Dominant Negative Effect of TGF-β1 and TNF-α on Basal and IL-6–Induced Lipoprotein(a) and Apolipoprotein(a) mRNA Expression in Primary Monkey Hepatocyte Cultures
Abstract—Lipoprotein(a) [Lp(a)] consists of apolipoprotein(a) [apo(a)] disulfide linked to apolipoprotein B-100 of LDL. Elevated plasma Lp(a) is an independent risk factor for a variety of vascular diseases. Lp(a) has been reported to be an acute-phase reactant, suggesting that cytokines may regulate its levels. To determine whether Lp(a) expression was subject to modulation by cytokines, primary monkey hepatocytes that endogenously express Lp(a) were used. Hepatocytes were treated with interleukin (IL)-6, the major mediator of the acute-phase response, and several other cytokines. IL-6 treatment (0.3 to 10 ng/mL) resulted in a marked, dose-dependent, 2- to 4-fold enhancement of Lp(a) accumulation in the hepatocyte culture media that was highly correlated with changes in apo(a) mRNA levels (r>0.9). Several other cytokines, such as IL-2, IL-8, and hepatocyte growth factor, had no significant effect on Lp(a) levels; however, transforming growth factor-β1 (TGF-β1) and tumor necrosis factor-α (TNF-α) were very active in inhibiting Lp(a) accumulation in the culture media, with IC50s of ≈0.3 and 1 ng/mL, respectively. Both TGF-β1 and TNF-α also decreased the apo(a) transcript. Mixing experiments, in which hepatocytes were treated with 10 ng/mL of IL-6 and 0.3 to 10 ng/mL of TGF-β1 or TNF-α, demonstrated that the IL-6–mediated induction of Lp(a) and apo(a) mRNA was ablated with very low levels of either inhibitory cytokine, suggesting a dominant negative effect of TGF-β1 and TNF-α. These results show that Lp(a) and apo(a) mRNA expression in primary monkey hepatocytes is subject to both positive (IL-6) and negative (TGF-β1 and TNF-α) regulation by physiological levels of cytokines. Thus, in vivo Lp(a) levels may be dependent on the balance between stimulatory and inhibitory cytokines.
- Received October 17, 1997.
- Accepted January 15, 1998.
Elevated plasma Lp(a) is an independent risk factor for the development of cardiovascular and cerebrovascular diseases.1 The Lp(a) particle is formed through the extracellular association of apoB-100–containing lipoproteins (eg, LDL) to apo(a) through a single disulfide bond.2 3 4 5 6 7 Apo(a) is highly variable in size and is thought to have arisen by multiple duplications of the plasminogen gene.8
The concentration of plasma Lp(a) is governed by its production rate and not its catabolism.9 Because apo(a) levels are limiting relative to those of apoB-100, Lp(a) generation is dependent on the amount of apo(a) available for coupling to apoB-100. Apo(a) is synthesized predominately by the liver10 11 12 ; therefore, characterization of factors that modulate hepatic apo(a) expression would lead to a better understanding of Lp(a) regulation.
Plasma Lp(a) levels increase after surgery and myocardial infarction.13 This change in Lp(a) is associated with elevations in several established acute-phase proteins, leading to the suggestion that Lp(a) is an acute-phase reactant and consequently responsive to cytokines such as IL-6.14 The latter observation is consistent with the presence of 6 IL-6 cis-acting REs in a 1.4-kb fragment of the human apo(a) gene 5′-flanking/promoter region.15 16 Of these 6 IL-6 REs, 5 are identical in the analogous region of the monkey apo(a) gene.17 Monkey IL-6 RE at position −88917 differs from the corresponding human IL-6 RE (−881)15 by containing A instead of G in the third position of the IL-6 RE consensus sequence 5′-CTGGGA-′3.14 Conservation of these IL-6 REs and the 88% nucleotide sequence identity between human and monkey apo(a) 5′-flanking fragments17 imply that important functions of this promoter region were maintained during primate evolution. This high degree of conservation between human and monkey apo(a) promoters suggests that the monkey is a suitable model system to investigate the influence of cytokines on endogenous Lp(a) and apo(a) mRNA expression.
In this study, using primary monkey hepatocyte cultures, we found that the major mediator of the acute-phase response, IL-6, significantly induced Lp(a)/apo(a) mRNA levels. In contrast, TGF-β1 and TNF-α had a dominant negative effect on basal and IL-6–induced Lp(a)/apo(a) mRNA expression.
Cynomolgus Monkey Hepatocyte Cell Culture
Primary liver hepatocytes (parenchymal cells) from healthy, adult, male cynomolgus monkeys were isolated and cultured exactly as described.18 In brief, livers were perfused for 13 minutes with a preperfusion buffer consisting of 0.149 mol/L NaCl, 10 mmol/L HEPES, pH 7.5, 16.6 mmol/L fructose, 0.5 mmol/L EGTA, 10 U/mL heparin, and 0.0003% phenol red. This preperfusion was followed by 40 minutes of perfusion with a digestion buffer consisting of 0.142 mol/L NaCl, 0.67 mol/L KCl, 10 mmol/L HEPES, pH 7.5, 5.0 mmol/L CaCl2, 16.6 mmol/L fructose, 0.2% BSA, 0.0003% phenol red, 20 U/mL collagenase (Worthington), 147 U/mL hyaluronidase (Sigma Chemical Co), and 160 U/mL trypsin inhibitor (Sigma). After digestion, the liver was transferred to a sterile, stainless steel pan, coarsely chopped, and gently shaken in a suspension buffer consisting of Dulbecco’s modified Eagle’s medium, 0.2% BSA, 5% FBS, 2.11 mmol/L fructose, and 11.0 mmol/L insulin (Sigma) to release the cells. The freed cells were passed through a series of three sieves of decreasing mesh size and were washed by pelleting the cells at 300 rpm in an IEC PR-6000 centrifuge followed by resuspension in suspension buffer without FBS. The washing step was repeated twice more. Washed cells were mixed with Percoll (Pharmacia) in 1× Hanks’ buffered saline solution (Life Technologies) and pelleted by centrifugation. The resulting hepatocyte pellet was washed twice in Dulbecco’s modified Eagle’s medium as described above and resuspended in the medium. Cell number was determined by mixing 10 μL of the cell suspension with an equal volume of 0.4% trypan blue, and the cells were counted on a hemacytometer and by phase-contrast microscopy. Hepatocytes were plated and cultured overnight in 6-well, collagen-coated plates (Collaborative Biomedical) at a density of 106 cells per well in Dulbecco’s modified Eagle’s medium, 10% FBS, and previously described supplements.18 It should be noted that, though unlikely, 10% FBS treatment could result in preexposure of the hepatocytes to cytokines contained in the serum. The next morning, the medium was changed to fresh medium not containing FBS. Cells were cultured under these conditions for a total of 2 days. The hepatocytes were then dosed with cytokines at the indicated concentrations for 3 days, with daily changes of the medium. Quadruplicate wells were used for each cytokine dose. Conditioned medium was collected after the last day of cytokine treatment, frozen at −80°C, and assayed for Lp(a) within 1 week by ELISA. Lp(a) levels were standardized for cell number. Total RNA was isolated by directly adding RNAzol (Biotecx) to the culture plates.
Human IL-1α, IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-11, IL-12, IL-13, hepatocyte growth factor, TGF-β1, and TNF-α were reconstituted according to the manufacturer (R & D Systems), stored at −80°C, and diluted as described below.
Lipoprotein and Apolipoprotein Analysis
Lp(a) total mass in the culture medium was measured by a commercial ELISA (ApoTek, PerImmune). The ELISA can measure monkey Lp(a), is independent of apo(a) isoform size, and does not cross-react with LDL or plasminogen.19 This ELISA measures only Lp(a) and not free apo(a). Because pooled human plasma was used as the standard in the ELISA, the monkey Lp(a) values do not represent absolute values. Lp(a) Western blot analyses were performed as previously described12 and developed by using a goat anti-human Lp(a) polyclonal antibody that was preabsorbed to LDL and plasminogen (Biodesign). Control experiments showed that the Lp(a) antibody did not cross-react with apoB-100 or plasminogen on Western blots. No bands were apparent when the Lp(a) antibody was omitted from the Western blot development.
A solution hybridization assay was used to measure cynomolgus monkey apo(a), plasminogen, and G3PDH mRNAs with riboprobes that were generated exactly as described.12 In brief, ≈100 pg of 32P-labeled anti-sense probe, 10 pg of cold sense-strand internal standard, and total RNA (3 to 10 μg) or yeast tRNA (50 μg) were added to a 0.5-mL GeneAmp thin-walled reaction tube (Perkin-Elmer) and dried in a SpeedVac. The pellet was completely dissolved in 4 μL of buffer consisting of 0.4 mol/L NaCl, 1 mmol/L EDTA, and 40 mmol/L MOPS, pH 7.0, followed by addition of 16 μL of deionized formamide. The hybridization solution was thoroughly mixed, placed in a Perkin-Elmer 9600 thermocycler, heated at 85°C for 15 minutes, and hybridized for 16 hours at 45°C. Tubes were removed from the thermocycler, and 280 μL of a solution containing 1.5 mol/L NaCl, 25 mmol/L EDTA, 50 mmol/L Tris HCl, pH 7.5, 0.05 mg/mL ribonuclease AI, and 0.2 U of ribonuclease TI were immediately added and incubated at 30°C for 1 hour. After the ribonuclease treatment, samples were adjusted to 0.7% SDS and 0.17 mg/mL proteinase K and incubated for 15 minutes at 37°C, followed by extraction with an equal volume of phenol/chloroform/isoamyl alcohol, 25:24:1, vol/vol/vol. The upper phase was removed to a clean tube, 20 μg of yeast tRNA was added, and the mixture was precipitated with 2.5 volumes of absolute ethanol. Tubes were centrifuged at full speed in a Microfuge (Eppendorf); the pellet was then washed once with 75% ethanol, thoroughly dissolved in 4 μL loading buffer (90% deinoized formamide, 20 mmol/L EDTA, pH 8.0, 0.02% bromophenol blue, and 0.02% xylene cyanol), and heated at 70°C for 10 minutes. The heated samples were immediately loaded onto 6% sequencing gels, and electrophoresis was carried out at 80 W until the bromophenol blue migrated to the bottom of the gel. Gels were transferred to exposed x-ray film, covered with plastic wrap, and exposed to a storage PhosphorImager screen (Molecular Dynamics) overnight. Exposed screens were scanned with a PhosphorImager, and the resulting bands were quantified by using ImageQuant software (Molecular Dynamics).
The modulating effect of IL-6 on Lp(a) levels was determined on hepatocytes isolated from two healthy cynomolgus monkeys expressing plasma Lp(a) levels of 6 (monkey 93–313) and 30 (monkey 93–303) mg/dL. IL-6 treatment (0.3 to 10 ng/mL) of the hepatocyte cultures resulted in a marked dose-dependent increase in Lp(a) levels in the culture media (Figure 1⇓). At 10 ng/mL of IL-6, hepatocytes from monkeys expressing plasma Lp(a) levels of 6 mg/dL and 30 mg/dL responded with a >250% and 425% increase in Lp(a), respectively. No further increases in Lp(a) were observed at IL-6 concentrations >10 ng/mL (data not shown). Lp(a) levels were elevated in the hepatocyte culture media with as little as 0.3 ng/mL of IL-6 (Figure 1⇓). Western blot analysis of the hepatocyte culture media demonstrated that both Lp(a) and apo(a) were increased by IL-6 treatment (Figure 2⇓). This suggested that the IL-6–mediated changes in Lp(a) were due to elevations in apo(a).
To determine whether the IL-6–mediated increase in Lp(a)/apo(a) was due to changes in apo(a) mRNA, the transcript was measured by using a specific ribonuclease protection assay (Figure 3⇓) in the IL-6–treated hepatocytes described above. These results indicated that the apo(a) mRNA levels were increased by IL-6 treatment in a dose-dependent fashion (Figure 1⇑). Treatment with 10 ng/mL of IL-6 resulted in a 275% and 425% increase in the apo(a) transcript levels in monkey 93–313 and monkey 93–303, respectively. Significant increases in apo(a) mRNA were observed with as little as 0.3 ng/mL of IL-6. The similarity in the IL-6–mediated induction of Lp(a) and apo(a) mRNA suggested that the levels were closely related. Indeed, regression analysis of Lp(a) and apo(a) mRNA levels in Figure 1⇑ revealed r values of 0.957 and 0.927 for monkey 93–313 and monkey 93–303, respectively. These results indicate that IL-6 increased Lp(a) levels by augmenting the apo(a) transcript.
Plasminogen and G3PDH mRNAs were measure by ribonuclease protection assays in the same RNA samples described above for monkey 93-303 (Figure 3⇑). These results demonstrated that IL-6 increased plasminogen mRNA a maximum of only 160% (Figure 4⇓), which was 2.6-fold lower than the maximum 425% increase observed for apo(a) mRNA. These results are consistent with the fact that the plasminogen promoter contains only two IL-6 cis-acting REs, whereas the apo(a) promoter has 6 of the corresponding elements.15 A similar pattern of IL-6 (10 ng/mL)–mediated changes in apo(a) and plasminogen mRNAs, of 275% and 148%, respectively, was observed in hepatocytes from the second monkey. The control G3PDH mRNA was unchanged by IL-6 treatment, indicating that the cytokine was not having a generalized effect (Figure 4⇓).
The responsiveness of Lp(a) and apo(a) mRNA to IL-6 prompted us to examine whether several other cytokines listed in the “Methods” section, were active modulators of their expression. None of these cytokines were capable of increasing Lp(a) levels at a concentration of 10 ng/mL. In contrast, IL-1α/β, IL-4, IL-13, TGF-β1, and TNF-α inhibited Lp(a) by 50% or greater at 10 ng/mL. Of these inhibitory cytokines, TGF-β1 and TNF-α were the most potent, decreasing Lp(a) levels to ≈10% of control; therefore, these two cytokines were investigated further.
TGF-β1 and TNF-α significantly inhibited Lp(a) levels in the hepatocyte cultures in a dose-dependent manner (Figure 5A⇓ and 5B⇓). These inhibitory cytokines effectively attenuated Lp(a) levels at concentrations as low as 0.3 ng/mL (Figure 5A⇓ and 5B⇓), with IC50s of ≈0.3 ng/mL and ≈1 ng/mL for TGF-β1 and TNF-α, respectively. As observed for IL-6, the modulation of Lp(a) levels by TGF-β1 and TNF-α was correlated with changes in apo(a) mRNA. However, unlike IL-6, the inhibitory cytokines decrease Lp(a) to a greater extent than the apo(a) mRNA. For example, 3 ng/mL of TGF-β1 caused a >90% suppression of Lp(a) while decreasing the apo(a) transcript by only 45%. Western bot analysis showed that the inhibitory cytokines decreased Lp(a) by lowering the amount of free apo(a) protein (data not shown). Therefore, besides modulating apo(a) transcription, the inhibitory cytokines may also be altering apo(a) at the posttranscriptional level. Apo(a) translation, processing, secretion, and release from the hepatocyte surface6 could potentially be modulated by TGF-β1 and TNF-α.
TGF-β1 and TNF-α decreased plasminogen mRNA by the same magnitude as observed for apo(a) mRNA (Figure 6A⇓). The level of G3PDH mRNA was unchanged by either TGF-β1 or TNF-α (Figure 6B⇓). Consistent with the latter observation, the hepatocytes showed no visible signs of toxicity when treated with up to 10 ng/mL of TGF-β1 or TNF-α.
The previous experiments indicated that Lp(a) expression was subject to both positive and negative cytokine regulation. To determine which of the two pathways played the dominate role in determining Lp(a) levels, competition experiments between IL-6 and TGF-β1 or TNF-α were carried out. Cells were treated with IL-6 (10 ng/mL) alone or with IL-6 (10 ng/mL) plus increasing concentrations of TGF-β1 or TNF-α (0.3 to 10 ng/mL). In this experiment, IL-6 alone increased Lp(a) levels to ≈350% above control values (Figure 7⇓). However, when the cells were dosed with a combination of 10 ng/mL IL-6 and 0.3 ng/mL TGF-β1, Lp(a) levels were elevated to only 150% above control levels (Figure 7A⇓). Increasing the TGF-β1 concentration to 1 ng/mL completely attenuated the IL-6 induction and, in addition, reduced Lp(a) levels to 40% of control values. Maximum inhibition of Lp(a) levels to 12% of control levels was seen at 3 ng/mL TGF-β1. Similarly, TNF-α also prevented the IL-6–mediated induction of Lp(a); however, it was less potent than TGF-β1 (Figure 7B⇓). At 0.3 ng/mL TNF-α plus 10 ng/mL IL-6, Lp(a) levels were increased to 250% of control levels compared with 350% with 10 ng/mL IL-6 alone (Figure 7B⇓). More notable changes were seen at 1 ng/mL TNF-α, which almost totally ablated the IL-6 induction of Lp(a) levels. Further increases of TNF-α to 3 and 10 ng/mL in the presence of 10 ng/mL IL-6 resulted in the inhibition of Lp(a) levels to 40% and 20%, respectively, of control values. In both competition experiments there was an excellent correlation between Lp(a) and apo(a) mRNA levels (Figure 7A⇓ and 7B⇓).
We have demonstrated that Lp(a) and apo(a) mRNA levels in primary monkey hepatocyte cultures are responsive to certain cytokines. This cytokine response could be either positive (IL-6) or negative (TGF-β1 and TNF-α), with the latter playing a dominant regulatory role. These observations reveal novel pathways for regulating Lp(a).
The positive response of Lp(a) and apo(a) mRNA to IL-6 in the monkey hepatocyte cultures is consistent with the presence of functional IL-6 cis-acting REs in the apo(a) promoter.15 Because IL-6 mediates the acute response, Lp(a) can be implicated as a positive acute-phase reactant. This finding is in keeping with a clinical study that demonstrated elevated levels of Lp(a) and acute-phase proteins after myocardial infarction and surgery.13 However, the kinetics of Lp(a) increase in the myocardial infarction group were not characteristic of a typical acute-phase reactant.13 Furthermore, there were subgroups of postsurgery patients whose Lp(a) levels remained elevated for prolonged periods or actually decreased to undetectable levels.13 The nonuniformity of the Lp(a) response in the latter study, together with a clinical investigation that has shown that Lp(a) is not an acute-phase reactant,20 questioned its role in the acute-phase response. In fact, a study using transgenic apo(a) mice implicated apo(a) as a negative acute-phase reactant.21
Based on our findings, the discrepancy in the Lp(a) response in the latter studies could be explained by the balance between stimulatory and inhibitory cytokines. It is possible that Lp(a) is observed to be elevated under conditions where IL-6 levels are much higher than are either TNF-α or TGF-β1 levels. Conversely, lower Lp(a) levels may be observed in situations where the cytokine balance is shifted toward the inhibitory cytokines. The latter may explain why apo(a) displayed characteristics of a negative acute-phase reactant in the apo(a) transgenic mice. To induce an acute-phase response in these animals, turpentine was administered subcutaneously. Although this treatment effectively increases IL-6, it also elevates TNF-α levels.22 Because we have shown that the inhibitory cytokines are effective in ablating the IL-6–induced and basal expression of Lp(a)/apo(a) mRNA, turpentine treatment may have decreased apo(a) levels in the transgenic mice owing to its ability to increase TNF-α levels.
A feedback loop has been described in apo(a) transgenic mice fed a high-fat diet to explain the inverse relationship between TGF-β1 activity and apo(a) accumulation in atherosclerotic lipid lesions.23 In this model, focal apo(a) accumulation in the vessel wall inhibits TGF-β1 activity, which activates smooth muscle cells to accumulate lipids and form lesions. In addition to the latter scenario, our data show that TGF-β1 directly effects hepatic Lp(a) production by decreasing apo(a) transcript and protein levels. Therefore, the relationship between TGF-β1 and Lp(a) is a complex one in which both molecules may alter the activity/level of the other. It is also interesting to note that tamoxifen, which significantly decreases circulating Lp(a)23 24 and liver apo(a) mRNA levels,25 has been reported to elevate TGF-β1 in vivo.26 Taken together, these independent observations provide circumstantial evidence that TGF-β1 may negatively regulate Lp(a) levels in vivo.
This study reveals a novel relationship between cytokines and Lp(a) levels in primary monkey hepatocytes. It will be important to determine whether this relationship is maintained in humans. The potential involvement of cytokines in modulating Lp(a) levels may have consequences for studies designed to characterize individuals at risk for developing vascular diseases due to their baseline Lp(a). These investigations may be beneficial by identifying those individuals whose Lp(a) levels are altered according to transient changes in their stimulatory/inhibitory cytokine profile rather than to unfavorable Lp(a) levels per se.
Selected Abbreviations and Acronyms
|FBS||=||fetal bovine serum|
|G3PDH||=||glyceraldehyde 3-phosphate dehydrogenase|
|TGF||=||transforming growth factor|
|TNF||=||tumor necrosis factor|
Koschinsky ML, Ramharack R. Structure, function and regulation of lipoprotein(a). Curr Pharm Des. 1996;2:121–138.
Chiesa G, Hobbs HH, Koschinsky ML, Lawn RM, Maika SD, Hammer RE. Reconstitution of lipoprotein(a) by infusion of human low density lipoprotein into transgenic mice expressing human apolipoprotein(a). J Biol Chem. 1992;267:24369–24374.
Koschinsky ML, Côté GP, Gabel B, Van der Hoek YY. Identification of the cysteine residue in apolipoprotein(a) that mediates extracellular coupling with apolipoprotein B-100. J Biol Chem. 1993;268:19819–19825.
White AL, Rainwater DL, Lanford RE. Intracellular maturation of apolipoprotein[a] and assembly of lipoprotein[a] in primary baboon hepatocytes. J Lipid Res. 1993;34:509–517.
White AL, Lanford RE. Cell surface assembly of lipoprotein(a) in primary cultures of baboon hepatocytes. J Biol Chem. 1994;269:28716–28723.
Callow MJ, Stoltzfus LJ, Lawn RM, Rubin EM. Expression of human apolipoprotein B and assembly of lipoprotein(a) in transgenic mice. Proc Natl Acad Sci U S A. 1994;91:2130–2134.
Rader DJ, Cain W, Zech LA, Usher D, Brewer HB Jr. Variation in lipoprotein(a) concentrations among individuals with the same apolipoprotein(a) isoform is determined by the rate of lipoprotein(a) production. J Clin Invest. 1993;91:443–447.
Tomlinson JE, McLean JW, Lawn RM. Rhesus monkey apolipoprotein(a): sequence, evolution, and sites of synthesis. J Biol Chem.. 1989;264:5957–5965.
Azrolan N, Gavish D, Breslow JL. Plasma lipoprotein(a) concentration is controlled by apolipoprotein(a) (apo(a)) protein size and the abundance of hepatic apo(a) mRNA in a cynomolgus monkey model. J Biol Chem. 1991;266:13866–13872.
Ramharack R, Spahr MA, Kreick JS, Sekerke CS. Expression of apolipoprotein[a] and plasminogen mRNAs in cynomolgus monkey liver and extrahepatic tissues. J Lipid Res. 1996;37:2029–2040.
Heinrich PC, Castell JV, Andus T. Interleukin-6 and the acute phase response. Biochem J. 1990;265:621–636.
Wade DP, Clarke JG, Lindahl GE, Liu AC, Zysow BR, Meer K, Schwartz K, Lawn RM. 5′ Control regions of the apolipoprotein(a) gene and members of the related plasminogen gene family. Proc Natl Acad Sci U S A. 1993;90:1369–1373.
Malgaretti N, Acquati F, Magnaghi P, Bruno L, Pontoglio M, Rocchi M, Saccone S, Della Valle G, D’Urso M, LePaslier D, Ottolenghi S, Taramelli R. Characterization by yeast artificial chromosome cloning of the linked apolipoprotein(a) and plasminogen genes and identification of the apolipoprotein(a) 5′ flanking region. Proc Natl Acad Sci U S A. 1992;89:11584–11588.
Hixson JE, Jett C, Birnbaum S. Identification of promoter sequences in the 5′ untranslated region of the baboon apolipoprotein[a] gene. J Lipid Res. 1996;37:2324–2331.
Ramharack R, Spahr MS, Hicks GW, Kieft KA, Brammer DW, Minton LL, Newton RS. Gemfibrozil significantly lowers cynomolgus monkey plasma lipoprotein(a)-protein and liver apolipoprotein(a) mRNA levels. J Lipid Res. 1995;36:1294–1304.
Taddei-Peters WC, Butman BT, Jones GR, Venetta TM, Macomber PF, Ransom JH. Quantification of lipoprotein(a) particles containing various apolipoprotein(a) isoforms by a monoclonal anti-apo(a) capture antibody and a polyclonal anti-apolipoprotein B detection antibody sandwich enzyme immunoassay. Clin Chem. 1993;39:1382–1389.
Lawn RM, Pearle AD, Kunz LL, Rubin EM, Reckless J, Metcalfe JC, Grainger DJ. Feedback mechanism of focal vascular lesion formation in transgenic apolipoprotein(a) mice. J Biol Chem. 1996;271:31367–31371.
Shewmon DA, Stock JL, Rosen CJ, Heiniluoma KM, Hogue MM, Morrison A, Doyle EM, Ukena T, Weale V, Baker S. Tamoxifen and estrogen lower circulating lipoprotein(a) concentrations in healthy postmenopausal women. Arterioscler Thromb. 1994;14:1586–1593.
Zysow BR, Kauser K, Lawn RM, Rubanyi GM. Effects of estrus cycle, ovariectomy, and treatment with estrogen, tamoxifen, and progesterone on apolipoprotein(a) gene expression in transgenic mice. Arterioscler Thromb Vasc Biol. 1997;17:1741–1745.
Butta A, MacLennan K, Flanders KC, Sacks NP, Smith I, McKinna A, Dowsett M, Wakefield LM, Sporn MB, Baum M, Colletta AA. Induction of transforming growth factor beta 1 in human breast cancer in vivo following tamoxifen treatment. Cancer Res. 1992;52:4261–4264.