|
- 积分
- 0
- 威望
- 0
- 包包
- 0
|
作者：Jill M. Schroeder, Wenlin Liu, and Norman P. Curthoys作者单位：Department of Biochemistry and Molecular Biology, Colorado StateUniversity, Fort Collins, Colorado 80523-1870 c _9 D$ B0 d% Z/ d% u# m
! C, F9 E7 Q/ c- t& P
2 W. s$ `) b9 A: N/ }& C) l$ ` ( B6 z6 J g3 \& b+ \4 G
% L$ n2 V2 P& u6 ?
: T; ?: M% {+ c7 m. Y7 N" f; w
. f/ ^3 Y* d7 D* T5 z
" m" j- i6 k! |5 X' U8 a0 w
2 ?4 D4 C2 s) V
- o6 z/ _$ v5 Q/ I0 q1 F4 @8 P " t3 L0 l; K# \
+ }7 Z" R2 J2 N5 f ^! c5 T
: {3 N4 C: C. t2 H
【摘要】% @' M( Y8 @0 x: }9 y2 o' e
During chronic metabolic acidosis, the adaptive increase in rat renalammoniagenesis is sustained, in part, by increased expression of mitochondrialglutaminase (GA) and glutamate dehydrogenase (GDH) enzymes. The increase in GAactivity results from the pH-responsive stabilization of GA mRNA. The3'-untranslated region (3'-UTR) of GA mRNA contains a directrepeat of an eight-base AU-rich element (ARE) that binds -crystallin/NADPH:quinone reductase ( -crystallin) with highaffinity and functions as a pH-response element. RNA EMSAs established that -crystallin also binds to the full-length 3'-UTR of GDH mRNA. Thisregion contains four eight-base sequences that are 88% identical to one of thetwo GA AREs. Direct binding assays and competition studies indicate that thetwo individual eight-base AREs from GA mRNA and the four individual GDHsequences bind -crystallin with different affinities. Insertion of the3'-UTR of GDH cDNA into a -globin expression vector (p G)produced a chimeric mRNA that was stabilized whenLLC-PK 1 -F cells were transferred to acidic medium. ApH-responsive stabilization was also observed using a G construct thatcontained only the single GDH4 ARE and a destabilizing element fromphospho enol pyruvate carboxykinase mRNA. Therefore, during acidosis,the pH-responsive stabilization of GDH mRNA may be accomplished by the samemechanism that affects an increase in GA mRNA.
; i" V# ]% m& i' }6 c 【关键词】 renal ammoniagenesis metabolic acidosis posttranscriptional regulation
7 T+ u4 o" I; a- h ----------
0 o" u8 }# U: A* w1 T; i: g; J4 N: f; p6 C! a
RENAL CATABOLISM OF GLUTAMINE is increased significantlyfollowing the onset of metabolic acidosis( 27 ). The resulting increases in renal ammoniagenesis and gluconeogenesis contribute to the maintenance ofacid-base balance ( 1 ). In ratkidney, this adaptation is sustained during chronic acidosis, at least in part, by an increased expression of mitochondrial glutaminase (GA) andglutamate dehydrogenase (GDH), cytosolic phospho enol pyruvate carboxykinase (PEPCK), and various transport proteins( 7 ). The resulting increases inthe activities of these enzymes and transport systems contribute to theincreased production and vectoral transport of ammonium and bicarbonate ionsto facilitate the excretion of acids and partially compensate for the decreasein blood pH, respectively. The increased level of PEPCK results from anincreased rate of transcription of the PEPCK gene( 18 ). However, the increasesin GA and GDH are caused by stabilization of their respective mRNAs( 19, 20 ).
' }3 x) `/ s p( q( b% P7 P3 C5 [4 o5 a' f5 O
A protein in rat renal cortical cytosolic extracts was observed to bindwith high affinity and specificity to the 3'-untranslated region(3'-UTR) of GA mRNA( 23 ). This protein wassubsequently identified as -crystallin/NADPH:quinone reductase( -crystallin) ( 29 ). Theprotein-binding element within GA mRNA was mapped to a 29-base sequence thatcontained a direct repeat of two 8-base AU-rich elements (AREs)( 23 ). The function of this sequence was established by characterizing the effect of medium pH on thehalf-lives ( t 1/2 ) of various chimeric -globin( G) mRNAs that were stably expressed in LLC-PK 1 -F cells ( 21 ). The inclusion of a76-nt GA mRNA segment containing the direct repeat of the two AREs wassufficient to produce a pH-responsive stabilization of a nonresponsive G-PEPCK mRNA. Furthermore, when the two AREs within a G-GA mRNAwere mutated, the pH-responsive stabilization was abolished. The cumulativedata indicate that the two AREs within GA mRNA function as a pH-responseelement (pHRE) and that enhanced binding of -crystallin to this sequence during acidosis may mediate the stabilization of GA mRNA( 7, 22 ).* s. T% Z# K4 D( ~
) [+ y1 o+ F) O0 {$ x* S# T
Previous experiments suggest that the adaptive increase in rat renal GDHmRNA is also mediated through stabilization of its mRNA. For example, theincrease in GDH mRNA following the acute onset of metabolic acidosis occurs inthe same tubular segments ( 8, 33, 34 ) and with similar kineticsas observed for GA mRNA ( 20 ).For both mRNAs, there is an 8- to 10-h lag between the onset of acidosis andthe initial increase in mRNA levels. In addition, whenLLC-PK 1 -F cells were transferred to acidic medium (pH6.9) and treated with actinomycin D to inhibit transcription, an apparentthreefold stabilization of endogenous GDH mRNA was observed( 20 ). These data suggest thatGDH mRNA may be stabilized by the same mechanism that is used to stabilize GA mRNA. The 3'-UTR of rat GDH mRNA( 10 ) contains four eight-base AREs that are 88% identical to one of the two sequences found in GA mRNA.However, unlike the direct repeat found in GA mRNA, the putative GDH pHREs aredistributed throughout the 3'-UTR and, based on previous data, it wasuncertain whether an individual ARE could function as a pHRE.
2 w7 r3 Z8 N8 I- z' {! R2 q% ]/ O' N3 L# l, N2 P& z) [
In the present study, direct binding and competition assays were performedto determine whether purified -crystallin binds to the 3'-UTR ofGDH mRNA and to the individual AREs from GA and GDH mRNAs. The 3'-UTR ofGDH mRNA and a representative GDH ARE were then cloned into G reporterconstructs and stably expressed in LLC-PK 1 -F cells. Todetermine whether the sequences also function as pHREs, t 1/2 analyses were performed. The cumulative data stronglysupport the conclusion that individual AREs within the two mRNAs can bind -crystallin with sufficient affinity to function as a pHRE.
8 p. Z# g/ F, [6 ?, M* b6 O
3 r# B4 B @# V8 f3 ]MATERIALS AND METHODS* p5 f5 G# U6 a: _( ]
$ ?8 M8 l; z5 N" a/ n4 }Materials. Male Sprague-Dawley rats (140-160 g) were purchased from Charles River. Slide-a-Lyzer cassettes were obtained fromPierce. [ - 32 P]UTP and [ - 32 P]dCTP (specificactivity 3,000 Ci/mmol) were purchased from ICN Biochemicals or AmershamPharmacia Biotechnology. The oligo labeling kit was from Amersham Pharmacia Biotechnology. Restriction enzymes, RNase T1, T7 RNA polymerase, and yeasttRNA were acquired from Roche, New England Biolabs, and MBI Fermentas.GENECLEAN kits were obtained from Bio101, and a PCRScript cloning kit wasobtained from Stratagene. Micro Bio-spin columns and chemicals for acrylamidegels were purchased from Bio-Rad. RNasin was obtained from Promega. GeneScreen Plus was purchased from New England Nuclear. DMEM/F-12 medium and Geneticin(G418) were products of Sigma and Mediatech, respectively. Guanidinethiocyanate and sodium- N -lauryl sarcosine were obtained from Fluka.All other biochemicals were acquired from Sigma.6 h% E* F% T1 a5 X. U
+ O) \1 w: z& V4 QPurification of -crystallin. -Crystallin waspurified from a rat kidney cortical cystosolic extract by affinitychromatography using a biotinylated RNA ligand( 29 ). The column wasequilibrated, washed, and eluted as described previously except that the bound proteins were eluted with an additional 2.5 ml of binding buffer containing500 mM potassium acetate and 50 mM MgCl 2. Aliquots of the elutedfractions were separated on a 10% polyacrylamide gel containing 1% SDS andstained with 0.1% silver nitrate. Fractions containing the 36-kDa -crystallin protein were pooled and dialyzed overnight vs. bindingbuffer using a Slide-A-Lyzer cassette. The solution was concentrated 5- to10-fold using a Microcon-30 column from Millipore." B6 Q( S N9 h2 `. N/ v% u
: m- M4 X9 l. q6 A2 i3 eSynthesis of RNAs. Rat liver GDH cDNA( 10 ) was digested with Xmn I and Eco RI to obtain a 1,178-nt fragment that contains the entire 3'-UTR. This fragment was then cloned intopBluescriptII-SK(-) (pBSSK) that had been previously digested with Asp 718, blunted with Klenow polymerase, and then cut with Eco RI to yield the pGDH1-4 plasmid (prepared by R. Gallien).The synthesis of the plasmid that encodes the R2-I RNA segment of GA mRNA wasdescribed previously ( 23 ).Transcription vectors containing only a single GA ARE and the individual GDHAREs were constructed by annealing complimentary oligonucleotides that weresynthesized by Macromolecular Resources (Ft. Collins, CO). The sequences ofthe coding strands of the oligonucleotides are the following:5 p6 {8 P/ i9 n3 p V4 }' g
! \3 {' g! O* e |& \* S5'-GTACCTGTGACTC TTTAAATA ATTACTACT-3' [GA(R2-IA)]
3 a1 O# ^, Y4 n, o' m- \; y0 S# d; d- [4 j' f
5'-GTACCTGTGACTC TTAAAATA ATTACTACT-3' [GA(R2-IB)]8 ~( ~4 z. R* x2 G
8 E. r! i/ F6 ~. |" L9 e/ i
5'-GTACCGTTTCGCT TTTAAGTA AAGTTTCT-3' (GDH1)
# ~* T* \! z. ?& K% z9 u
/ z/ A8 F# Z5 d' X8 I+ q5'-GTACCGTGAACGC CTAAAATA TAGGAAATT-3' (GDH1)
. {) E) a: V6 Y+ n y% J9 B" ^
) O1 j+ ^8 v) U% l1 u- Z5'-GTACCAACCATAC TTCAAATA GAGTGTCCT-3' (GDH3)
) F9 m4 T( h! q0 O _6 E
- f/ w- }$ ? s3 `$ x1 i5'-GTACCAGACATTA TTTATATA AGAATGAGT-3' (GDH4)5 c. S0 Q$ P/ L6 \% ?
0 ?7 P. d7 m" ^& Z2 R7 vThe italicized letters designate the AREs. The resulting double-stranded DNAs encode the ARE sequences and form Asp 718 and Xba Ioverhangs. The annealed oligonucleotides were inserted into pBSSK that hadbeen restricted with Asp 718 and Xba I.4 L+ g" S Z7 Y+ }6 @8 d% A
5 o* d+ [ P: W2 S1 q! I- }A DNA template containing a T7 promoter was obtained by digesting thepGDH1-4 plasmid with BssH II, which flanks the multicloning siteof pBSSK. Generating the other templates required additional digestion with Xba I and Sac I. Xba I cleaves DNA immediately afterthe ARE sequence, and Sac I cleaves the similarly sized nontemplateDNA into two pieces so that the template and promoter-less DNA fragments canbe separated on an 8% polyacrylamide gel. The templates were extracted fromthe gel using the crush and soak method( 26 ). The eluted DNA wasprecipitated with 2 vol of 100% ethanol, washed with 200 µl 70% ethanol,and resuspended in 11 µl diethylpyrocarbonate-treated water. In vitrotranscription reactions using T7 RNA polymerase to synthesize 32 P-labeled and unlabeled RNAs were performed as describedpreviously ( 23 ). Theconcentrations of the 32 P-labeled and unlabeled RNAs weredetermined by scintillation counting and by measuring the absorbance at 260 nmand using specific extinction coefficients calculated from the nucleotidecomposition, respectively.
! G6 O! j# y2 w! I7 ^8 v
@7 P6 i- p6 T1 j; a, ?" s2 Y& oRNA electrophoretic mobility shift assay. This assay was performed as reported previously ( 23 ).Briefly, 10-50 ng of purified -crystallin were incubated for 10min at room temperature in a 10-µl reaction containing 10 mM HEPES, pH 7.4,25 mM potassium acetate, 2.5 mM magnesium acetate, 2 µg yeast tRNA, 0.5%Nonidet P-40, 5% glycerol, 1 mM dithiothreitol, and 10 U of RNasin.Approximately 20 fmol of labeled RNA were then added, and the sample wasincubated at room temperature for 20 min. For the competition studies, a 30-,100-, or 300-fold excess of an unlabeled RNA was added along with the labeled RNA. To compare the binding of the GDH1-4 and GA(R2-I) RNAs, the sampleswere also incubated for 10 min with 15 U of RNase T1 to reduce the length ofthe GDH1-4 RNA. The samples were subjected to electrophoresis for 2h at 170 V on a 5% polyacrylamide gel using a 90 mM Tris, 110 mM boric acid, 2mM EDTA running buffer. Gels were then dried and exposed to a PhosphorImagerscreen.
3 k1 Z0 Q9 r0 d7 |" T8 U9 I' L9 k9 D/ ?; Q/ `- B) ]+ J' F
G expression vectors. Various G constructs weresynthesized to contain either the 3'-UTR of the GDH cDNA or anindividual ARE. A 930-nt segment containing the 3'-UTR of GDH cDNA was PCR-amplified from pGDH1-4 using primers that add Spe I and Xba I sites to the 5'- and 3'-ends, respectively. The PCR product was cloned into the Srf I site of pCR-Script-SK( ), and the Spe I/ Xba I fragment was excised from the plasmid and inserted into the Xba I site within the multicloning site of p G( 17 ). The p G-GA(R2-I)and p G-GDH4 vectors were constructed by annealing complimentaryoligonucleotides (Macromolecular Resources, Ft. Collins, CO) that encode theGA(R2-I) or the GDH4 sequence and form Spe I and Xba Ioverhangs and inserting them into the Xba I site of p G. Thecoding sequences of the GA(R2-I) and GDH4 oligonucleotides are5'-CTAGTTC TTTAAATATTAAAATA ATTCTAAT-3' and5'-CTAGTAGACATTA TTTATATA AGAATGAGT-3', respectively. The italicized letters in the oligonucleotide sequences designate the AREs. Thep G-GDH4-PEPCK vector was constructed by K. Propst. Complimentaryoligonucleotides (Macromolecular Resources) that encode the GDH4 sequence andform Xba I and Nhe I overhangs were annealed and inserted intothe Xba I and Nhe I sites of pGEM4Z-PEPCK ( 24 ). The sequence of thecoding strand of the GDH4 oligonucleotide is5'-CTA GATATC AGACATTA TTTATATA AGAATGAGG-3',where the bold italicized letters designate an Eco RV site that wasused to detect the presence of the inserted sequence, and the italicized lightface letters designate the ARE. The pGEM4Z-GDH4-PEPCK vector was digestedwith Xba I and Spe I to obtain the GDH4-PEPCK sequence thatwas inserted into Xba I/ Spe I-digested p G to producep G-GDH4-PEPCK. y8 q. G. n9 u6 }! v
9 x" U* a; }# S! ?
Cell culture. LLC-PK 1 -F cells were obtainedfrom Gerhard Gstraunthaler and cultured as described previously( 16 ). Cells were grown in a50:50 mixture of DMEM and Ham's F-12 containing 5 mM glucose and 10% fetalbovine serum at 37°C in a 5% CO 2 -95% air atmosphere. Normalmedium (pH 7.4) contains 25 mM sodium bicarbonate, whereas acidic medium (pH6.9) contains 10 mM sodium bicarbonate supplemented with 15 mM sodium chlorideto maintain an equivalent osmolarity and sodium ion concentration.LLC-PK 1 -F cell lines that stably express the chimericmRNAs were produced by transfection of 3-day postsplit cells with calciumphosphate-precipitated DNA ( 3 )and selection with medium containing 0.8 mg/ml G418. The medium was changedevery 2 days. After 14-21 days, three 10-cm plates containing multiplecolonies were combined. Following the next split, the cells were grown innormal medium containing 0.2 mg/ml G418.
1 k- Q! p" _: E6 W O; S! h, N9 {) `1 ?$ a. F
Analysis of t 1/2. The various transfectedLLC-PK 1 -F cell lines were generally split 1:10 andgrown for 7-10 days in pH 7.4 medium containing 0.2 mg/ml G418. Theywere then maintained in pH 7.4 medium without G418 for 24 h and subsequentlytreated for 12 h with normal or acidic medium. At time 0, 65 µM 5-6 dichloro-1- -ribofuranosylbenzimidazole (DRB), a specific inhibitor of RNA polymerase II transcription( 11 ), dissolved in 95% ethanolwas added to each plate. An equivalent concentration of 95% ethanol was addedto control plates. The final concentration of ethanol never exceeded 0.5%. At0, 3, 6, and 9 h post-DRB treatment, total cellular RNA was isolated using themethod of Chomczynski and Sacchi( 6 ). The RNA concentration wasdetermined by measuring the absorbance of the RNA at 260 nm.
) M! P* `- S9 D l% j, |
6 P3 t. r* S4 r3 Q* F( D/ LNorthern blot analysis. A 507-bp fragment of rabbit G cDNA was excised by restricting pRSV- G( 15 ) with Hin dIII and Bgl II. A 2.0-kb fragment of the 18S ribosomal RNA cDNA from Acanthamoeba castellanii was excised by restricting pAr2 with Hin dIII and Eco RI( 9 ). The fragments wereseparated on 1% agarose gels, excised, and purified using a GENECLEAN kit. Asynthesis of oligolabeled cDNA probes and Northern blot analysis was performed as described previously ( 17 ).The blots were exposed to a PhosphorImager screen and quantified usingMolecular Dynamics software. The level of the chimeric G mRNA wasdivided by the level of the corresponding 18S rRNA to correct for errors insample loading. The log of normalized data was then plotted vs. the time of treatment with DRB.
7 z! O% F+ D1 _' a* C0 }- b, h, R% e7 U* |/ C: e1 e
RESULTS
8 ~* S6 X7 p* H; o/ J( y% J
' o# A- A. U: g$ l' a* UAn RNA/protein complex was formed when GDH1-4 RNA, containing theentire 3'-UTR of GDH mRNA (bases 1696-2874), was incubated withpurified -crystallin ( Fig.1 ). The shifted complex had the same mobility as the complex thatis formed when -crystallin is bound to GA(R2-I) RNA. However, relative to the GA(R2-I)/ -crystallin complex, very little of the total radioactivity is contained in the GDH1-4/ -crystallin complex. Theobserved difference could be due, in part, to the relative sizes of the twoRNAs. To resolve the unbound GDH1-4 RNA from the shifted complex, it wasnecessary to treat the samples with RNase T1. This treatment had no effect onthe complex formed with GA(R2-I) RNA. Thus most, if not all, of thenucleotides in bound GA(R2-I) RNA are protected from digestion. However, onlya few of the 1,200 nucleotides within GDH1-4 RNA are retained inthe complex. As a result, most of the label in GDH1-4 RNA is digestedand migrates as short oligonucleotides, whereas nearly all of the boundGA(R2-I) RNA appears in the shifted complex. Despite this difference, theobserved complex indicates that the 3'-UTR of GDH mRNA contains one ormore elements that bind -crystallin./ p# X) B; V) p1 c) C
0 h e8 Z; U, X9 |, k4 G9 \0 ?7 P' w
Fig. 1. Binding of -crystallin to the R2-I RNA segment of glutaminase[GA(R2-I)] and glutamate dehydrogenase oligonucleatide (GDH1-4) RNAs.GA(R2-I) RNA is a 29-base segment from the 3'-untranslated region (UTR)of GA mRNA that contains the direct repeat of the two 8-base AU-rich elements(AREs). GDH1-4 RNA (1.2 kb) contains the entire 3'-UTR of the GDHmRNA. Either 2 ( lanes 3, 4, and 6 ) or 4 µl( lane 7 ) of purified -crystallin ( -crys) were added tothe samples. Samples in lanes 2 and 4-7 were treatedwith RNase T1 to digest the unbound RNA. The RNA/protein complexes wereresolved on a nondenaturing polyacrylamide gel. The gel was then dried andimaged with a PhosphorImager screen. The arrow indicates the position of -crystallin/RNA complex. The percentage of total radioactivity containedin the RNA/protein complex is indicated as % shifted. The reported data arerepresentative of 3 separate experiments.
4 A6 q/ O* ^$ @2 N
9 l" \; h8 }7 Y/ b" {7 Z$ C% L5 rTo determine whether -crystallin can bind to individual eight-base ARE sequences, it was necessary to synthesize short RNAs of similar lengthsthat contain a single ARE. As a control, RNAs containing the individual AREsfrom GA mRNA were synthesized. GA(R2-IA) RNA contained the UUUAAAUA element(bases 2596-2603), and GA(R2-IB) RNA contained the UUAAAAUA element(bases 2604-2611) in the context of the surrounding sequence of the3'-UTR of GA mRNA ( 28 ).Purified -crystallin binds to the two RNAs to form complexes that havemobilities identical to those observed with GA(R2-I) RNA( Fig. 2 ). Thus it is unlikelythat multiple copies of -crystallin bind to GA(R2-I) RNA even though itcontains two potential binding elements.0 p- j) g. I+ W- g
& ^5 ]5 i& s q8 H. H
Fig. 2. Binding of -crystallin to ribonucleotides containing the individualAREs from GA and GDH mRNAs. 32 P-labeled GA(R2-I), GA(R2-IA),GA(R2-IB), GDH1, GDH2, GDH3, and GDH4 RNAs were incubated in the absence(-) or presence ( ) of 2 µl of purified -crystallin. Thesamples were resolved on a nondenaturing polyacrylamide gel. The gel was thendried and imaged with a PhosphorImager screen. The arrow indicates theposition of -crystallin/RNA complex. The percentage of totalradioactivity contained in the RNA/protein complex is indicated as % shifted.The reported data are representative of 3 separate experiments.4 h2 P* J5 b' U: Z
) J9 }4 i9 Q& _ N/ P2 @; b* P
Experiments were also performed to determine whether -crystallin binds to the individual eight-base AREs from the 3'-UTR of GDH mRNA. Thesequences and locations of the AREs within the 3'-UTR of GDH areUUUAAGUA (GDH1; bases 1973-1980); CUAAAAUA (GDH2; bases2313-2320); UUCAAAUA (GDH3; bases 2537-2544); and UUUAUAUA (GDH4;bases 2749-2756) ( 10 ).Purified -crystallin binds to each of the RNAs containing a single ARE from the GDH 3'-UTR ( Fig.2 ) but to a lesser extent than observed for GA(R2-I) RNA or theindividual GA AREs. Of the four GDH AREs, GDH2 and GDH4 RNAs demonstrate thehighest apparent affinity for -crystallin. Thus the individual GDH AREscan function as binding elements for -crystallin. An additional, moreslowly migrating band was reproducibly observed only with GDH4 RNA. Thencomposition or the significance of this band is unkown.
8 U$ i% B0 V3 ?/ z/ J; k% e Q% ]5 X# }* I! P
Competition studies were performed to assess the relative affinity of -crystallin binding to individual GA and GDH AREs. Purified -crystallin was bound to 32 P-labeled GA(R2-I) RNA, andincreasing amounts of unlabeled RNAs were added as competitors. Competition studies demonstrated that a 100-fold excess of unlabeled GA(R2-IA) RNA orGA(R2-IB) RNA was required to produce a level of competition similar to thatobserved with a 30-fold excess of GA(R2-I) RNA( Fig. 3 ). Thus the twoindividual elements exhibit similar affinities for -crystallin.Furthermore, the relative affinity for the two individual elements isapproximately one-third of that for the RNA that contains the direct repeat ofthe individual binding sites.
$ r9 f" @3 B3 q% Y* V( V' \) W4 }
Fig. 3. Competition analysis of the relative binding of -crystallin toGA(R2-I), GA(R2-IA), and GA(R2-IB) RNAs. 32 P-labeled GA(R2-I) RNAwas incubated with purified -crystallin ( lanes 2-10 ). A30-, 100-, or 300-fold excess of an unlabeled RNA competitor was added asindicated. The samples were resolved on a nondenaturing gel. The gel was thendried and imaged with a PhosphorImager screen. The arrow indicates theposition of -crystallin/RNA complex. The percentage of totalradioactivity contained in the RNA/protein complex is indicated as % shifted.The reported data are representative of 3 separate experiments.3 j: M/ K& q0 o+ Z( Q
5 f: r% j! l' j9 n
RNAs containing individual GDH AREs were also tested as competitors oflabeled GA(R2-I) RNA ( Fig. 4 ).GDH2 and GDH4 RNAs were more effective competitors than GDH3 and GDH1 RNAs. A300-fold excess of GDH2 or GDH4 RNA competes slightly less effectively than a100-fold excess of GA(R2-I) RNA. In contrast, GDH1 and GDH3 RNAs were weakercompetitors. Thus the competition pattern observed with individual GDHelements confirms the results of the direct binding studies. The lower bandsobserved when GA(R2-IA) and GA(R2-IB) RNAs( Fig. 2 ) and GDH2 RNA( Fig. 3 ) were added ascompetitors represent undissociated dimers of 32 P-labeled GA(R2-I)RNA./ C0 ?. u1 [0 l" w! G
, e1 \. ~6 |* F5 ~ r6 w
Fig. 4. Competition analysis of the relative binding of -crystallin to GDH1,GDH2, GDH3, and GDH4 RNAs. 32 P-labeled GA(R2-I) RNA was incubatedwith purified -crystallin ( lanes 2-10 ). A 100- or300-fold excess of an unlabeled RNA competitor was added as indicated. Thesamples were resolved on a nondenaturing gel. The gel was then dried andimaged with a PhosphorImager screen. The arrow indicates the position of -crystallin/RNA complex. The percentage of total radioactivity containedin the RNA/protein complex is indicated as % shifted. The reported data arerepresentative of 3 separate experiments.2 n7 K& D7 m( L" N. m
7 o& J8 X" R# c! \9 g8 S/ E4 M
Functional studies were performed to determine whether the 3'-UTR ofGDH mRNA is sufficient to produce a pH-responsive stabilization of a reportermRNA. This was accomplished by measuring the half-lives of various chimeric G mRNAs expressed in LLC-PK 1 -F cells treated witheither normal (pH 7.4, 25 mM ) oracidic (pH 6.9, 10 mM ) medium. Thep G plasmid contains a promoter derived from the Rous sarcoma virus longterminal repeat, a transcription start site, the coding region from rabbit G genomic DNA containing three exons and two introns, a multicloningsite, and a 3'-UTR and polyadenylation site from bovine growth hormone(bGH) cDNA ( 17 ).
, u; q# M* C. u& N' W7 a3 A J, h4 [7 ]9 r
As a control, p G was stably expressed in the split ofLLC-PK 1 -F cells used in the present experiments.Neither the level nor the t 1/2 30 h) of G mRNAwas affected by growing the cells in either normal or acidic medium (data notshown). As a second control, the LLC-PK 1 -F cells werealso stably transfected with the p G-GA plasmid that contains 955-basesfrom the 3'-UTR of GA mRNA( 17 ). The p G-GA vectorproduced an unstable mRNA that had a t 1/2 of 12 h incells maintained in normal medium. However, G-GA mRNA exhibited apH-responsive stabilization ( t 1/2 30 h) when the cellswere transferred to acidic medium (data not shown). These observations confirmthe previous conclusion that the 3'-UTR of GA mRNA contains aninstability element and imparts a pH-responsive stabilization to G mRNA( 17 ).+ q- ~% T% [/ f; F8 k
8 D% b% S- @0 \4 v- Y% N) o
Functional studies were performed using the p G-GDH vector that encodes the 930-base 3'-UTR of GDH mRNA including all four of theeight-base AREs ( Fig. 5 ). Theresulting G-GDH mRNA had a t 1/2 of 13 h inLLC-PK 1 -F cells that were grown in normal medium. Whenthe cells were transferred to acidic medium, the t 1/2 of 30 h. Thus this mRNA also exhibits apH-responsive stabilization. This observation indicates that the 3'-UTRof GDH mRNA contains one or more pHREs that impart selective stabilization toGDH mRNA in response to treatment with an acidic medium.
. D1 t2 r1 w+ d# {' ]
8 L- k# |- Y( _Fig. 5. Half-life analysis of -globin ( G)-GDH mRNA. A :schematic of G-GDH mRNA. B : Northern blot analysis of G-GDH mRNA isolated from LLC-PK 1 -F cells stablytransfected with p G-GDH. Cells were either maintained in normal (pH 7.4)medium or transferred to acidic (pH 6.9) medium for 12 h and then treated with65 µM 5-6-dichloro-1- -ribofuranosylbenzimidazole (DRB) for 0,3, 6, and 9 h. C : relative levels of G-GDH mRNA were determinedby PhosphorImager analysis. The level of G-GDH mRNA was divided by thecorresponding level of 18S rRNA to correct for errors in sample loading. Thelog of normalized data was then plotted vs. the time of treatment with DRB.The reported data are the mean of duplicate experiments. The error barsrepresent the range of the 2 data points for the 9-h time points.& T2 f: q- z B) Q
! E7 U4 E8 N% v9 {% A* n
To test the function of shorter segments, the 29-base R2-I fragment of GAcDNA encoding the direct repeat of the two eight-base AREs of GA mRNA wascloned into p G. This construct lacks the 3'-UTR of PEPCK mRNA thatwas included in previous studies ( 21 ). Therefore, it was usedto determine whether the GA sequence can function as both a destabilizingelement and a pHRE. When LLC-PK 1 -F cells were stablytransfected with p G-GA(R2-I) and grown in normal medium, G-GA(R2-I) mRNA decayed with a t 1/2 of 12 h( Fig. 6 ). When the same cellswere transferred to acidic medium, the chimeric RNA was degraded with a t 1/2 of 30 h. This indicates that the sequencecontained within the GA(R2-I) region of GA 3'-UTR is sufficient tofunction as an instability element and a pHRE.4 `0 q( I$ C/ x+ p# \; y
& X+ U% i2 E/ t' D! @, WFig. 6. Half-life analysis of G-GA(R2-I) mRNA. A : schematic of G-GA(R2-I) mRNA. B : Northern blot analysis of G-GA(R2-I)mRNA isolated from LLC-PK 1 -F cells stably transfectedwith p G-GA(R2-I). Cells were either maintained in normal (pH 7.4) mediumor transferred to acidic (pH 6.9) medium for 12 h and then treated with 65µM DRB for 0, 3, 6, and 9 h. C : relative levels of G-GA(R2-I) mRNA were determined by PhosphorImager analysis. The level of G-GA(R2-I) mRNA was divided by the corresponding level of 18S rRNA tocorrect for errors in sample loading. The log of normalized data was thenplotted vs. the time of treatment with DRB. The reported data are the mean ofduplicate experiments. The error bars represent the range of the 2 data pointsfor the 9-h time points.8 A$ Q8 d: y2 t7 Y
5 c' S- W+ H, I* _& u9 TThe direct binding and competition analyses indicated that the fourth AREwas one of two sites within GDH mRNA that binds -crystallin with thegreatest affinity. Therefore, the GDH4 sequence was cloned into p G totest whether a single GDH element was sufficient to function as a pHRE. The t 1/2 of G-GDH4 mRNA inLLC-PK 1 -F cells maintained in either normal or acidic 30 h (data not shown). Thus the 35-base GDH4 sequence does notsignificantly destabilize G mRNA. Given the inherent stability of thisconstruct, it could not be used to determine whether the GDH4 sequence canfunction as a pH-responsive stabilizing element.
3 C! A/ ^# r# o- I% ~3 H9 w: h3 X% |0 V" N% X7 f9 s
To assess whether the GDH4 sequence can function as a pHRE, the3'-UTR of PEPCK cDNA was cloned downstream of the GDH4 element toproduce the p G-GDH4-PEPCK plasmid. Insertion of the PEPCK sequencedestabilizes the resulting G-GDH4-PEPCK mRNA( Fig. 7 ). The t 1/2 of this mRNA in LLC-PK 1 -F cells grown in pH 7.4 medium was 16 h. However, when the cells were treatedwith pH 6.9 medium, the t 1/2 30 h.Therefore, the GDH4 sequence can function as a pHRE but not as a destabilizingelement.
5 s/ t5 Z$ J w6 o! f. W! z l5 Q9 S9 U, Q) T) @( f6 t
Fig. 7. Half-life analysis of G-GDH4-phospho enol pyruvatecarboxykinase (PEPCK) mRNA. A : schematic of the G-GDH4-PEPCKmRNA. B : Northern blot analysis of G-GDH4-PEPCK mRNA isolatedfrom LLC-PK 1 -F cells stably transfected withp G-GDH4-PEPCK. Cells were either maintained in normal (pH 7.4) medium ortransferred to acidic (pH 6.9) medium for 12 h and then treated with 65 µMDRB for 0, 3, 6, and 9 h. C : relative levels of G-GDH4-PEPCKmRNA were determined by PhosphorImager analysis. The level of G-GDH4-PEPCK mRNA was divided by the corresponding level of 18S rRNA tocorrect for errors in sample loading. The log of normalized data was thenplotted vs. the time of treatment with DRB. The reported data are the mean ofduplicate experiments. The error bars represent the range of the 2 data pointsfor the 9-h time points.
/ ~2 z ?: k% O2 B j9 T
+ D1 i) B, U8 b K* r+ m1 {DISCUSSION
% x a3 [; E0 q' _* x8 { s" j1 s. W B! o8 W! i3 p
The control of mRNA degradation plays an important role in the posttranscriptional regulation of gene expression. Some mRNAs turnover rapidlywith a t 1/2 of * z/ q* f$ T1 S* _
+ ]3 g$ _) L1 u0 r: A+ gThe 5'- and 3'-ends of most eukaryotic mRNAs are protected from exonuclease degradation by the binding of eIF4E and poly(A) binding proteins(PABP) to the 7-methyl-guanosine cap and the poly(A) tail, respectively. Thetwo ends are then linked together by eIF4G that acts as a bridging protein bybinding to eIF4E and PABP ( 13, 14 ). Deadenylation is theinitial step in the degradation of most mRNAs( 4, 31 ). In yeast, deadenylation is usually followed by decapping and 5' 3' exonucleolytic degradation. However, in mammals the prominent pathway involves initialdeadenylation followed by 3' 5' degradation. Various AREs,including the canonical AUUUA sequence( 4 ), function as instabilityelements by binding proteins, such as TTP( 2 ), that recruit the 3' 5' poly(A)-specific deadenylase DAN/PARN ( 13 ) and a complex of 3' 5' exoribonucleases that has been termed the exosome( 30 ). The deadenylation andsubsequent degradation of an mRNA can be averted by the alternativerecruitment of a stabilizing ARE-binding protein such as HuR( 5 ).
5 D! k% y9 h% c9 z1 |7 i- _
8 C* k6 }- `& m. ]On the basis of previous experiments, it was hypothesized that the pHRE ofGA mRNA also functions as the recognition site for a sequence-specificendonuclease ( 7 ). During normalacid-base balance, the weak interaction of -crystallin with the pHRE may allow for recruitment of the endonuclease that initiates the rapid degradationof GA mRNA. The onset of acidosis leads to an enhanced interaction of -crystallin with the pHRE( 22 ) that may be mediated by akinase that is upstream of the p38 stress-activated protein kinase( 12 ). The enhanced binding of -crystallin may stabilize GA mRNA by blocking the recruitment of thesequence-specific endonuclease. However, more recent pulse-chase experimentsindicate that the degradation of the chimeric G-GA mRNA inLLC-PK 1 -F cells is preceded by deadenylation and occurswithout apparent endonucleoytic cleavage (Schroeder JM and Curthoys NP,unpublished observations). Thus the pHRE may function as a site thatalternatively binds the stabilizing protein, -crystallin, or anARE-binding protein that recruits a deadenylase and the exosome.
0 l# T5 W% f b& t
8 P! t" n. c% U3 @% l8 uThe revised hypothesis is also supported by the analysis of the turnover ofthe chimeric G-GA(R2-I) mRNA that contains the direct repeat of theeight-base AREs of GA mRNA ( Fig.6 ). Insertion of this segment was sufficient to destabilize G mRNA in LLC-PK 1 -F cells maintained in normalmedium (pH 7.4) and produce a pH-responsive stabilization when the cells weretransferred to acidic medium (pH 6.9). Furthermore, the magnitude of botheffects was similar to that observed in the control experiment that utilized G-GA mRNA. Therefore, the 29-base R2-I segment functions as both adestabilizing element and a pHRE. This may occur through the alternaterecruitment of a destabilizing ARE-binding protein and -crystallin." b, u9 s) ?9 c
: e* X, |* q* h- Q& v a! L w3 E: sThe binding of -crystallin to the individual AREs of the GA 3'-UTR was previously studied using a crude rat renal cytosolic extractand GA(R2-I) RNAs in which the eight-base AREs were mutated to include fiveguanine and cytosine residues( 23 ). The RNAs containing asingle mutated ARE exhibited a large reduction in apparent binding affinityfor -crystallin compared with wild-type GA(R2-I) RNA. When both of theAREs were mutated, binding was abolished. In contrast, both the direct binding studies ( Fig. 2 ) and thecompetition analysis ( Fig. 3 )performed with GA(R2-IA) and GA(R2-IB) RNAs, which contain a single ARE in thecontext of the surrounding sequence from the 3'-UTR of GA mRNA, indicatethat a single eight-base element binds -crystallin with only slightlyless affinity than the complete R2-I sequence. Thus the decreased bindingaffinity observed with the single-site mutations may have been caused by the increase in the GC content of the mRNAs and not the missing ARE.
9 ~& E4 y8 c8 G- `1 U- u- [6 o! d$ J8 l5 Y
Previous studies suggested that the increase in renal GDH mRNA duringacidosis is also mediated by a cell-specific stabilization ( 20 ). The 3'-UTR of GDHmRNA contains four well-spaced eight-base segments that are 88% identical toone of the AREs in the GA mRNA( 9 ). The reported experimentsestablish that the 3'-UTR of GDH mRNA binds -crystallin andcontains both an instability element and a pHRE. Furthermore, the directbinding studies and the competition analysis indicate that all four putative AREs bind -crystallin with varying affinities, all of which were lessthan that observed for the individual elements from GA mRNA. The GDH2 and GDH4sequences showed a greater binding affinity than the GDH1 and GDH3 elements.The putative element with GDH4 RNA contains only A and U residues, whereas theGDH2 element contains a single C residue at the 5'-end of the eight-base sequence. In contrast, the GDH3 and GDH1 elements contain C and Gsubstitutions at the number 3 and 6 positions, respectively. Therefore, substitutions that interrupt the stretch of AU nucleotides may havea greater negative impact on the binding of -crystallin.% N/ C: W! C e6 P- k8 G) ~( Q8 l
L, X: x) {1 Q. S8 j
The greater affinity of -crystallin for the direct repeat of the AREswithin GA mRNA compared with the individual AREs within GDH mRNA is consistentwith the observed changes in the levels of the two enzymes that occur duringacidosis. The GA activity is increased 8- to 20-fold within the renal proximalconvoluted tubule during chronic acidosis( 8, 34 ), whereas the GDH activity within the same cells is increased only three-fold( 33 ). Therefore, thepreferential binding of a limiting amount of -crystallin couldcontribute to the greater fold-stabilization of GA mRNA.+ B' n0 x" f( D; C, ?3 F
7 d. u" [) E! B, BFunctional studies using p G-PEPCK mRNA indicated that the GDH4 AREcan function as a pHRE. However, either the multiple AREs or an alternativeelement within the 3'-UTR is needed to impart a significantdestabilization to GDH mRNA. Identification of the destabilizing element willrequire the synthesis and characterization of additional constructs thatindividually assess the function of the three additional AREs or that mutatethe individual sites within the pGDH1-4 plasmid.
1 j9 H4 k+ V! V: ^" p% F7 M3 T4 b r i' n3 y* \
The combined binding and functional studies establish that individual AREswithin GA and GDH mRNAs bind -crystallin with different affinities andthat a single ARE from GDH mRNA is sufficient to mediate a pH-responsivestabilization. A recent analysis using cDNA microarrays indicated that theonset of metabolic acidosis leads to increased expression of a large number of genes within the renal cortex( 32 ). Thus it will beinteresting to determine how many of these genes encode an eight-base sequence in their 3'-UTR that is similar to pHRE sequences of GA or GDH mRNAs.
& {$ U5 c/ X! Q% b3 {) ?) Y8 a4 q& J# e7 W% P2 p+ o2 ]8 y+ W
DISCLOSURES
# A- R7 i/ D& ^* D4 y. g, Z; t' r+ Y, K
This work was supported by the Public Health Service through NationalInstitute of Diabetes and Digestive and Kidney Diseases Grant DK-37124.: D; c3 U' A/ s) d
【参考文献】
& U4 Z0 P9 z5 j; u9 C( J, U6 m Brosnan JT,Vinay P, Gougoux A, and Halperin ML. Renal ammonia production and itsimplications for acid-base balance. In: pH Homeostasis, Mechanismand Control, edited by Haussinger D. New York: Academic,1988, p. 281-304.
! q, }( [! V( @ E0 Q) l0 @; g' |0 R7 R5 t3 F" u! `( ~' R
& P0 @" u" d4 B8 A! u9 g" J# c. C
2 t: X* u0 e; W5 r% {6 P
Chen CY, GherziR, Ong SE, Chan EL, Raijmakers R, Pruijn GJ, Stoecklin G, Moroni C, Mann M,and Karin M. AU binding proteins recruit the exosome to degradeARE-containing mRNAs. Cell 107:451-464, 2001.
( b: [# k, j" A l4 ^) D
; e- j2 `" N5 B+ K# N) f! U3 W; [) |& R
, Q1 e7 U3 g! W# R6 r' d$ p) n
Chen C andOkayama H. High-efficiency transformation of mammalian cells by plasmidDNA. Mol Cell Biol 7:2745-2752, 1987.) n. { m- i, r* L# Y% [
( Z8 a- T( E# D
% u4 Z& l: \$ L
: L6 y/ V5 {" l. s! @0 C% ~Chen CY andShyu AB. AU-rich elements: characterization and importance in mRNAdegradation. Trends Biochem Sci 20: 465-470,1995.
- j) t2 s! E7 s7 h' Y
7 x/ }) v2 d4 Z" r) J8 \* j
5 g* f4 I5 p5 S5 i9 D- _# x( U- m0 S& l2 t8 v9 ?
Chen CY, Xu N,and Shyu AB. Highly selective actions of HuR in antagonizing AU-richelement-mediated mRNA destabilization. Mol Cell Biol 22: 7268-7278,2002.
+ ^0 I& {& z7 p' n7 _" L0 o; g5 H& h
$ x4 {9 b% F- _" ]: {. r
' v" n9 a4 K" m3 Q- Z: X$ V
Chomczynski P and Sacchi N. Single-step method of RNA isolation by acid guanidiniumthiocyanate-phenol-chloroform extraction. Anal Biochem 162: 156-159,1987.. x4 @. ~+ K( w5 X
) _$ p# K8 i5 X+ W K3 D
8 r, N1 }, `' {
5 ^+ B: b, _' V" t& a0 ]" A0 ACurthoys NP andGstraunthaler G. Mechanism of increased renal gene expression duringmetabolic acidosis. Am J Physiol Renal Physiol 281: F381-F390,2001.' ]# @8 E' h8 r. ^( U9 J" ~
% p" q, B0 `5 ] H! o- A/ T; ~% A2 U- {9 s9 e
5 J+ P* W4 {5 x8 ?. U
Curthoys NP andLowry OH. The distribution of glutaminase isoenzymes in the variousstructures of the nephron in normal, acidotic and alkalotic rat kidney. J Biol Chem 248:162-167, 1973., f: L& g4 B/ a7 e
+ ~) N3 N3 P; D. g* r6 d
/ G9 j. A$ `' n: U, u2 \6 }. _. h: D- }% ]
D'Alessio JM,Harris GH, Perna PJ, and Paule MR. Ribosomal ribonucleic acid repeat unitof Acanthamoeba castellani : cloning and restriction edonuclease map. Biochemistry 20:3822-3827, 1981.
. t# `9 b$ [/ h0 B0 c% }" d4 T( f; i/ f% B# j( F- @- ^
& A2 |9 N3 ~8 {5 r' t. T$ u2 D5 r* [- J- p4 f; A% O% s
Das AT, MoererP, Charles R, Moorman AFM, and Lamaers WH. Nucleotide sequence of ratliver glutamate dehyrogenase cDNA (Abstract). Nucleic AcidsRes 17: 2355,1989.
: _8 \3 e% H' ~3 \8 r) }4 I4 \
2 ]4 ]# A' a* G7 p6 E4 @+ R& w* B+ M8 I5 [" f
2 u8 {* k3 b% k, r1 TDubois MF,Nguyen VT, Bellier S, and Bensuade O. Inhibitors of transcription such as5,6-dichloro-1-b- D -ribofuranosyl-benzimidazole and isoquinolinesulfonamide derivaties (H-8 and H-7) promote dephosphorylation of thecarboxyl-terminal domain of RNA polymerase II largest subunit. JBiol Chem 269:13331-13336, 1994., {8 Z7 M1 }. T' I" [6 F3 y0 ?) O" g
+ e, h7 ~1 ?# \/ Q& }4 q9 s4 i/ Y2 Y0 F1 q) z2 f
l( i) [$ D- ?; q4 E7 {Feifel E,Obexer P, Andratsch M, Euler S, Taylor L, Tang A, Wei Y, Schramek H, CurthoysNP, and Gstraunthaler G. p38 MAPK mediates the acid-induced transcriptionof phospho enol pyruvate carboxykinase inLLC-PK 1 -FBPase cells. Am J Physiol RenalPhysiol 283:F678-F688, 2002.
6 O3 d6 u: `6 w" T0 K: P2 d
, r! Y a/ b6 ]- h8 b' L% d& ^' T1 n6 h9 `5 C5 ]1 O
, [4 g# x5 m# g0 s; EGao M, FritzDT, Ford LP, and Wilusz J. Interaction between a poly(A)-specificribonuclease and the 5' cap influences mRNA deadenylation rates invitro. Mol Cell 5:479-488, 2000.; z7 I4 C2 a; U: k& B: M
8 j: h! i- Q7 V" \; d p c
, s' H# T3 s3 b7 o- I* P6 P( E
: D% y o. {" q% p2 Q% G
Gao M, WiluszCJ, Peltz SW, and Wilusz J. A novel mRNA-decapping activity in HeLacytoplasmic extracts is regulated by AU-rich elements. EMBOJ 20:1134-1143, 2001.& D$ W, o9 t1 e" m% A, M
: T& R3 f7 h/ d) {! T
' D* a" J$ q: `/ ^; V* y) S* H: D$ v. n% C! u9 P0 t& i
Gorman C,Padmanabhan R, and Howard BH. High efficiency DNA-mediated transformationof primate cells. Science 221:551-553, 1983., J, S1 T& D3 Q( v( X' n4 m1 a
' O4 @6 x+ C. p: o( ?/ [, [4 J
. N6 G2 y! z- @# O
8 p$ R6 Q4 F+ v: o. |5 E$ t
Gstraunthaler G and Handler JS. Isolation, growth and characterization of a gluconeogenicstrain of renal cells. Am J Physiol Cell Physiol 252: C232-C238,1987.
& H4 q$ [& S3 ~2 a F5 W& m9 k) C6 ]% ^+ |
) ]1 I1 e" ]) S, }( c
6 c" Q: ` J$ f5 q+ l
Hansen WR,Barsic-Tress N, Taylor L, and Curthoys NP. The 3'-nontranslatedregion of the rat renal glutaminase mRNA contains a pH-responsive stabilityelement. Am J Physiol Renal Fluid Electrolyte Physiol 271: F126-F131,1996.
9 m' u/ K& v7 s# M6 ]( E. I' ]8 L6 a3 }! h$ z
% Y7 @) [/ f7 a, J) K4 }3 S0 s9 {# k/ ^! m$ v
Hwang JJ andCurthoys NP. Effect of acute alterations in acid-base balance on rat renalglutaminase and phospho enol pyruvate carboxykinase gene expression. J Biol Chem 266:9392-9396, 1991.9 z; y( D" |5 X- w' ^" `
, s. ?" _. |/ n& \# |( P$ E4 ^3 Q6 n! _1 q! g
8 L& r: L2 [! Q: ^5 H. z, `. sHwang JJ,Perera S, Shapiro RA, and Curthoys NP. Mechanism of altered renalglutaminase gene expression in response to chronic acidosis. Biochemistry 30:7522-7526, 1991.
# z' r: d" ^2 s/ A2 K$ o% j, F6 @: x7 G8 ~4 S8 ~2 S
" R+ h$ i" a9 ^2 w9 u8 r2 \, F5 C' y2 p9 r! ], \4 J
Kaiser S, HwangJJ, Smith H, Banner C, Welbourne TC, and Curthoys NP. Effect of alteredacid-base balance and of various agonists of the levels of renal glutamatedehydrogenase mRNA. Am J Physiol Renal Fluid ElectrolytePhysiol 262:F507-F512, 1992.# q5 {, }# O/ ?2 a" \- K p
, D2 q! k6 S# W; s5 _) G8 S( H/ Q2 w! v. C* }# @
3 t. d# @1 V- ~1 A% l8 k
Laterza OF andCurthoys NP. Specificity and functional analysis of the pH-responsiveelement within renal glutaminase mRNA. Am J Physiol RenalPhysiol 278:F970-F977, 2000./ u2 V/ |9 d) n7 C8 f
) Z" E! }( J. J( V- W
) V0 o! P! Z# \/ u ]7 ^0 T K8 Z- e w; f
Laterza OF andCurthoys NP. Effect of acidosis on properties of the glutaminase mRNApH-response element binding protein. J Am Soc Nephrol 11: 1583-1588,2000.
; W* {6 D' T1 m- b( p
& |( O' {9 C1 E9 `! J0 v1 b- _- f5 a+ U/ B9 e
* b& `: F& V0 E, j& N, L" L
Laterza OF,Hansen WR, Taylor L, and Curthoys NP. Identification of an mRNA-bindingprotein and the specific elements that may mediate the pH-responsive inductionof renal glutaminase mRNA. J Biol Chem 272: 22481-22488,1997.0 Y+ V7 D* M6 s/ b' P
0 F$ z' C V* }) u& w! ?! ~2 [5 E+ f4 Z
* Z: L5 C$ c" m; a: @! h# J! k
Laterza OF,Taylor L, Unnithan S, Nguyen L, and Curthoys NP. Mapping and functionalanalysis of an instability element in phospho enol pyruvatecarboxykinase mRNA. Am J Physiol Renal Physiol 279: F866-F873,2000.
: e7 r( m9 K. P
6 J7 s+ f" f$ ]- u' U' ]5 }; r# u U# @
) _: t }/ E2 X8 d, YRoss J. mRNA stability in mammalian cells. Microbiol Rev 59: 423-450,1995.' v3 f& m( R4 o$ }% z3 n: n4 `9 m" H
% g c* Z' I9 M, I, \# C3 C& K5 Q) H5 Q
& x1 U9 D7 D* K* o! O7 H6 [
Sambrook J,Fritsch EF, and Manniatis T. Molecular Cloning: A LaboratoryManual (2nd ed.). Cold Spring Harbor, NY: Cold Spring HarborLaboratory, 1989, p.6.1-6.48.8 U I- \* q" S) ^6 u/ ^- W
- k6 i# h& l* p! I# |7 P+ z8 u+ t
5 D, k" K: v/ O5 o
% d, @ w% d/ }/ f* X
Schoolwerth AC. Regulation of renal ammoniagenesis inmetabolic acidosis. Kidney Int 40: 961-973,1991.
+ B' L# x0 P& r0 n& {7 n
( `7 x' o( d" z( H$ ?' T5 s
/ v1 j: k, i; t' D7 q/ _
( W' b d8 E1 R3 k/ W6 o) b5 hShapiro RA,Farrell L, Srinivasan M, and Curthoys NP. Isolation, characterization, andin vitro expression of a cDNA that encodes the kidney isoenzyme of themitochondrial glutaminase. J Biol Chem 266: 18792-18796,1991.
& U8 P- R2 K% v
. ]& j* R- i; \2 M! g$ w1 N/ a# y+ b6 ^ x S
! `! y% A9 s' w: ^$ \( K3 y) w9 pTang A andCurthoys NP. Identification of -crystallin/NADPH:quinone reductaseas a renal glutaminase mRNA pH-response element binding protein. JBiol Chem 276:21375-21380, 2001.
2 f! Q1 \+ m# T! F: ~; q
( q) V9 A# v! Z2 e
, @+ k( `5 f- i5 l5 R i I
1 J+ o+ v& e! q TVan Hoof A andParker R. The exosome: a proteasome for RNA? Cell 99: 347-350,1999.# j% a: \5 @7 \* Y8 }# L# ?3 d8 u
% y: u) y! a9 o; W) n! C
. N, c* s; O" V/ m( i) D
$ W$ [ {0 m* w x( DVan Hoof A andParker R. Messenger RNA degradation: beginning at the end. CurrBiol 12:R285-R287, 2002.
- R' ~3 U! |3 M' j: p! y
7 U7 |0 h5 L7 |+ c$ j6 q$ D' r+ |
+ R2 X6 ^& |: _! O. H5 `. O% |& B7 g4 D+ z- h. M: [& _: f
Wang Z, BurnhamCE, and Soleimani M. Gene microarray identification of elements activatedby metabolic acidosis (Abstract). J Am Soc Nephrol 11: 11A,2000.
. X2 d7 M& F8 c8 y3 L. I: X7 u$ ?# D( S2 B- D h
" C% p: Z1 o" e1 q* C' N, W
z2 ]2 ^6 J6 G$ @Wright PA andKnepper MA. Glutamate dehydrogenase activities in microdissected ratnephron segments: effects of acid-base loading. Am J Physiol RenalFluid Electrolyte Physiol 259:F53-F59, 1990.
$ s) c2 _2 a R, [
& e+ X+ Y4 Y2 a' y$ |3 ]7 Y) P) k/ R1 j8 _! F# H* E
! M9 W' \! i* E8 e# R7 F* g, E: r8 _
Wright PA andKnepper MA. Phosphate-dependent glutaminase activity in rat renal corticaland medullary tubule segments. Am J Physiol Renal Fluid ElectrolytePhysiol 259:F961-F970, 1990. |
|
发表于 2009-4-21 13:42
