nitrate tolerance.
In summary, our study has identified several interesting candidate genes that are associated with nitrate tolerance. These genes have not been previously discussed in this context, although there is evidence showing potential involvement of these genes in the regulation of vascular tone. They also represent potential genes for study in endothelial dysfunction caused by pathophysiological states such as hypertension and hyperlipidemia since nitrate tolerance resembles endothelial dysfunction. Finally, these genes may represent possible targets for the development of new therapeutics for nitrate tolerance and hypertension.
This work was supported by NIH grants HL49192, 62826 and 63462 to B.C.B. and the AHA Research Grant (No. 0030302T) to C.Y. We would like to acknowledge the generous help of Dr Joseph M. Miano.
Dr. Bradford C. Berk
Department of Medicine
Box MED, University of Rochester School of Medicine
601 Elmwood Avenue, Rochester, NY 14642 (USA)
Tel. +1 585 275 0810, Fax +1 585 273 1497, E-Mail Bradford_Berk@urmc.rochester.edu
Received: Received: October 11, 2001
Accepted after revision: January 7, 2002
Number of Print Pages : 7
Number of Figures : 4, Number of Tables : 2, Number of References : 29
Journal of Vascular Research
Founded 1964 as Angiologica by M. Comèl and L. Laszt (1964–1973) continued as Blood Vessels by J.A. Bevan (1974–1991)
Vol. 39, No. 4, Year 2002 (Cover Date: July-August 2002)
Journal Editor: M.J. Mulvany, Aarhus
ISSN: 1018–1172 (print), 1423–0135 (Online)
For additional information: http://www.karger.com/journals/jvr