Inhibition of E2F Abrogates the Development of Cardiac Myocyte Hypertrophy

Growth of the post-natal mammalian heart occurs primarily by cardiac myocyte hypertrophy. Previously, we and others have shown that a partial re-activation of the cell cycle machinery occurs in myocytes undergoing hypertrophy such that cells progress through the G1/S transition. In this study, we have examined the regulation of the E2F family of transcription factors that are crucial for the G1/S phase transition during normal cardiac development and the development of myocyte hypertrophy in the rat. Thus, mRNA and protein levels of E2F-1, 3, and 4 and DP-1 and DP-2 were down-regulated during development to undetectable levels in adult myocytes. Interestingly, E2F-5 protein levels were substantially up-regulated during development. In contrast, an induction of E2F-1, 3, and 4 and the DP-1 protein was observed during the development of myocyte hypertrophy in neonatal myocytes treated with serum or phenylephrine, whereas the protein levels of E2F-5 were decreased with serum stimulation. E2F activity, as measured by a cyclin E promoter luciferase assay and E2F-DNA binding activity, increased significantly during the development of hypertrophy with serum and phenylephrine compared with non-stimulated cells. Inhibiting E2F activity with a specific peptide that blocks E2F-DP heterodimerization prevented the induction of hypertrophic markers (atrial natriuretic factor and brain natriuretic peptide) in response to serum and phenylephrine, reduced the increase in myocyte size, and inhibited protein synthesis in stimulated cells. Thus, we have shown that the inhibition of E2F function prevents the development of hypertrophy. Targeting E2F function might be a useful approach for treating diseases that cause pathophysiological hypertrophic growth. Growth of the post-natal mammalian heart occurs primarily by cardiac myocyte hypertrophy. Previously, we and others have shown that a partial re-activation of the cell cycle machinery occurs in myocytes undergoing hypertrophy such that cells progress through the G1/S transition. In this study, we have examined the regulation of the E2F family of transcription factors that are crucial for the G1/S phase transition during normal cardiac development and the development of myocyte hypertrophy in the rat. Thus, mRNA and protein levels of E2F-1, 3, and 4 and DP-1 and DP-2 were down-regulated during development to undetectable levels in adult myocytes. Interestingly, E2F-5 protein levels were substantially up-regulated during development. In contrast, an induction of E2F-1, 3, and 4 and the DP-1 protein was observed during the development of myocyte hypertrophy in neonatal myocytes treated with serum or phenylephrine, whereas the protein levels of E2F-5 were decreased with serum stimulation. E2F activity, as measured by a cyclin E promoter luciferase assay and E2F-DNA binding activity, increased significantly during the development of hypertrophy with serum and phenylephrine compared with non-stimulated cells. Inhibiting E2F activity with a specific peptide that blocks E2F-DP heterodimerization prevented the induction of hypertrophic markers (atrial natriuretic factor and brain natriuretic peptide) in response to serum and phenylephrine, reduced the increase in myocyte size, and inhibited protein synthesis in stimulated cells. Thus, we have shown that the inhibition of E2F function prevents the development of hypertrophy. Targeting E2F function might be a useful approach for treating diseases that cause pathophysiological hypertrophic growth. Shortly after birth, the majority of rat cardiac myocytes lose the ability to undergo cell division such that post-natal growth of the heart occurs primarily by myocyte hypertrophy (1Poolman R.A. Brooks G. J. Mol. Cell. Cardiol. 1998; 30: 2121-2135Abstract Full Text PDF PubMed Scopus (86) Google Scholar, 2Li J.-M. Brooks G. Eur. Heart J. 1999; 20: 406-420Crossref PubMed Scopus (82) Google Scholar). This adaptive physiological growth response enables the heart to grow to an adequate size in order to maintain sufficient cardiac output. However, myocardial injury, e.g. myocardial infarction and/or prolonged periods of chronic stress on the heart (e.g. hypertension), leads to excessive hypertrophic growth, commonly referred to as “decompensated hypertrophy,” which adversely affects cardiac function and accelerates heart failure (3Piano M.R. Bondmass M. Schwertz D.W. Heart Lung. 1998; 27: 3-19Abstract Full Text PDF PubMed Scopus (40) Google Scholar). Clinically, there is a need to develop new therapeutic entities that target specific molecules involved in this detrimental growth, which will ultimately improve the prognosis for patients. Previously, we have reported that a partial re-activation of specific components of the cell cycle machinery occurs during the development of cardiac hypertrophy, leading to transition of myocytes through the G1/S cell cycle checkpoint (2Li J.-M. Brooks G. Eur. Heart J. 1999; 20: 406-420Crossref PubMed Scopus (82) Google Scholar, 4Li J.M. Poolman R.A. Brooks G. Am. J. Physiol. 1998; 275: H814-H822Crossref PubMed Google Scholar). E2F is a family of transcription factors that are known to play a pivotal role in the G1/S phase transition of the cell cycle (5Dyson N. Genes Dev. 1998; 12: 2245-2262Crossref PubMed Scopus (1981) Google Scholar, 6Nevins J.R. Leone G. DeGregori J. Jakoi L. Nevins J.R. J. Cell. Physiol. 1997; 173: 233-236Crossref PubMed Scopus (176) Google Scholar); however, very little is known about the potential role(s) of these transcription factors in cardiac hypertrophy. Structurally and functionally, the E2F transcription factors can be divided into three main categories as follows: (i) E2Fs 1–3, which are involved in proliferation; (ii) E2F-4 and E2F-5 that are thought to play a role in differentiation; and (iii) E2F-6, which is described as a transcriptional repressor (7Trimarchi J.M. Lees J.A. Nat. Rev. Mol. Cell Biol. 2002; 3: 11-20Crossref PubMed Scopus (970) Google Scholar). For full transcriptional activity, the E2F transcription factors heterodimerize with a DP partner protein, of which two mammalian forms have been described, namely DP-1 and DP-2 (8Rogers K.T. Higgins P.D. Milla M.M. Phillips R.S. Horowitz J.M. Rogers K.T. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 7594-7599Crossref PubMed Scopus (38) Google Scholar). The transcriptional activity of E2F members is sterically regulated by the pRb family of pocket proteins that consists of pRb, p107, and p130 (5Dyson N. Genes Dev. 1998; 12: 2245-2262Crossref PubMed Scopus (1981) Google Scholar, 7Trimarchi J.M. Lees J.A. Nat. Rev. Mol. Cell Biol. 2002; 3: 11-20Crossref PubMed Scopus (970) Google Scholar). E2F activation occurs when cyclin D-cyclin-dependent kinase (CDK) 4/6 complexes in the G1 phase of the cell cycle phosphorylate the pRb pocket proteins, which then dissociate from the E2F-DP complex, resulting in E2F-mediated gene transcription and progression to the S phase of the cell cycle (5Dyson N. Genes Dev. 1998; 12: 2245-2262Crossref PubMed Scopus (1981) Google Scholar, 6Nevins J.R. Leone G. DeGregori J. Jakoi L. Nevins J.R. J. Cell. Physiol. 1997; 173: 233-236Crossref PubMed Scopus (176) Google Scholar). Indeed, various studies have shown that E2F is instrumental in cardiac myocyte cell cycle progression. Thus, Flink et al. (11Flink I.L. Oana S. Maitra N. Bahl J.J. Morkin E. J. Mol. Cell. Cardiol. 1998; 30: 563-578Abstract Full Text PDF PubMed Scopus (66) Google Scholar) showed that a switch occurs from the E2F-p107 complex in proliferating fetal myocytes to E2F-p130 in 2-day-old neonatal cells. Also, von Harsdorf et al. (13von Harsdorf R. Hauck L. Mehrhof F. Wegenka U. Cardoso M.C. Dietz R. Circ. Res. 1999; 85: 128-136Crossref PubMed Scopus (78) Google Scholar) showed that adenoviral delivery of E2F-1 drove a significant number of adult cardiac myocytes into the S phase of the cell cycle; Agah et al. have reported similar findings (12Agah R. Kirshenbaum L.A. Abdellatif M. Truong L.D. Chakraborty S. Michael L.H. Schneider M.D. J. Clin. Invest. 1997; 100: 2722-2728Crossref PubMed Scopus (182) Google Scholar). Clinically, E2F has proved to be a viable target in the treatment of certain vasculoproliferative diseases. Thus, Mann et al. (10Mann M.J. Whittemore A.D. Donaldson M.C. Belkin M. Conte M.S. Polak J.F. Orav E.J. Ehsan A. Dell'Acqua G. Dzau V.J. Mann M.J. Lancet. 1999; 354: 1493-1498Abstract Full Text Full Text PDF PubMed Scopus (396) Google Scholar) showed that blocking E2F-1 function in human vein grafts ex vivo, using a decoy oligonucleotide strategy, reduced the incidence of neointimal hyperplasia and subsequent graft failure. The importance of E2F in the heart was shown recently by Cloud et al. (14Cloud J.E. Rogers C. Reza T.L. Ziebold U. Stone J.R. Picard M.H. Caron A.M. Bronson R.T. Lees J.A. Mol. Cell. Biol. 2002; 22: 2663-2672Crossref PubMed Scopus (74) Google Scholar), who reported that E2F-3a and E2F-3b null mice die from symptoms of congestive heart failure during development, indicating a crucial role for E2F3 in normal cardiac growth. Although much is known about E2F function in a variety of diseases, currently no study has characterized expression and/or inhibited E2F function during the development of cardiac myocyte hypertrophy. In the present study, we have determined the expressions of E2F and DP family members at the mRNA and protein levels in myocytes obtained from the developing rat heart. In addition, we have determined the expressions and activities of these molecules during the development of myocyte hypertrophy in vitro and shown that blocking E2F function with a peptide that abrogates E2F-DP heterodimerization (15Bandara L.R. Girling R. La Thangue N.B. Nat. Biotechnol. 1997; 15: 896-901Crossref PubMed Scopus (36) Google Scholar) inhibits the development of hypertrophy. Our results suggest that E2F plays an important role in cardiac myocyte growth and that targeting these transcription factors could provide us with an effective therapy for treating detrimental left ventricular hypertrophy (LVH) leading to heart failure. Myocyte Isolation—Fetal Wistar rat cardiac myocytes were isolated from 70 fetuses at 18 days of gestation (E18), and neonatal rat myocytes were isolated from 30 Wistar pups (2–3 days post-natal (P2)), according to the method described by Poolman and Brooks (1Poolman R.A. Brooks G. J. Mol. Cell. Cardiol. 1998; 30: 2121-2135Abstract Full Text PDF PubMed Scopus (86) Google Scholar) with minor modifications, including enzymatic digestions for 10 min, preplating for 1 h, and centrifugation and re-suspension of cells in myocyte media (one part M199, four parts Dulbecco's modified Eagle's medium, and 5% fetal calf serum (FCS) 1The abbreviations used are: FCS, fetal calf serum; BrdUrd, bromodeoxyuridine; RT, reverse transcription; ANF, atrial natriuretic factor; BNP, brain natriuretic peptide; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; EMSA, electromobility shift assay; PE, phenylephrine; AH2, peptide sequence targeted to the DEF box region of the E2F heterodimerization domain; AHS2, scrambled sequence based upon AH2; ANT, penetratin sequence. with 100 IU of penicillin/streptomycin) containing 500 μm 5-bromo-3-deoxyuridine (BrdUrd) prior to culture on Primaria plates (Falcon). Adult myocytes were isolated as described by Brooks et al. (16Brooks G. Poolman R.A. McGill C.J. Li J.M. J. Mol. Cell. Cardiol. 1997; 28: 2261-2271Abstract Full Text PDF Scopus (70) Google Scholar), except that 5 mg of protease (Sigma) and 50 mg of bovine serum albumin (Sigma) were added to the digestion mixture. Peptide Studies—E2F blocking and were used at of 30 μm in myocyte media as described by et al. (15Bandara L.R. Girling R. La Thangue N.B. Nat. Biotechnol. 1997; 15: 896-901Crossref PubMed Scopus (36) Google Scholar). Peptide were as follows: AH2, AHS2, and ANT, was as described by Poolman and Brooks (1Poolman R.A. Brooks G. J. Mol. Cell. Cardiol. 1998; 30: 2121-2135Abstract Full Text PDF PubMed Scopus (86) Google Scholar), except that was added to myocyte media at a of 10 Also, was when myocytes were to and cells were for 10 the was modified by with of and and adult myocytes were in mg of was then from the in with the was on 5 of using and was using the E2F-1, GAPDH, ANF, BNP, were on were with and was using and a sequence was using and and using an was using the as by the was from isolated and adult rat ventricular myocytes as described by Brooks and (16Brooks G. Poolman R.A. McGill C.J. Li J.M. J. Mol. Cell. Cardiol. 1997; 28: 2261-2271Abstract Full Text PDF Scopus (70) Google Scholar) except that and were used as protease in proteins were to with for at E2F-1 E2F-3a E2F-4 E2F-5 DP-1 or DP-2 and were used at a of in containing were then in or for 1 were in for 10 min, except the which was myocytes were in culture at and left in 5% myocyte containing 500 μm then were with of a cyclin E luciferase or the cyclin E and of the luciferase using were for and then treated with various hypertrophic for h, after which cells were and using a assay were according to La Thangue et al. Thangue Res. PubMed Scopus Google Scholar), except that was used as the The E2F binding sequence used was Myocyte were used in and studies were and were and using the was to of and was using the of were to be of E2F and DP in 1 mRNA and protein expressions of E2F and DP family members during development in fetal (E18), neonatal and adult rat cardiac myocytes. The mRNA expressions of E2F family members were down-regulated during normal cardiac myocyte development, E2F-5 levels in adult cells compared with family members similar of regulation was at the protein such that E2F-1, and and DP-1 and DP-2 were down-regulated in ventricular and in isolated cardiac myocytes Interestingly, levels of E2F-5 protein were to increase during development levels of in ventricular and cardiac myocytes were very and were when the were a present in This is with the of et al. M.M. J. Dev. Biol. Google Scholar) in the that expression of protein was in cardiac compared with the expressions of E2F of E2F and DP during the of Myocyte that E2F and DP family members are during normal cardiac development, we determined the regulation of these molecules was during the development of myocyte hypertrophy. an in vitro of hypertrophy, cardiac myocytes were stimulated for with of two hypertrophic or 100 μm phenylephrine that hypertrophy been we measured the induction of the hypertrophic ANF, by cell cycle progression by and cell size that was significantly in cells treated for with and compared with the levels in cells in was on and treated cell using the method to the cell cycle of these cells. that of the cells in were in G1 with 5% in the S However, upon with for h, the S phase of cells increased to with a of cells in the G1 increased the S phase by with a of cells in the G1 Thus, a significant number of cells treated with hypertrophic progress from G1 into S with a partial re-activation of the cell cycle during the development of hypertrophy J.M. Poolman R.A. Brooks G. Am. J. Physiol. 1998; 275: H814-H822Crossref PubMed Google Scholar). that cell size was increased significantly in treated with or 100 μm PE, compared with cells these that a significant of hypertrophy was in cardiac myocytes stimulated with determined the expression of E2F and DP family members during the development of myocyte hypertrophy. on protein expression of E2F and DP family that the E2F-1 protein was up-regulated with for h, whereas 100 μm E2F-1 protein levels when compared with similar was observed E2F-3a and and DP-1 proteins were up-regulated with or expression of DP-2 in stimulated and non-stimulated cells was very using and DP-2 This which showed that isolated neonatal myocytes of DP-2 Interestingly, E2F-5 protein was regulated in cells stimulated with at whereas levels were down-regulated significantly at and In with the protein was in stimulated or non-stimulated cells. This of regulation for E2F family members and/or a of expression during the development of hypertrophy. in E2F in during the of studies showed that the E2F and DP proteins are regulated during the development of cardiac myocyte hypertrophy, we determined the transcriptional activities and binding activities of these proteins was during this by using a of luciferase and measured E2F activity in myocytes by a containing the luciferase gene transcriptional of the mammalian cyclin E promoter into neonatal myocytes. containing the luciferase gene transcriptional of the was into myocytes to provide a which could be used to E2F activity and provide an of E2F-mediated luciferase that E2F activity was when myocytes were stimulated to undergo hypertrophy with FCS, whereas increased E2F activity compared with levels in cells that E2F-DNA binding was increased treatment with hypertrophic by a of The results of these studies from the luciferase and showed an in E2F-DNA binding of myocytes with or 100 μm in studies were with an E2F and to and E2F were in namely that the E2F-DP and that E2F-DP in complex with pocket in which was in was observed the E2F the of E2F activity is increased during the development of cardiac myocyte hypertrophy we determined E2F activity this E2F-DP heterodimerization is for transcriptional E2F activity, and function can be by this Indeed, a specific peptide sequence targeted to the DEF box region of the heterodimerization of has been shown to the growth of cells with inhibition of E2F activity (15Bandara L.R. Girling R. La Thangue N.B. Nat. Biotechnol. 1997; 15: 896-901Crossref PubMed Scopus (36) Google Scholar). have used this peptide sequence to E2F function in myocytes stimulated to undergo hypertrophy with or 100 μm were to the penetratin sequence to delivery into cells G. A. J. Biol. Full Text PDF PubMed Google Scholar). The penetratin sequence to as and a scrambled sequence based upon were as for this study E2F a of and we determined the of blocking E2F function on the development of hypertrophy. that myocytes stimulated to undergo hypertrophy in the of peptide or to the and showed a significant induction of and BNP, that hypertrophic growth However, cells stimulated to undergo hypertrophy with in the of showed a in and levels compared with cells treated with This was in cells the induction of and showed serum the of inhibited induction levels were decreased when compared with normal cells and with serum and and to that with the induction was reduced in cells when compared with cells or cells in the of and protein of myocytes was compared in myocytes to or for in the or of E2F Thus, cells stimulated to undergo hypertrophy with protein a increase in cell and cells showed an of protein indicating that the of a peptide in the myocyte culture adversely the development of hypertrophy. In contrast, the of the increase in protein in myocytes treated with such that levels similar to non-stimulated cells. was no significant in cell number treatment of myocytes with for h, that the of this peptide was to hypertrophic growth. were observed for cells substantially the increase in protein that hypertrophic growth The and ANT, no such This that blocking E2F function inhibits the protein synthesis that is known to during cardiac myocyte hypertrophy. In addition, myocytes increase in size with FCS, whereas cells treated with or or peptide in size E2F-DP heterodimerization prevents the increase in protein with hypertrophic growth. were stimulated to undergo hypertrophy with or 100 μm in the or of AH2, AHS2, or μm compared with cells protein was determined by assay obtained from three In this study, we have shown for the that E2F and DP family members are regulated during normal rat cardiac development and during the development of hypertrophy in neonatal rat myocytes. has been shown that neonatal rat myocytes the cell cycle days and 4 after and in the of the cell cycle (1Poolman R.A. Brooks G. J. Mol. Cell. Cardiol. 1998; 30: 2121-2135Abstract Full Text PDF PubMed Scopus (86) Google Scholar, F. J.M. A.M. J. Mol. Cell. Cardiol. 1996; 28: Full Text PDF PubMed Scopus Google Scholar). from has shown that certain cell cycle molecules that of E2F are down-regulated during normal development are during adaptive hypertrophic growth (2Li J.-M. Brooks G. Eur. Heart J. 1999; 20: 406-420Crossref PubMed Scopus (82) Google Scholar, 4Li J.M. Poolman R.A. Brooks G. Am. J. Physiol. 1998; 275: H814-H822Crossref PubMed Google Scholar, G. Poolman R.A. McGill C.J. Li J.M. J. Mol. Cell. Cardiol. 1997; 28: 2261-2271Abstract Full Text PDF Scopus (70) Google Scholar). Our that E2F the of and DP family members are down-regulated in the cardiac myocyte at the mRNA and protein levels from the fetal to adult of ventricular development. In contrast, E2F-5 protein expression was increased development. This expression for E2F-5 is with a potential role for this cell cycle in myocyte Indeed, in which has a E2F-5 significant levels of expression through to M.M. Res. PubMed Scopus Google Scholar). E2F-5 mRNA levels were down-regulated during development, E2F-5 expression is to be at the in adult cardiac myocytes. an in vitro of myocyte hypertrophy, we determined the expression of E2F and DP family members during this Interestingly, a expression of E2F and DP members was observed in myocytes undergoing hypertrophic growth such that there was a the fetal of protein expressions with the in specific and cyclin D-cyclin-dependent during the development of left ventricular hypertrophy J.M. Poolman R.A. Brooks G. Am. J. Physiol. 1998; 275: H814-H822Crossref PubMed Google Scholar). E2F-1 expression levels were significantly when myocytes were treated with PE, a significant was with stimulation. Previously, E2F-1 has been in cardiac myocyte cell cycle progression the E2F-1 gene a G1/S phase transition (13von Harsdorf R. Hauck L. Mehrhof F. Wegenka U. Cardoso M.C. Dietz R. Circ. Res. 1999; 85: 128-136Crossref PubMed Scopus (78) Google Scholar). However, cells progress the phase of the cell The for the regulation of E2F-1 with serum compared with might in growth for the for that the hypertrophic Our studies that E2F-3a to a are the E2F up-regulated and with hypertrophic an important role for these transcription factors in the hypertrophic growth of myocytes. is during the development of hypertrophy, cardiac myocytes a expression of E2F as described by et al. for human cells Genes Dev. PubMed Google Scholar). In this study, cells were into the phase of the cell cycle by serum and then stimulated to the cell cycle by subsequent serum stimulation. The results showed that E2F-1 promoter activity was at and then with a leading role in the G1 phase when to for cyclin p107, and E2F-1, of which are for S phase progression. E2F-5 activity was in these as a of the that were a proliferating is that a similar in cardiac myocytes such that E2F-1 activity could be on in hypertrophy and down-regulated to E2F-3a to cell cycle progression and growth of stimulated myocytes. E2F-4 is to be important for E2F-5 (7Trimarchi J.M. Lees J.A. Nat. Rev. Mol. Cell Biol. 2002; 3: 11-20Crossref PubMed Scopus (970) Google Scholar). However, we observed a significant of the E2F-4 protein in myocytes undergoing hypertrophy with that E2F-4 be in with E2F-1 and to hypertrophy or that could be from the hypertrophic stimulation. Indeed, the E2Fs and E2F-4 and E2F-5 (7Trimarchi J.M. Lees J.A. Nat. Rev. Mol. Cell Biol. 2002; 3: 11-20Crossref PubMed Scopus (970) Google Scholar). is that E2F-4 is from the upon hypertrophic of the cardiac In to the expression levels of the E2F proteins, we the activity of these transcription a cyclin E myocytes stimulated with showed a increase in E2F activity, whereas cells stimulated with showed a increase in E2F activity when compared with cells results suggest that E2Fs play an important role in the development of hypertrophy growth or stimulation. these findings and showed an increase in E2F-DNA binding during the development of hypertrophy with serum and Inhibiting E2F activity with a specific peptide sequence that inhibits E2F-DP heterodimerization to a significant inhibition in the development of myocyte hypertrophy as determined by a of and mRNA expressions and protein synthesis and showed that blocking E2F function to significant inhibition of hypertrophic growth, as levels of and were reduced to in non-stimulated myocytes. Interestingly, showed that blocking E2F activity in cells with has a and induction were reduced to undetectable levels This is with luciferase assay results that hypertrophic growth is E2F function in cells prevented protein which in cells to and in cells treated with and the or have shown that targeting E2F can the growth of cells. Thus, et al. R. M. M. L. Dzau V.J. Proc. Natl. Acad. Sci. U. S. A. PubMed Scopus Google Scholar) showed that an E2F decoy oligonucleotide into rat cells decreased of these cells and decreased the expression of and the proliferating cell Mann et al. (10Mann M.J. Whittemore A.D. Donaldson M.C. Belkin M. Conte M.S. Polak J.F. Orav E.J. Ehsan A. Dell'Acqua G. Dzau V.J. Mann M.J. Lancet. 1999; 354: 1493-1498Abstract Full Text Full Text PDF PubMed Scopus (396) Google Scholar) used a similar approach to neointimal hyperplasia in vein a that to vein graft failure. an ex delivery of an E2F decoy oligonucleotide into a of was and graft failure to compared with for the blocking E2F function in the can provide significant for patients. In the present study, we have shown that blocking E2F function in cardiac myocytes provide similar for with hypertrophic cardiac In we have shown that E2F and DP family members are regulated during normal rat cardiac myocyte development such that transcription with the of are down-regulated as the heart This of expression is during the development of myocyte hypertrophy, E2F-1, and and DP-1 are whereas E2F-5 is down-regulated stimulation. E2F activity is increased during the development of hypertrophy, and blocking E2F function inhibits the development of hypertrophy, that E2F activity in certain cardiac diseases might provide a therapeutic approach for the development of myocyte hypertrophy that might and the of heart failure.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.