Induction of 15-Lipoxygenase mRNA and Protein in Early Atherosclerotic Lesions
Background 15-Lipoxygenase (15-LO) may be involved in atherogenesis and in oxidative modification of LDL. In this study, we investigated 15-LO expression in developing atherosclerotic lesions and verified the exact type of the atherosclerosis-associated LO at the nucleotide level.
Methods and Results Quantitative reverse transcription–polymerase chain reaction, in situ hybridization, and immunocytochemistry were used in two models of experimental atherosclerosis. New Zealand White rabbits were given a 1% cholesterol diet for 0 (control group), 3, 6, or 14 weeks. 15-LO mRNA was undetectable in the aortic intima-medias of the control group, whereas it was clearly induced as early as after 3 weeks. 15-LO expression increased further in the 6- and 14-week groups. According to in situ hybridization and immunocytochemical studies, 15-LO was localized to macrophage-rich areas. In Watanabe heritable hyperlipidemic rabbits, 15-LO mRNA was undetectable in normal aortic intima-medias. 15-LO mRNA was markedly induced in fatty streaks but less so in more advanced lesions. Identification of the induced LO as reticulocyte-type 15-LO was done by cloning and sequencing. We also observed a distinct basal expression of copper-zinc and extracellular superoxide dismutases in normal aortic intima-medias, but no clear induction of these mRNAs was detected in atherosclerotic aortas.
Conclusions The results show that, in contrast to copper-zinc and extracellular superoxide dismutases, the expression of reticulocyte-type 15-LO is markedly induced in rabbit fatty streaks. This may lead to an increase in the oxidative potential during the early phase of atherogenesis and contribute to the development of atherosclerotic lesions.
Oxidative modification of LDL plays an important role in atherogenesis.1 In vivo, several mechanisms may lead to LDL oxidation. These include 15-LO, hydroxyl radical, superoxide anion, peroxynitrite, myeloperoxidase, and ceruloplasmin (for review see Reference 2). In more advanced lesions, autooxidation may contribute to the peroxidation of lesion lipids.3 4
LOs catalyze dioxygenation of fatty acids containing a 1,4-cis,cis-pentadiene structure. Their nomenclature is based on the predominantly oxygenated carbon of free arachidonic acid, but some LOs can also use other substrates.5 Several lines of evidence suggest that LOs are involved in LDL oxidation. Soybean 15-LO,6 recombinant human reticulocyte 15-LO,7 and LO reaction products8 can cause LDL oxidation in vitro. On the other hand, various LO inhibitors block cell-mediated oxidation of LDL.9 10 Reticulocyte-type 15-LO is expressed in human and rabbit atherosclerotic lesions.11 12 LO reaction products have been detected in atherosclerotic lesions,13 14 and transfer of 15-LO cDNA into fibroblasts leads to enhanced levels of lipid hydroperoxides in LDL.15 In addition, transfer of 15-LO gene into rabbit iliac arteries leads to the appearance of oxidation-specific lipid-protein adducts in the transduced arteries.16 However, verification of the lesion-associated LO type at the nucleotide level has not been reported. Little is also known about the time course and regulation of 15-LO expression during the early phase of atherogenesis.
Several antioxidative mechanisms protect cells and LDL from oxidative damage.2 3 These include various antioxidative enzymes, such as CuZn-SOD, EC-SOD, and Mn-SOD, glutathione peroxidases, and catalase.3 CuZn-SOD is mainly a cytoplasmic enzyme. Mn-SOD is located in mitochondria, whereas EC-SOD functions principally at cell membranes and in extracellular space.17 To obtain an estimate about the expression of antioxidative enzymes in developing atherosclerotic lesions, we analyzed the expression of CuZn-SOD and EC-SOD in the same samples that were used for the 15-LO analyses.
RT-PCR has substantially improved the detection of gene transcripts in experimental models in which only limited amounts of tissue are available for analyses. Competitive RT-PCR is based on the simultaneous amplification of a target gene and an internal standard that have identical primer binding sites.18 19 20 In this study, we used competitive RT-PCR to quantify the expression of 15-LO, CuZn-SOD, and EC-SOD mRNAs at various phases of atherogenesis. The 15-LO findings were confirmed with in situ hybridization and immunocytochemistry. The results suggest that, in early atherosclerotic lesions, a clear increase occurs in the expression of reticulocyte-type 15-LO mRNA and protein, whereas the expression of CuZn-SOD and EC-SOD mRNAs remains unchanged.
Male NZW and WHHL rabbits were used for the studies. The NZW rabbits were 3 months old at the beginning of the experiment. They consumed either a standard diet (control rabbits; n=2) or a diet containing 1% cholesterol (wt/wt) for 3, 6, or 14 weeks (n=5 in each group). To further verify the level of 15-LO mRNA expression in aortic intima-medias of control NZW rabbits, five additional control animals were used for the analysis of 15-LO mRNA expression in normal aortas. One rabbit in the 14-week group ceased eating before the end of the experiment and was excluded from the study. WHHL rabbits were maintained on a standard diet, and 2 of them were killed at the age of 3 months, 1 at 7 months, and 1 at 18 months. In addition, 7- and 11-month-old WHHL rabbits were perfusion-fixed for in situ hybridization studies. Blood samples were collected and analyzed for plasma cholesterol content by enzymatic methods (Boehringer Mannheim). The mean plasma cholesterol concentrations of control and cholesterol-fed NZW rabbits were 1.3±0.8 and 50.5±13.8 mmol/L, respectively. WHHL rabbits had plasma cholesterol levels of 23.6±11.3 mmol/L.
The rabbits were anesthetized with an intravenous injection of phentanyl-fluanisone (0.3 mL/kg, Hypnorm, Jansen Pharmaceuticals) and midazolam (1 mg/kg, Dormicum, Hoffmann-La Roche). Aortas were opened, and the extent of atherosclerosis was recorded by naked-eye evaluation. Aortas were classified, according to the extent of macroscopic atherosclerotic lesions, into the following categories: normal intima-medias, fatty streaks, and more advanced fatty plaques. NZW rabbit controls and the 3-week group had no visible atherosclerotic lesions. After 6 weeks on cholesterol feeding, ≈20% of the surface area of the aortic arch was covered with fatty streaks. Rabbits in the 14-week group had extensive fatty streaks and plaques covering almost the entire surface area of the aortic arch. Thus, RT-PCR samples (see below) from control and 3-week groups were macroscopically normal, whereas samples from 6- and 14-week groups represented an average involvement of fatty streaks and/or fatty plaques after the indicated lengths of cholesterol feeding. To minimize the possibility of early fatty streak formation undetectable by naked-eye examination, macroscopically normal intima-medias were obtained from thoracic aortas (excluding orifices of intercostal arteries) of the two young, 3-month-old WHHL rabbits. Excluded parts of the thoracic aortas and the entire aortic arches that were covered with fatty streaks were pooled and analyzed separately. More advanced fatty plaques were obtained from arches and thoracic aortas of the 7- and 18-month-old WHHL rabbits. Intima-medias were dissected free of the adventitia and stored in liquid nitrogen until isolation of RNA. All animal studies were approved by the Experimental Animal Committee of the University of Tampere.
RNA Isolation and RT-PCR Conditions
Total RNA was isolated by a guanidinium thiocyanate method.21 On average, 34±11 μg total RNA was obtained per 100 mg (wet wt) tissue. The quality of the RNAs was checked on agarose gel electrophoresis, which showed intact ribosomal RNA bands. All RNA preparations were found to be free of DNA contamination by amplifying the samples with all PCR primer pairs used in this study.
RNA (3.5 μg per reaction) was reverse-transcribed to first-strand cDNA with avian myeloblastosis virus reverse transcriptase (5 U per reaction) and random hexamer primers (1 μg per reaction) (cDNA Cycle Kit, Invitrogen). Usually, 1/200 of the reverse-transcribed cDNA was used for a single PCR. With 15-LO, the amount of cDNA was doubled. Negative controls were included in every assay as previously described.22 All RT reagents were tested to be free of any DNA contamination with the primers used in this study. PCRs were performed in a 15-μL volume containing 50 mmol/L KCl, 10 mmol/L Tris-HCl (pH 8.8), 0.1% Triton X-100, 1 μmol/L each of the gene-specific primers, and 0.30 U (0.36 U for actin) Dynazyme DNA polymerase (Finnzymes) per reaction. The Table⇓ lists gene-specific dNTP and Mg2+ concentrations, annealing temperatures, and the numbers of PCR cycles. dATP, dCTP, dGTP, and dUTP were used, except for 15-LO amplifications, in which dTTP was substituted for dUTP for better sensitivity. The denaturation step was 3 minutes at 94°C in the first cycle and 1 minute thereafter. The annealing step was 1 minute at the temperature indicated in the Table⇓. The extension step was 1 minute at 72°C, except for the last cycle (11 minutes). For quantification, 1.5 μCi of [α-33P]dCTP (2000 Ci/mmol, Du Pont NEN) was used per reaction in competitive PCR. Because only the internal standard for actin was not homologous with the corresponding cDNA fragment, there was a possibility for heterodimer formation during the late PCR cycles.18 Consequently, except for actin, the final cycle was performed after 1:4 dilution of the amplified sample to bring PCR back into its exponential phase.18
PCR primer sequences and lengths of the amplified fragments are given in the Table⇑. Actin primers were designed according to rabbit α-smooth muscle and β-nonmuscle actin cDNA sequences.23 GAPDH primers were based on rabbit cDNA sequence.24 15-LO primers were designed according to rabbit reticulocyte 15-LO cDNA sequence,25 and they avoided all known 5- and 12-LO sequences. CuZn-SOD primers were designed according to human and rabbit cDNA sequences.26 27
RT-PCR was used to clone a partial cDNA sequence of rabbit EC-SOD. Rabbit aortic RNA was subjected to RT-PCR with human primers spanning a 375-bp fragment of the human cDNA (nucleotides 171 through 545).17 The amplified fragment was isolated, cloned according to standard procedures (TA Cloning Kit, Invitrogen and pGEM-T, Promega), and sequenced from both ends with an Autoread DNA Sequencing Kit (Pharmacia P-L Biochemicals).28 The sequence between the primers has been submitted to EMBL Data Library (accession number X78139). Rabbit EC-SOD cDNA sequence shows an 86% homology with the corresponding human sequence. EC-SOD primers were designed according to the sequence information and spanned a 221-bp fragment of human and rabbit cDNAs.
Internal Standards and Competitive PCR
The same sets of primers were used to amplify both the internal PCR standards and the target cDNAs (Table⇑). The standard for actin analyses was made from a nonhomologous DNA (BamHI/EcoRI fragment of the v-erb B gene, PCR Mimic Construction Kit, Clontech). The CuZn-SOD standard was prepared by amplifying human aortic cDNA with the CuZn-SOD primers. The human-derived standard has a Dde I (Boehringer Mannheim) restriction site that is missing from the rabbit amplification product. The EC-SOD standard was prepared in the same way as the CuZn-SOD standard, but it contained a human-specific restriction site for Sma I (Boehringer Mannheim). The GAPDH and 15-LO standards were generous gifts from Dr H. Lukhaup and Dr H.A. Dresel. The GAPDH standard was initially derived from rabbit cDNA but was manipulated to have a 102-bp deletion between the primer binding sites. The 15-LO standard was prepared analogously, and it had a 117-bp deletion between the primer binding sites.
For quantification of gene transcripts, an equal amount of cDNA was pipetted to each of the PCRs, but different amounts of internal standard were added.20 Quantification was started with an 8- or 10-fold dilution series of the PCR standard to get an initial approximation of the cDNA concentration. After that, a twofold dilution series of the standard was made, and competitive PCR was performed with the radioactive label. However, because no clear induction of CuZn-SOD or EC-SOD mRNA expression was detected during atherogenesis, their cDNA concentrations were determined according to the intensities of the target and standard bands in ethidium bromide–stained gels (SigmaScan/Image Software, Jandel Scientific).18
The PCR products were separated (4 V/cm, 2 hours) in 2% MetaPhor agarose (FMC BioProducts) in 45 mmol/L Tris-borate buffer (pH 8.0) containing 1 mmol/L EDTA. The reactions were performed in duplicate or triplicate, and two thirds of the reactions were used for the electrophoresis. The cDNA- and standard-derived fragments were cut out from the gel, and their radioactivities were determined (1210 Ultrobeta liquid scintillation counter, LKB/Wallac). The ratio of standard versus cDNA radioactivity was counted after the background was subtracted and was further corrected according to the dCTP content of the PCR products. The ratio was plotted against the amount of the added standard. A regression line was drawn through the plot, and the amount of cDNA in the samples was deduced from the point at which the ratio of the radioactivities was 1. The cDNA concentrations were further corrected according to the lengths of the amplified fragments. The mean interassay coefficients of variation for the quantification analyses were as follows: actin 15%, GAPDH 10%, 15-LO 16%, CuZn-SOD 11%, and EC-SOD 22%.
Immunocytochemistry and In Situ Hybridization
Tissue samples for immunocytochemistry and in situ hybridization were taken in a standardized way from aortas of comparable animals at the same age. Aortic tissue was rapidly removed and immersion-fixed for 6 hours in 4% paraformaldehyde.11 Serial paraffin- or OCT-embedded tissue sections (10 μm) were used for the analyses. 15-LO antisense and sense riboprobes were synthesized in the presence of 35S-UTP (Du Pont NEN) from a pBluescript SK plasmid (Stratagene) as described.11 In situ hybridizations were performed on pretreated tissue sections (6×106 cpm/mL). Final wash was with 0.1×SSC at 60°C for 30 minutes. Autoradiography (NTB-2, Eastman-Kodak) was used for signal detection. Nonhybridizing sense probes were used as controls.11
The following antibodies were used for the immunostainings: mouse mAb against macrophages (RAM-11; dilution 1:500)29 ; mouse mAb against smooth muscle cells (HHF-35; dilution 1:500)30 ; and goat antiserum against human reticulocyte–type 15-LO (dilution 1:1000), which recognizes rabbit 15-LO but does not crossreact with 5-LO or platelet-type 12-LO.11 Avidin–biotin–horseradish peroxidase system (Vector Laboratories) was used for the signal detection.11 Class- and species-matched immunoglobulins were used as controls for the immunostainings together with incubations in which the primary antibodies were omitted.11 12
All data are expressed as mean±SD. Quantitative RT-PCR results were first analyzed by one-way ANOVA. After that, Student’s t test was used to compare the NZW rabbit control and WHHL rabbit normal groups with the other groups. Values of P<.05 were considered statistically significant.
Actin and GAPDH mRNA Expression
An example of actin quantification in normal NZW rabbit aorta is shown in Fig 1A⇓. The target cDNA and the internal standard are approximately equally amplified in the middle lane of the insert in Fig 1A⇓. After additional parallel amplifications, a quantification plot was established. Similarly, the GAPDH target cDNA from WHHL rabbit fatty streaks and the internal standard are about equally amplified in the middle lane of the insert in Fig 1B⇓. The visual estimation was confirmed by a quantification plot.
Our results indicate that a twofold to fivefold induction occurs in the amount of GAPDH mRNA in both rabbit models with the development of atherosclerotic lesions (Fig 2⇓). Because actin mRNA levels remained more stable during atherogenesis, actin expression was used to normalize the expression of other genes.
15-LO mRNA Expression
15-LO mRNA expression was barely detectable in the normal NZW aortas (Fig 2A⇑). Similar results were obtained from normal intima-medias of five additional control NZW rabbits, in which 15-LO expression was also below the detection limit (ie, <3×10−4 amol/μg total RNA). After 3 weeks on the cholesterol-rich diet, there was a 15-fold increase in the amount of 15-LO mRNA in NZW rabbit aortas (Figs 1C⇑ and 2A⇑). After 14 weeks on the cholesterol-rich diet, the expression of 15-LO mRNA was increased more than 100-fold compared with that of the control aorta (Fig 2A⇑).
In normal WHHL rabbit aortas, 15-LO mRNA expression was below the detection limit, whereas fatty streaks showed a remarkable induction of 15-LO mRNA expression (Fig 2B⇑). When atherosclerosis had progressed further to more advanced fatty plaques, the content of 15-LO mRNA was 10 times lower than in the fatty streaks.
Verification of 15-LO PCR Product
Although 15-LO PCRs were performed under stringent conditions (Table⇑) and the primers were designed to avoid other known LOs, the identities of PCR amplification products were confirmed by cloning and sequencing. Five insert-containing clones were sequenced from both NZW and WHHL rabbit lesions. They all were identical to the published sequence of rabbit reticulocyte 15-LO cDNA,25 except for a constant G-to-A substitution at nucleotide 607, which thus represents an allelic variation or a true sequencing error in the original sequence.
In Situ Hybridization and Immunocytochemistry
Strong expression of 15-LO mRNA was detected in both WHHL and NZW rabbit atherosclerotic arteries. Fig 3a⇓ shows an example of 15-LO mRNA expression in WHHL rabbit lesions. As a control, a serial section was hybridized with a nonhybridizing 15-LO sense probe (Fig 3b⇓). Immunostainings of serial sections with mAb RAM-11 revealed that most of the cells in 15-LO–expressing areas were macrophages (data not shown). An example of 15-LO expression in early NZW rabbit atherosclerotic lesions is shown in Fig 3c⇓ through 3h: 15-LO protein (Fig 3d⇓) and mRNA (Fig 3e⇓) were expressed in the same area that immunostained for macrophages (Fig 3g⇓). Control hybridizations with the 15-LO sense probe (Fig 3f⇓) and control immunostainings (Fig 3h⇓) were negative. Macrophages expressing 15-LO mRNA were present in the NZW rabbit intima as early as after 3 weeks on the cholesterol-rich diet (data not shown). Under stringent conditions, no 15-LO mRNA or protein was detected in medial smooth muscle cells. However, we cannot fully exclude a possibility for a low-level expression of 15-LO–like molecules in arterial smooth muscle cells.
CuZn-SOD and EC-SOD mRNA Expressions
CuZn-SOD and EC-SOD expressions were analyzed with competitive RT-PCR from the same samples that were used for the 15-LO quantification. There was a clear basal expression of both CuZn-SOD and EC-SOD mRNA in normal aortas. However, no major changes were detected in the expression of these genes with the progression of atherosclerosis (Fig 2⇑).
RT-PCR was chosen for the analysis of arterial wall gene expression instead of Northern blotting because of its better sensitivity. The usefulness of RT-PCR is clearly illustrated by the fact that, on average, ≈240 μg total RNA (or 10 μg poly[A] RNA) is obtained from aortic arch intima-medias of five rabbits, which permits only a few Northern blotting experiments but is still sufficient for more than 10 000 PCRs. Various strategies for semiquantitative PCR can be used. First, PCR can be performed without any added standard, but it is very difficult to standardize the assay without coamplification of a standard molecule. It is also possible to add to the PCR tube another set of primers designed to amplify an unrelated standard gene (such as actin). However, the standard gene is usually expressed in higher amounts, and its amplification has already reached a plateau when the gene of interest is barely detectable. In addition, multiple primer sets in the same tube can interfere with the amplification process.31 We used a competitive PCR technique with added standards that can be amplified in the same tube with the same set of primers as the target cDNA.18 19 20 This approach eliminates most of the problems related to the above-mentioned amplification methods and can be used to study differences in the amounts of gene transcripts between various samples. The standard may have either a homologous or a heterologous sequence between the PCR primers, since amplification efficiency seems to be determined primarily by the primer sequences.19 31
Initially, we aimed to use two genes, actin and GAPDH, for the normalization of the results. However, a twofold to fivefold induction of GAPDH mRNA expression was detected in both rabbit models with the development of atherosclerotic lesions (Fig 2⇑). The increase in GAPDH expression may be due to an increasing number of macrophages in the arterial wall during progression of atherosclerosis, to a change in smooth muscle cells from a contractile to a synthetic phenotype,32 or to the effect of high blood cholesterol content on glycolytic metabolism in the cells. Additionally, nuclear uracil DNA glycosylase is a product of the same gene as GAPDH,33 which brings on another source of variation in GAPDH mRNA expression. These data suggest that GAPDH should not be used for the normalization of mRNA expression at different phases of atherogenesis. Conversely, only random changes were detected in actin expression between the samples. Consequently, actin was chosen as the primary gene for the normalization of the quantification results. Nevertheless, no qualitative changes in the 15-LO quantification results would have been caused by the use of GAPDH as a normalizing gene; only the magnitude of 15-LO induction would have been smaller (Fig 2⇑).
No 15-LO activity measurements were performed in the present study. However, our results are in close agreement with those of Kühn et al,14 who measured 15-LO reaction products in NZW rabbit atherosclerotic arteries using a study design very similar to our study. In their study, the amount of the S-enantiomer of 13-HODE, compared with the R-enantiomer, was slightly increased after 6 weeks and substantially increased after 14 weeks on a 1% cholesterol diet, which indicates increased enzymatic oxidation of linoleic acid. Thus, their results imply an induction of 15-LO expression in early atherosclerotic lesions, which was found in the present study.
Our results show that 15-LO mRNA expression is induced in NZW rabbit aortas after as little as 3 weeks on a cholesterol-rich diet before the appearance of the first macroscopic lesions. At this stage, some macrophages were already present in the arterial wall. The concentration of 15-LO mRNA was highest in NZW rabbit aortas after 14 weeks on the cholesterol-rich diet and in aortic fatty streaks of 3-month-old WHHL rabbits. On the basis of in situ hybridization and immunocytochemical studies, it is likely that most of the 15-LO expression occurs in macrophages. This study confirms previous findings11 12 of the 15-LO mRNA and protein expression in lesion macrophages. When atherosclerosis in WHHL rabbits had further progressed to more advanced fatty plaques, the 15-LO mRNA levels had fallen to levels about 1/10 of the levels found in the fatty streaks (Fig 2⇑). Results of Kühn et al14 also suggest a decreased role for enzymatic (15-LO) oxidation in the later phase of atherogenesis, in which it is more likely that unspecific lipid peroxidation plays an important role. Reduction of 15-LO expression in the advanced lesions may be due to a lower number of macrophages or to a downregulation of 15-LO expression and/or macrophage activity by various cytokines present in atherosclerotic lesions.34 We did not detect any major induction of CuZn-SOD or EC-SOD mRNA in rabbit atherosclerotic aortas. Thus, a definite increase in 15-LO versus SOD mRNA ratio was observed during the early phase of atherogenesis.
Previous reports of 15-LO expression in atherosclerotic lesions were based on qualitative RT-PCR, in situ hybridization, immunoblotting, immunocytochemistry,11 12 and the demonstration of the presence of 15-LO reaction products in atherosclerotic aortas.13 14 In the former studies, leukocyte-type 5-LO and platelet-type 12-LO mRNA could not be detected in the lesion area.11 12 However, a possibility of cross-reaction between members of the same gene family exists. Also, detection of 13-HODE does not unequivocally prove 15-LO function, since leukocyte-type 12-LO converts linoleic acid to 13-HODE.35 36 37 Consequently, we cloned and sequenced the atherosclerosis-associated LO. The results confirmed the induction of the reticulocyte-type 15-LO expression in rabbit atherosclerotic lesions. The present results do not exclude the possibility of an additional expression of a leukocyte-type 12-LO or some unknown LOs in lesion macrophages, endothelial cells, or smooth muscle cells. However, the results suggest that 15-LO expression in macrophages may play an important role in the early phase of atherogenesis and could be a potential target for interventions.
Selected Abbreviations and Acronyms
|CuZn-SOD||=||copper-zinc superoxide dismutase|
|EC-SOD||=||extracellular superoxide dismutase|
|Mn-SOD||=||manganese superoxide dismutase|
|NZW||=||New Zealand White|
|RT-PCR||=||reverse transcription–polymerase chain reaction|
|WHHL||=||Watanabe heritable hyperlipidemic|
This study was supported by grants from The Elli and Elvi Oksanen Fund of the Pirkanmaa Regional Fund under the auspices of the Finnish Cultural Foundation, The Academy of Finland, Sigrid Jusélius Foundation, Finnish Heart Foundation, and Aarne Koskelo Foundation. The authors thank Drs Joseph L. Witztum, Daniel Steinberg, and Elliott Sigal for providing reticulocyte 15-LO antibodies and cDNA, Dr Allen Gown for RAM-11 and HHF-35 antibodies, and Drs Heike Lukhaup and Hans A. Dresel for GAPDH and 15-LO standards. Eija Kyrölä’s expertise in typing the manuscript is acknowledged.
- Received March 7, 1995.
- Revision received July 11, 1995.
- Accepted July 16, 1995.
- Copyright © 1995 by American Heart Association
Witztum JL, Steinberg D. Role of oxidized low density lipoprotein in atherogenesis. J Clin Invest. 1991;88:1785-1792.
Ylä-Herttuala S. Role of lipid and lipoprotein oxidation in the pathogenesis of atherosclerosis. Drugs Today. 1994;30:507-514.
Halliwell B, Gutteridge JMC. Free Radicals in Biology and Medicine. Oxford, England: Clarendon Press; 1989.
Smith C, Mitchinson MJ, Aruoma OI, Halliwell B. Stimulation of lipid peroxidation and hydroxyl-radical generation by the contents of human atherosclerotic lesions. Biochem J. 1992;286:901-905.
Sparrow CP, Parthasarathy S, Steinberg D. Enzymatic modification of low density lipoprotein by purified lipoxygenase and phospholipase A2 mimics cell-mediated oxidative modification. J Lipid Res. 1988;29:745-753.
O’Learly VJ, Darley-Usmar VM, Russell LJ, Stone D. Pro-oxidant effects of lipoxygenase-derived peroxides on the copper-initiated oxidation of low-density lipoprotein. Biochem J. 1992;282:631-634.
Parthasarathy S, Wieland E, Steinberg D. A role for endothelial cell lipoxygenase in the oxidative modification of low density lipoprotein. Proc Natl Acad Sci U S A. 1989;86:1046-1050.
Rankin SM, Parthasarathy S, Steinberg D. Evidence for a dominant role of lipoxygenase(s) in the oxidation of LDL by mouse peritoneal macrophages. J Lipid Res. 1991;32:449-456.
Ylä-Herttuala S, Rosenfeld ME, Parthasarathy S, Glass CK, Sigal E, Witztum JL, Steinberg D. Colocalization of 15-lipoxygenase mRNA and protein with epitopes of oxidized low density lipoprotein in macrophage-rich areas of atherosclerotic lesions. Proc Natl Acad Sci U S A. 1990;87:6959-6963.
Ylä-Herttuala S, Rosenfeld ME, Parthasarathy S, Sigal E, Särkioja T, Witztum JL, Steinberg D. Gene expression in macrophage-rich human atherosclerotic lesions: 15-lipoxygenase and acetyl low density lipoprotein receptor messenger RNA colocalize with oxidation specific lipid-protein adducts. J Clin Invest. 1991;87:1146-1152.
Kühn H, Belkner J, Zaiss S, Fährenklemper T, Wohlfeil S. Involvement of 15-lipoxygenase in early stages of atherogenesis. J Exp Med. 1994;179:1903-1911.
Benz DJ, Mol M, Ezaki M, Mori-Ito N, Zalán I, Miyanohara A, Friedmann T, Parthasarathy S, Steinberg D, Witztum JL. Enhanced levels of lipoperoxides in low density lipoprotein incubated with murine fibroblasts expressing high levels of human 15-lipoxygenase. J Biol Chem. 1995;270:5191-5197.
Ylä-Herttuala S, Luoma J, Viita H, Hiltunen T, Sisto T, Nikkari T. Transfer of 15-lipoxygenase gene into rabbit iliac arteries results in the appearance of oxidation-specific lipid-protein adducts characteristic of oxidized low density lipoprotein. J Clin Invest. 1995;95:2692-2698.
Hjalmarsson K, Marklund SL, Engström Å, Edlund T. Isolation and sequence of complementary DNA encoding human extracellular superoxide dismutase. Proc Natl Acad Sci U S A. 1987;84:6340-6344.
Becker-André M, Hahlbrock K. Absolute mRNA quantification using the polymerase chain reaction (PCR): a novel approach by a PCR aided transcript titration assay (PATTY). Nucleic Acids Res. 1989;17:9437-9446.
Wang AM, Doyle MV, Mark DF: Quantitation of mRNA by the polymerase chain reaction. Proc Natl Acad Sci U S A. 1989;86:9717-9721.
Gilliland G, Perrin S, Blanchard K, Bunn HF. Analysis of cytokine mRNA and DNA: detection and quantitation by competitive polymerase chain reaction. Proc Natl Acad Sci U S A. 1990;87:2725-2729.
Kingston RE. CsCl purification of RNA from tissue. In: Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K, eds. Current Protocols in Molecular Biology. New York, NY: Greene Publishing Assocs and John Wiley & Sons; 1991:4.2.3-4.2.4.
Applequist S, Keyna U, Beck-Engeser G, Raman C, Jack HM. Nucleotide sequence of rabbit glyceraldehyde-3-phosphate dehydrogenase mRNA. EMBL Data Library 1993: Accession No. L23961. European Bioinformatics Institute, Hinxton Genome Campus, Cambridge CB10 1RQ, England. E-mail firstname.lastname@example.org.
Fleming J, Thiele BJ, Chester J, O’Prey J, Janetzki S, Aitken A, Anton IA, Rapoport SM, Harrison PR. The complete sequence of the rabbit erythroid cell-specific 15-lipoxygenase mRNA: comparison of the predicted amino acid sequence of the erythrocyte lipoxygenase with other lipoxygenases. Gene. 1989;79:181-188.
Sherman L, Dafni N, Lieman-Hurwitz J, Groner Y. Nucleotide sequence and expression of human chromosome 21-encoded superoxide dismutase mRNA. Proc Natl Acad Sci U S A. 1983;80:5465-5469.
Jackson RM, Ho YS. Nucleotide and deduced amino acid sequence of rabbit CuZnSOD. EMBL Data Library 1993: Accession No. L12405. European Bioinformatics Institute, Hinxton Genome Campus, Cambridge CB10 1RQ, England. E-mail email@example.com.
Ansorge W, Sproat B, Stegemann J, Schwager C, Zenke M. Automated DNA sequencing: ultrasensitive detection of fluorescent bands during electrophoresis. Nucleic Acids Res. 1987;15:4593-4602.
Tsukada T, Rosenfeld M, Ross R, Gown AM. Immunocytochemical analysis of cellular components in atherosclerotic lesions: use of monoclonal antibodies with the Watanabe and fat-fed rabbit. Arteriosclerosis. 1986;6:601-613.
Meyer-Siegler K, Mauro DJ, Seal G, Wurzer J, deRiel JK, Sirover MA. A human nuclear uracil DNA glycosylase is the 37-kDa subunit of glyceraldehyde-3-phosphate dehydrogenase. Proc Natl Acad Sci U S A. 1991;88:8460-8464.
Conrad DJ, Kühn H, Mulkins M, Highland E, Sigal E. Specific inflammatory cytokines regulate the expression of human monocyte 15-lipoxygenase. Proc Natl Acad Sci U S A. 1992;89:217-221.
Yokoyama C, Shinjo F, Yoshimoto T, Yamamoto S, Oates JA, Brash AR. Arachidonate 12-lipoxygenase purified from porcine leukocytes by immunoaffinity chromatography and its reactivity with hydroperoxyeicosatetraenoic acids. J Biol Chem. 1986;261:16714-16721.
Chen XS, Kurre U, Jenkins NA, Copeland NG, Funk CD. cDNA cloning, expression, mutagenesis of C-terminal isoleucine, genomic structure, and chromosomal localizations of murine 12-lipoxygenases. J Biol Chem. 1994;269:13979-13987.