A Comparison of Aorta and Vena Cava Medial Message Expression by cDNA Array Analysis Identifies a Set of 68 Consistently Differentially Expressed Genes, All in Aortic Media

A Comparison of Aorta and Vena Cava Medial Message Expression by cDNA Array Analysis Identifies a Set of 68 Consistently Differentially Expressed Genes, All in Aortic Media Lawrence D. Adams, Randolph L. Geary, Bruce McManus, Stephen M. Schwartz Abstract—We performed a systematic analysis of gene expression in arteries and veins by comparing message profiles of macaque aorta and vena cava media using a cDNA array containing 4048 known human genes, ⬇35% of currently named human genes (⬇11 000). The data show extensive differences in RNA expression in artery versus vein media. Sixty-eight genes had consistent elevation in message expression by the aorta, but none were elevated in the vena cava. The most differentially expressed gene was regulator of G-protein signaling (RGS) 5, at an expression ratio of 46.5⫾12.6 (mean⫾SEM). The data set also contained 2 genes already known to be expressed in the aorta, elastin at 5.0⫾1.4, and the aortic preferentially expressed gene 1 (APEG-1) at 2.3⫾0.6. We chose to analyze RGS5 expression further because of its high level of differential expression in the aorta. Levels of RGS5 mRNA were confirmed by Northern analysis and in situ hybridization. A human tissue RNA dot blot showed that RGS5 message is highest in aorta, followed by small intestine, stomach, and then heart. Northern analysis confirmed that RGS5 expression in human aorta is higher than in any region of the heart. RGS5 is a G-protein signaling regulator of unknown specificity most homologous to RGS4, an inhibitory regulator of pressure-induced cardiac hypertrophy. The expression pattern of the 68 differential genes as a whole is a start toward identifying the molecular phenotypes of arteries and veins on a systematic basis. (Circ Res. 2000;87:623-631.) Key Words: cDNA array 䡲 aorta 䡲 vena cava 䡲 expression profile 䡲 RGS5

W

e have begun a systematic analysis of expression differences between vascular smooth muscle tissues. We chose to begin with an analysis of the media of aorta and vena cava because of the possibility of isolating relatively homogenous cellular populations and because of the distinct functional requirements, embryological origins, and hemodynamic environments of these 2 tissues. The present study, using a cDNA array containing 35% of human named genes (4048 of the ⬇10 900 functionally identified human genes in UniGene Build No. 109 of the National Center for Biotechnology Information human Unigene collection), identified 68 genes that are consistently expressed at higher levels in the aorta. No genes were more consistently highly expressed by the vena cava. One of these aortic markers, regulator of G-protein signaling (RGS) 5, showed extreme levels of differential expression. Together, these genes offer a molecular definition of the phenotypes of these 2 types of smooth muscle.

Materials and Methods Array Hybridization Research Genetics GF211 cDNA arrays (4048 human genes) were used as specified. Equivalent amounts of RNA per vessel pair were

used to synthesize cDNA probes, and equivalent counts per minute were added to hybridizations. Washed blots were exposed to phosphor image screens and scanned on a Storm phosphor imager (Molecular Dynamics). Expression was quantified from scans with similar intensity and background, using Pathways software (Research Genetics). Ratios were calculated for each animal and then averaged.

Northern, Dot Blot, and In Situ Hybridization

Total RNA (15 ␮g) from macaque or rat was run, transferred, and hybridized as described1 with multiprime probes (Amersham) synthesized from sequence-verified clones of human and mouse RGS5 (IMAGE clones 853809 and 1853118, respectively [American Type Culture Collection]) and of human nonmuscle myosin heavy chain B (NMMHC-B) (IMAGE clone 823886), and then probed with a 28S rRNA primer as described.2 Washes were as follows: 3 cycles of 5 minutes each at room temperature in 5⫻ SSC, 0.1% SDS, and 2 cycles of 20 minutes each at 57°C (rat) or 55°C (macaque) in 0.3⫻ SSC, 0.1% SDS. Human Northern blot No. 7791-1 (Clontech Laboratories, Inc) containing 2 ␮g poly(A) RNA from adult and fetal heart, aorta, and 5 regions of the heart was probed with human RGS5, then G3PDH as specified. Human RNA master dot blot No. 7770-1 (Clontech) containing 33 tissue, 14 brain region, and 7 fetal poly(A) RNAs (89 to 514 ng each) was probed as specified with human clones for RGS5 and then human ubiquitin. In situ hybridization was performed on cross sections of aorta and vena cava taken directly adjacent to vessel segments snap-frozen for

Received July 26, 2000; revision received August 22, 2000; accepted August 23, 2000. From the Department of Pathology (L.D.A., S.M.S.), University of Washington, Seattle, Wash; Department of Surgery (R.L.G.), Wake Forest University School of Medicine, Winston-Salem, NC; and Department of Pathology and Laboratory Medicine (B.M.), St. Paul’s Hospital, University of British Columbia, Vancouver, British Columbia, Canada. Correspondence to Lawrence D. Adams, Department of Pathology, University of Washington, Box 357335, 1959 NE Pacific St, Seattle, WA 98195-7335. E-mail [email protected] © 2000 American Heart Association, Inc. Circulation Research is available at http://www.circresaha.org

623 Downloaded from http://circres.ahajournals.org/ by guest on July 26, 2015

624

Circulation Research

September 29, 2000

array analyses. These tissues were immersion-fixed for 48 hours in 10% formalin and embedded in paraffin. Sections were cut onto coated slides and in situ hybridization was performed using 35Slabeled riboprobes (sense and antisense) transcribed from human RGS5 cDNA No. 853809 as described.3

TABLE 1.

Gene Expression Categories

Hybridization Classes for Genes, Defined by the Animal Sample With the Lowest Fold Hybridization (Above the Average Background [AB] Before Normalization) in the Higher-Expressing Tissue

Tissue Samples and RNA Isolation

Expression Class

Three human aortas, 2 from males (age 32 and 33 years) and 1 from a female (age 47 years), were isolated from donor hearts obtained from the University of British Columbia–St. Paul’s Hospital. Samples were stripped of adventitia and endothelium and snap-frozen in liquid nitrogen. Macaque tissues were obtained from 5 surgically menopausal (ovariectomized) female cynomolgus macaque monkeys (Macaca fascicularis).4 These monkeys made up the control group from a study of the effects of isoflavones on bypass graft intimal hyperplasia and iliac artery angioplasty. Animals were not administered isoflavones but underwent bypass graft placement between the distal aorta and right iliac artery and angioplasty of the left common iliac artery. The bypass graft originated at the most distal abdominal aorta, with the proximal anastomosis adjacent to the aortic bifurcation and distal anastomosis to the external iliac artery. The vena cava and aorta used for arrays were removed well proximal to the site of surgery (distal thoracic aorta; vena cava from the diaphragm to just above the aortic anastomosis) 6 weeks after surgery. After sedation with ketamine (15 mg/kg IM) and butorphanol (0.05 mg/kg IM), animals were anticoagulated with heparin (300 U/kg IV) and deeply anesthetized with sodium pentobarbital (100 mg/kg IV). The adventitia was stripped from both the aorta and the vena cava, and the endothelium was removed from the aorta but not the vena cava. Tissues were snap-frozen in liquid nitrogen. All animal care and procedures were performed at the Comparative Medicine Clinical Research Center of the Wake Forest University School of Medicine in accordance with state and federal law. Animal protocols were approved by the Wake Forest University Animal Care and Use Committee and conformed to guidelines set forth in the Principles of Laboratory Animal Care (formulated by the National Society for Medical Research), and in the Guide for the Care and Use of Laboratory Animals (NIH publication No. 86-23, revised 1985). Twenty male Sprague-Dawley rats (Rattus norvegicus) (ZivigMiller) were housed and fed according to protocols approved by the Animal Care Committee of the University of Washington. Rats were anesthetized and euthanized, and tissues were processed as previously described.2 All RNA was isolated as previously described.5

I

Results General Comparison of Human and Macaque Array Hybridization Samples of RNA from 5 pairs of macaque aortas and vena cavas were used to probe a cDNA array containing 4048 human genes. Genes were defined as expressed if hybridization was ⱖ1.5 times the average background (before normalization in Pathways software to overall hybridization strength). The informativeness of macaque probes on human filters in terms of the total number of positive responding spots was a concern, because all arrayed targets consist of ⬇1000 bp of the 3⬘ end of each gene and contain complete 3⬘ untranslated regions, areas of low interspecies homology. To quantify general hybridization patterns, we hybridized cDNA probes from human aorta (n⫽3) and from macaque aorta and vena cava (both n⫽5), then compared the average number of genes expressed by the above definition. The number of genes detected in human was ⬇200 to 500 greater than in macaque (human aorta, 3855⫾115 [mean⫾SEM]; macaque

Lowest Sample Expression Above AB ⱖ5

II

2 to 5

III

1.5 to 2

IV

⬍1.5

Ratio Classes Defined by the Lowest Animal Ratio Ratio Class

Lowest Animal Ratio of Aorta/Vena Cava

A

ⱖ1.5

B

1.3 to 1.5

aorta, 3367⫾115; and macaque vena cava, 3635⫾181), indicating possible species specific differences in 3⬘ noncoding domains or, possibly, species differences in expression.

Identification of Differentially Expressed Genes To have objective criteria for identifying consistently differentially expressed genes, we established categories of ratio and array hybridization level for this study (Table 1). After analyzing the data for the 5 independent macaque comparisons, there were 68 genes that fit these criteria. All were higher in the aorta (Table 2). Of these, only RGS5 was expressed ⬎10-fold, at 46.5⫾12.6 SEM. RGS5 had the highest level of differential expression in each macaque. Levels were undetectable, or barely detectable, in vena cava, suggesting an “all-or-nothing” expression. The array hybridization images for class IA genes are shown in Figure 1A. Expression ratios of the 68 differentially expressed genes are graphed in Figure 1B.

Northern Analysis and In Situ Hybridization We verified the array-based quantification by further examining the expression of the 2 most highly expressed genes, RGS5 and NMMHC-B. First, we examined macaque aorta and vena cava by Northern analyses (Figure 2A). RGS5 was most highly expressed in macaque aorta, followed by carotid artery, heart, and vena cava, with little expression detected in skeletal muscle or liver. NMMHC-B was most highly expressed in aorta, followed by vena cava and then the carotid artery. After normalization to 28S rRNA hybridization (Table 3), the Northern blot aorta-to-vena cava ratio was quantified as 45-fold for RGS5 and 5-fold for NMMHC-B, very close to the mean array values. Macaque message sizes were consistent with the published human sizes,6 with 2 major RGS5 bands at ⬇7.0 and ⬇2.1 kb and 1 NMMHC-B band at ⬇7.5 kb.7 We further examined RGS5 because of its large differential expression ratio. To compare human aorta RGS5 message levels to heart, the highest-expressing human tissue examined in the literature,6 we probed a Clontech Northern blot containing RNA samples of aorta, fetal and adult heart, and 5 distinct heart regions (Figure 2B). RGS5 message levels in

Downloaded from http://circres.ahajournals.org/ by guest on July 26, 2015

Adams et al

Expression Profiles of Aorta and Vena Cava

625

Figure 1. A, Array hybridization images of class IA genes for each macaque (1–5). Ao indicates aorta; VC, vena cava. B, Graphs of expression levels of all 68 consistent aorta marker genes, sorted by ratio. Larger graph presents all genes except RGS5 sorted in descending ratio order, listed as 2 through 68 (accession Nos., A490477, AA459308, AA486321, AA485748, R66310, H06516, H95960, AA487623, AA875933, H79047, AA489331, AA040703, R50953, AA490473, R22219, N53065, T96082, N64384, H25917, AA504461, AA873159, H09959, AA287323, T69450, R43778, H46554, AA777192, AA668681, T72076, AA845178, R75820, AA863149, R83876, H39192, AA127014, R76437, H59204, AA454160, R15111, AA630507, AA169645, AA455652, H23979, AA873152, AA706968, AA453750, R83837, AA644092, AA148548, AA877082, AA489201, AA448711, H17943, AA706987, AA188179, N73827, R92994, T73556, AA490945, AA421518, W51985, R48320, N32768, AA488406, AA599187, N79534, and R63811). Smaller, inset graph presents ratios for RGS5 (AA668470) and the next 9 genes in descending ratio order, listed as 1 through 10. PAM indicates peptidylglycine ␣-amidating mono-oxygenase; SPARC, secreted protein acidic and rich in cysteine.

Downloaded from http://circres.ahajournals.org/ by guest on July 26, 2015

626

Circulation Research TABLE 2.

September 29, 2000

Class I to III Differentially Expressed Genes, All Aortic Accession no.

Mean R⫾SEM

Mean Int Ao

Mean Int VC

RGS5

AA668470

46.5⫾12.6

18 589

517

NMMHC-B

AA490477

7.5⫾2.3

5147

898

Vimentin

AA486321

4.9⫾1.4

18 440

4812

Peptidylglycine ␣-amidating mono-oxygenase

R66310

4.2⫾1.8

3113

1080

SPARC/osteonectin

H95960

3.8⫾0.9

12 178

3813

␣2-Macroglobulin, field 2*

H06516

3.9⫾2.2

3980

1945

␣2-Macroglobulin, field 1*

H06516

3.5⫾1.7

3330

1555

FK506-binding protein precursor

R75820

2.0⫾0.3

3134

1783

Thromboxane-A synthase

R76437

2.0⫾0.3

2361

1152

Pregnancy specific ␤1 glycoprotein 5

W51985

1.7⫾0.1

1656

1022

Pregnancy-specific ␤1-glycoprotein 7

N32768

1.6⫾0.1

2779

1786

Nel-related protein

N79534

1.5⫾0.1

2400

1535

Fibromodulin

AA485748

4.6⫾0.6

1685

388

Cardiac gap junction protein

AA487623

3.7⫾0.7

1545

487

Extracellular protein S1-5

AA875933

3.7⫾0.4

2324

666

H79047

3.3⫾0.5

961

328

AA040703

2.8⫾0.3

1006

377

Proteasome subunit HsC7-I

N53065

2.6⫾0.6

777

327

Class IA

Class IB

Class IIA

Insulin-like growth factor binding protein 2 (36 kDa) Profilin 2 Integrin, ␣X

N64384

2.5⫾0.6

1906

910

Transcription factor AREB6

H46554

2.2⫾0.3

1376

664

Proteasome subunit XAPC7

AA863149

2.0⫾0.2

1511

826

H39192

2.0⫾0.2

1065

502

AA127014

2.0⫾0.2

1657

903

AA490473

2.7⫾1.2

1668

1017

H25917

2.5⫾0.4

2613

1162

Low-density lipoprotein receptor precursor

AA504461

2.4⫾0.4

1423

576

DNA-repair protein complementing XP-C cells

AA287323

2.3⫾0.8

1918

1201

APEG-1

T69450

2.3⫾0.6

1131

538

Plasma-cell membrane glycoprotein PC-1

R43778

2.3⫾0.4

818

386

DNA-directed RNA polymerase II (14.5 kDa)

AA777192

2.2⫾0.3

1846

809

Cell division cycle 42 (GTP binding protein, 25 kDa)

AA668681

2.1⫾0.3

1832

1054

Carboxypeptidase A1

AA845178

2.0⫾0.4

1033

609

R83876

2.0⫾0.3

700

331

AA454160

1.9⫾0.3

2836

1628

BMK1 ␣ kinase Mitochondrial NADH dehydrogenase-ubiquinone Fe-S protein 8, 23-kDa subunit precursor Class IIB Protein phosphatase 2A (␤ type) Arp2/3 protein complex subunit p34-Arc

ATP-binding cassette protein mRNA (06B09 clone) Superoxide dismutase 3, extracellular RNA-binding protein CUG-BP/hNab50

R15111

1.9⫾0.2

1050

532

AA630507

1.9⫾0.2

1753

1049

AA455652

1.9⫾0.2

817

462

H23979

1.9⫾0.2

759

441

Complement C3 precursor

AA873152

1.9⫾0.3

1273

791

M-phase phosphoprotein (mpp5)

AA706968

1.8⫾0.2

1650

897

cAMP-dependent protein kinase (regulatory subunit type I) Arginyl-tRNA synthetase OX-2 membrane glycoprotein precursor

V-yes-1 oncogene homologue

R83837

1.8⫾0.2

738

390

Nucleoside diphosphate kinase A

AA644092

1.8⫾0.2

794

409

Zinc finger protein 10 (KOX 1)

AA877082

1.8⫾0.3

1194

772

PHAPI2b protein

AA489201

1.8⫾0.3

929

574

N73827

1.8⫾0.3

1292

832

SA

Downloaded from http://circres.ahajournals.org/ by guest on July 26, 2015

Adams et al TABLE 2.

Expression Profiles of Aorta and Vena Cava

627

Continued Accession no.

Mean R⫾SEM

Mean Int Ao

Mean Int VC

R48320

1.7⫾0.1

3380

2171

AA599187

1.6⫾0.1

1013

697

R63811

1.5⫾0.0

1219

804

Elastin

AA459308

5.0⫾1.4

1515

381

dsRNA adenosine deaminase DRADA2b

AA489331

3.0⫾0.5

782

313

Low-Mr GTP-binding protein (RAB31)

T96082

2.6⫾0.2

916

369

Choline kinase

H09959

2.3⫾0.2

1840

777

AA873159

2.7⫾1.0

586

241

Eph-like receptor tyrosine kinase hEphB1c Phosphoglycerate kinase 1 Ribosomal protein S25 Class IIIA

Class IIIB B-lymphocyte serine/threonine protein kinase Phosphoethanolamine cytidylyltransferase

T72076

2.7⫾0.9

705

308

Apolipoprotein CI

H59204

2.4⫾0.4

929

474

Plasma retinol binding protein precursor

AA169645

2.0⫾0.4

460

272

Cdc6-related protein

AA148548

2.0⫾0.3

645

334

Splicing factor SRp55-2

H17943

1.9⫾0.2

456

249

Human D9 splice variant A mRNA

T73556

1.8⫾0.3

611

336

Fatty acid binding protein

AA453750

1.8⫾0.2

685

397

Zinc finger protein, clone:RES4-26

AA448711

1.8⫾0.2

665

361

AU RNA-binding protein/enoyl–coenzyme A hydratase

AA706987

1.8⫾0.2

696

364

UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferase (T1)

AA188179

1.8⫾0.2

752

440

Arp2/3 protein complex subunit p41-Arc

AA490945

1.8⫾0.3

440

268

Secretory carrier-associated membrane protein (SCAMP)

AA488406

1.7⫾0.2

805

442

Clathrin-associated protein

R50953

1.7⫾0.2

654

430

Long-chain fatty acid acyl-coA ligase

R22219

1.7⫾0.1

794

474

AA421518

1.6⫾0.1

487

300

Pre-promegakaryocyte potentiating factor

Ratios are the average of individual macaque aorta (Ao) to vena cava (VC) comparisons (n⫽5). Mean intensity (Int) values are included as a guide for expression levels; individual macaque intensities were used to calculate mean ratios (R). All 68 consistently differentially expressed genes had SEMs less than two-thirds of their average ratios. *Array housekeeping genes.

the aorta were at least 12-fold higher than any region of the heart (Table 4, top). RGS5 message is reported to be highly expressed in mouse heart and very low in liver,8 but has never been examined in rodent vessels. We compared RGS5 expression in rat aorta, carotid artery, and vena cava with that in heart and liver by Northern analysis (Figure 2C). The RGS5 message level was equally high in rat aorta and heart (Table 4, middle) and 2.3 times higher than the vena cava, whereas the carotid artery had lower expression than the vena cava, differing from macaque and human expression. The rat message consisted of a single band the same size as mouse heart (data not shown). The rat and macaque data establish RGS5 as a marker differentiating aorta from carotid artery and aorta from vena cava, but show that there are distinct species differences in expression levels in vessels. Finally, we examined the expression of RGS5 on a human dot blot (Clontech) containing RNAs from 47 adult and 7 fetal tissues (Figure 3). Twentynine tissues had higher expression than skeletal muscle, a moderate-expression tissue6; the top 15 are quantified in Table 4, bottom. The 6 highest-expressing tissues in order (beginning with the highest) were aorta, small intestine,

stomach, heart (4.6-fold lower than the aorta), adrenal gland, and fetal heart. To localize the RGS5 transcript in vivo, we performed in situ hybridization on macaque aorta and vena cava (Figure 4). Hybridization in the aorta from the antisense probe was seen uniformly throughout the media, well above the sense probe background level, whereas the vena cava antisense probe hybridization was very similar to the background sense hybridization. No hybridization was seen in the endothelium. The presence of endothelium was confirmed by examination of hematoxylin and eosin–stained sections (Figures 4a and 4b) and by immunohistochemistry using Von Willebrand factor antibody staining (data not shown). One possible explanation of differential expression of RGS5 could have been the presence of RNA from vasa vasorum, as they are present in arteries but not veins.9 This remains a formal possibility for other aorta differentially expressed genes.

Discussion Our analysis of aorta versus vena cava reveals 68 genes that mark the aorta by overexpression, including 2 transcripts

Downloaded from http://circres.ahajournals.org/ by guest on July 26, 2015

628

Circulation Research

September 29, 2000 TABLE 3. Normalized Expression Levels of RGS5 and NMMHC-B in Macaque Tissues

Tissue

Ratio Tissue/Skeletal Muscle

3

Aorta

184.3

4

Carotid

14.5

1

Heart

5.1

5

Vena cava

4.1

6

Liver

1.2

2

SkM

1.0

Lane

Tissue

NMMHC-B Normalized/Liver

3

Aorta

370.7

5

Vena cava

71.0

4

Carotid

55.6

1

Heart

1.5

6

Liver

1.3

2

SkM

1.0

Lane

RGS5 (top) and NMMHC-B (bottom) hybridizations were normalized, first to 28S rRNA hybridization to adjust to loading differences, then to normalized skeletal muscle (Skm) hybridization. All hybridizations were quantified from phosphor imager scans.

Figure 2. Northern hybridizations of RGS5 and NMMHC-B. Blots were stripped and reprobed with 28S rRNA hybridizations in panels A and C and with G3PDH in panel B for loading quantification. A, Macaque tissues. H indicates heart; Skm, skeletal muscle; Ao, aorta; CA, carotid artery; VC, vena cava; and Lv, liver. B, Human aorta and heart tissue (Clontech). FH indicates fetal heart; H, heart; Ao, aorta; AP, apex of heart; LA, left atrium; RA, right atrium; LV, left ventricle; and RV, right ventricle. C, Rat tissues. Abbreviations as in panels A and B. Panels A and C, Locations of 28S and 18S rRNAs are shown on the left at 4.9 and 1.7 kb, respectively.

previously identified with the aorta, aortic preferentially expressed gene 1 (APEG-1)10 and elastin.11,12 It is important to stress that we have examined 5 individual animals to produce this data set, eliminating false-positive results due to animal variation in elevations of gene expression or random measurement error. Genes were only declared differentially expressed if mean expression ratios were ⬎1.5fold. All 5 individual animal ratios were ⬎1.3-fold; SEMs were less than two-thirds (given that a gene had to be differential in all animals by 1.3-fold, this fact in effect ensured that the SEM was a measurement of variation in differential ratio). All differential genes hybridized consistently above 1.5-fold the average background. Although an expression ratio of 1.3 is small, in an analysis of message expression using cDNA arrays, Geiss et al13 demonstrated that even ratios as low as 1.3 to 1.4 on the array can be statistically meaningful for genes with low SE. They showed that Northern blots produce higher ratios than do arrays. Thus, we expect that most of the 68 genes are truly differentially expressed. We have shown previously that this Research Genetics cDNA array produces values that agree with independent methods of expression measurement.14 Moreover, 44 of the 68 genes were expressed at ratios of 2-fold or greater. The most striking aspect of the data set is the expression of one gene, RGS5. RGS5 is remarkably higher in average ratio than any other transcript compared between aorta and vena cava. Nevertheless, we suggest that it is the data set as a whole that begins the molecular description of the arterial and venous phenotypes. Golub et al15 have recently compared acute lymphoblastic leukemia with acute myeloid leukemia. Using more sophisticated analysis than currently available to us, these authors found a set of only 50 genes out of 6817 that

Downloaded from http://circres.ahajournals.org/ by guest on July 26, 2015

Adams et al

Expression Profiles of Aorta and Vena Cava

629

TABLE 4. Quantification of RGS5 Expression in the Human Heart and Human Tissues

Lane

Tissue

Ratio RGS5/Tissue/ Fetal Heart*

3

Aorta

31.5

5

Left atrium

2.6

2

Heart

2.5

8

Right ventricle

2.4

7

Left ventricle

2.1

6

Right atrium

1.6

4

Heart apex

1.4

1

Fetal heart

1.0

Lane

Tissue

Ratio Tissue/Skeletal Muscle†

1

Heart

3.1

3

Aorta

3.0

5

Vena cava

1.3

2

Skeletal muscle

1.0

4

Carotid artery

0.2

6

Liver

0.2

Tissue

Ratio Tissue/Skeletal Muscle‡

C2

Aorta

23.0

E3

Small intestine

6.7

C8

Stomach

6.6

C1

Heart

5.0

D5

Adrenal gland

3.8

G1

Fetal heart

3.7

D6

Thyroid gland

3.6

F1

Appendix

3.0

Blot Location

C5

Bladder

2.5

C7

Prostate

2.5

G7

Fetal lung

2.1

B6

Subthalamic nucleus

2.0

E1

Kidney

1.9

E7

Lymph node

1.8

A8

Medulla oblongata

1.8

*Quantified Northern analysis of RGS5 in human aorta and heart. RGS5 hybridizations were normalized first to GAPDH hybridizations for loading adjustment and then to the fetal heart hybridization level. †Quantified expression of RGS5 in rat tissues. RGS5 levels were normalized first to 28S rRNA levels to adjust to loading differences and then to normalized skeletal muscle RGS5 hybridization. Hybridizations were quantified from phosphor imager scans. ‡The 15 tissues with highest RGS5 message levels. RGS5 hybridizations were normalized first to ubiquitin hybridization to adjust to loading differences and then to normalized skeletal muscle RGS5 hybridization. Hybridizations were quantified from phosphor imager scans.

could be used as a pattern to distinguish between these 2 generally related neoplasms. Similarly, the 68 genes reported here represent a consistently differentially expressed set that

Figure 3. Dot blot of human tissue poly(A) RNAs (89 to 514 ng each) (Clontech). A1 through A8, respectively, Whole brain, amygdala, caudate nucleus, cerebellum, cerebral cortex, frontal lobe, hippocampus, and medulla oblongata. B1 through B8, Occipital lobe, putamen, substantia nigra, temporal lobe, thalamus, subthalamic nucleus, spinal chord, and blank. C1 through C8, Heart, aorta, skeletal muscle, colon, bladder, uterus, prostate, and stomach. D1 through D8, Testis, ovary, pancreas, pituitary gland, adrenal gland, thyroid gland, salivary gland, and mammary gland. E1 through E8, Kidney, liver, small intestine, spleen, thymus, peripheral leukocyte, lymph node, and bone marrow. F1 through F4, Appendix, lung, trachea, and placenta. F5 through F8, Blank. G1 through G8 (fetal tissue), brain, heart, kidney, liver, spleen, thymus, lung, and blank. H1 through H8 (controls), 100 ng yeast total RNA, 100 ng yeast tRNA, 100 ng Escherichia coli rRNA, 100 ng E. coli DNA, 100 ng poly r(A), 100 ng human Cot-1 DNA, 100 ng human DNA, and 500 ng human DNA.

together constitute a phenotype for aortic and vena cava smooth muscle. A few of the genes in this set of 68 have been examined previously in vascular smooth muscle contexts. Four genes showing aortic expression, APEG-1, elastin, NMMHC-B, and vimentin have previously been identified as markers of the intima. APEG-1 has been shown in rats to be differentially expressed between aorta and vena cava and downregulated after carotid balloon injury.16 NMMHC-B, also called SMemb, has been previously reported to be differentially expressed in vivo in vascular smooth muscle, in human normal intima and intima of coronary arteries after angioplasty,17 and in 2 rabbit models of intima formation.18 Vimentin has been shown to be expressed in balloon-injured rat carotid neointima, whereas another intermediate filament, desmin, disappears19 and is expressed in “synthetic” smooth muscle cells in culture.20 Elastin has also been shown to be a marker between rat smooth muscle cell types of different embryological origin and is upregulated after carotid artery balloon injury in vivo.21 Secreted protein acidic and rich in cysteine, seen here as an arterial marker, has also been shown to be differential between smooth muscle subtypes in vitro but is not modulated in expression after carotid artery balloon injury.22

Downloaded from http://circres.ahajournals.org/ by guest on July 26, 2015

630

Circulation Research

September 29, 2000

Figure 4. Composite photomicrograph showing cross sections of macaque aorta (a, c, and e) and vena cava (b, d, and f) hybridized with 35S-labeled riboprobes to human RGS5. Hematoxylin counterstain provides orientation of the aortic wall (a, lumen at right, adventitia at left) and vena cava (b, lumen and thin media at center with surrounding adventitia). c and d, Same sections viewed with darkfield microscopy show intense hybridization of the antisense probe to RGS5 message in the aorta (c) but minimal hybridization in vena cava (d). Adjacent sections (e and f) were hybridized with sense probe to control for nonspecific background. Magnification, ⫻200, all panels.

As for the remainder of the data set, it is intriguing to see that only a minority of these genes has been of traditional interest to vascular biologists, suggesting that there are many novel aspects of arteries and veins yet to functionally characterize. Among the aortic genes not previously studied in vascular biology, the high level of differential expression of RGS5 singles it out for our attention. RGS5 is currently an “orphan” G-protein pathway regulator, and its exact functions need to be characterized. The core function of RGS proteins is to regulate G-protein signal strength and duration by binding to and dephosphorylating G␣ subunits.8,22–26 Additionally, there is evidence for both inhibitor and effector (signal integration) G-protein signaling from 2 multidomain RGS proteins, p115RhoGEF27,28 and PDZRhoGEF,29 suggesting that the role RGS5 plays in arteries may be quite complex. RGS4, RGS5, RGS8, and RGS16 are among the shortest RGS proteins and share homology outside their RGS domains, suggesting they constitute a subfamily, with RGS4 and RGS5 being the most similar.30 –33 In transgenic mouse overexpression studies, high levels of RGS4 inhibit cardiac ventricular hypertrophy in response to cardiac pressure overload after thoracic aortic constriction.34 These remodeling processes are required for adaptation and survival of transgenics. Also, in transgenic mouse overexpression studies of G␣q (a binding partner of RGS4), high expression levels cause cardiac hypertrophy.35,36 These data for RGS4 overexpression in heart, and our expression data for RGS5 in vessels, suggest that RGS5 could

be a counterpart in arteries to the RGS4-regulated system of balance between activation and deactivation of G protein– induced hypertrophic signaling in the heart. Arteries, like hearts, adapt to increased tension by remodeling, producing a thick vessel wall able to generate a stronger contractile response.37 Considering the known role for RGS4 in heart accommodation to increased pressure and its structural similarities to RGS5, we speculate that RGS5 might also be involved in the adaptation of arteries to normal and pathological pressure changes. However, until a transgenic overexpression of RGS5 is performed and tested after aortic constriction, there will be no direct evidence to determine whether RGS5 is regulating pressure-induced hypertrophy in arteries. In summary, our analysis of 4048 genes, combined with the ⬇20 000 genes expected to be expressed in an average tissue,38 suggests there will be roughly 340 genes marking the aortic versus the venous phenotype, including perhaps 4 genes of the ratio magnitude of RGS5. It is intriguing that we found no genes overexpressed in vein versus artery. One possible explanation is that the arterial phenotype is built on a basic veinlike phenotype. We speculate that the most highly overexpressed aortic gene, RGS5, could play a role in control of transcription of the entire aortic phenotype and, based on homology to other RGS proteins, may play an aortic-specific role in regulation of responses to G-protein–mediated vasoactive signals.

Acknowledgments This work was supported by NIH Grants RO1HL 57557, RO1HL58083, and PO1HL03174. We thank Isa Werny, Jonathan McBride, and Colette Norby-Slycord for their excellent technical assistance. We thank Dr Linda Harris for critical reading of the manuscript and many helpful discussions on cDNA array issues, as well as Dr Roger Bumgarner and Dr Eileen Mulvihill for useful discussions on cDNA array analysis methodologies.

References 1. Giachelli CM, Majesky MW, Schwartz SM. Developmentally regulated cytochrome P450IA1 expression in cultured rat vascular smooth muscle cells. J Biol Chem. 1991;266:3981–3986. 2. Adams LD, Lemire JM, Schwartz SM. A systematic analysis of 40 random genes in cultured vascular smooth muscle subtypes reveals a heterogeneity of gene expression and identifies the tight junction gene zonula occludens 2 as a marker of epithelioid “pup” smooth muscle cells and a participant in carotid neointimal formation. Arterioscler Thromb Vasc Biol. 1999;19:2600 –2608. 3. Geary RL, Nikkari ST, Wagner WD, Williams JK, Adams MR, Dean RH. Wound healing: a paradigm for lumen narrowing after arterial reconstruction. J Vasc Surg. 1998;27:96 –106. 4. Geary RL, Williams JK, Golden D, Brown DG, Benjamin ME, Adams MR. Time course of cellular proliferation, intimal hyperplasia, and remodeling following angioplasty in monkeys with established atherosclerosis: a nonhuman primate model of restenosis. Arterioscler Thromb Vasc Biol. 1996;16:34 – 43. 5. Giachelli CM, Bae N, Lombardi DM, Majesky MW, Schwartz SM. Molecular cloning and characterization of 2B7, a rat mRNA which distinguishes smooth muscle cell phenotypes in vitro and is identical to osteopontin (secreted phosphoprotein I, 2aR). Biochem Biophys Res Commun. 1991;177:867– 873. 6. Seki N, Sugano S, Suzuki Y, Nakagawara A, Ohira M, Muramatsu M, Saito T, Hori T. Isolation, tissue expression, and chromosomal assignment of human RGS5, a novel G-protein signaling regulator gene. J Hum Genet. 1998;43:202–205. 7. Simons M, Wang M, McBride OW, Kawamoto S, Yamakawa K, Gdula D, Adelstein RS, Weir L. Human nonmuscle myosin heavy chains are

Downloaded from http://circres.ahajournals.org/ by guest on July 26, 2015

Adams et al

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

encoded by two genes located on different chromosomes. Circ Res. 1991;69:530 –539. Chen C, Zheng B, Han J, Lin SC. Characterization of a novel mammalian RGS protein that binds to G␣ proteins and inhibits pheromone signaling in yeast. J Biol Chem. 1997;272:8679 – 8685. Barger AC, Beeuwkes R III, Lainey LL, Silverman KJ. Hypothesis: vasa vasorum and neovascularization of human coronary arteries: a possible role in the pathophysiology of atherosclerosis. N Engl J Med. 1984;310: 175–177. Hsieh CM, Yoshizumi M, Endege WO, Kho CJ, Jain MK, Kashiki S, de los SR, Lee WS, Perrella MA, Lee ME. APEG-1, a novel gene preferentially expressed in aortic smooth muscle cells, is down-regulated by vascular injury. J Biol Chem. 1996;271:17354 –17359. Sauvage M, Hinglais N, Mandet C, Badier C, Deslandes F, Michel JB, Jacob MP. Localization of elastin mRNA and TGF-␤1 in rat aorta and caudal artery as a function of age. Cell Tissue Res. 1998;291:305–314. Belknap JK, Grieshaber NA, Schwartz PE, Orton EC, Reidy MA, Majack RA. Tropoelastin gene expression in individual vascular smooth muscle cells: relationship to DNA synthesis during vascular development and after arterial injury. Circ Res. 1996;78:388 –394. Geiss GK, Bumgarner RE, An MC, Agy MB, van’t Wout AB, Hammersmark E, Carter VS, Upchurch D, Mullins JI, Katze MG. Large-scale monitoring of host cell gene expression during HIV-1 infection using cDNA microarrays. Virology. 2000;266:8 –16. Schaub FJ, Han DK, Conrad Liles W, Adams LD, Coats SA, Ramachandran RK, Seifert RA, Schwartz SM, Bowen-Pope DF. Fas/ FADD-mediated activation of a specific program of inflammatory gene expression in vascular smooth muscle cells. Nat Med. 2000;6:790 –796. Golub TR, Slonim DK, Tamayo P, Huard C, Gaasenbeek M, Mesirov JP, Coller H, Loh ML, Downing JR, Caligiuri MA, Bloomfield CD, Lander ES. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science. 1999;286:531–537. Hsieh CM, Yoshizumi M, Endege WO, Kho CJ, Jain MK, Kashiki S, de los SR, Lee WS, Perrella MA, Lee ME. APEG-1, a novel gene preferentially expressed in aortic smooth muscle cells, is down-regulated by vascular injury. J Biol Chem. 1996;271:17354 –17359. Aikawa M, Sakomura Y, Ueda M, Kimura K, Manabe I, Ishiwata S, Komiyama N, Yamaguchi H, Yazaki Y, Nagai R. Redifferentiation of smooth muscle cells after coronary angioplasty determined via myosin heavy chain expression. Circulation. 1997;96:82–90. Kuro-o M, Nagai R, Nakahara K, Katoh H, Tsai RC, Tsuchimochi H, Yazaki Y, Ohkubo A, Takaku F. cDNA cloning of a myosin heavy chain isoform in embryonic smooth muscle and its expression during vascular development and in arteriosclerosis. J Biol Chem. 1991;266:3768 –3773. Gabbiani G, Rungger-Brandle E, de Chastonay C, Franke WW. Vimentin-containing smooth muscle cells in aortic intimal thickening after endothelial injury. Lab Invest. 1982;47:265–269. Palmberg L, Sjolund M, Thyberg J. Phenotype modulation in primary cultures of arterial smooth-muscle cells: reorganization of the cytoskeleton and activation of synthetic activities. Differentiation. 1985; 29:275–283. Majesky MW, Giachelli CM, Schwartz SM, Reidy MA. Rat carotid neointimal smooth muscle cells re-express a developmentally regulated phenotype during repair of arterial injury. Circ Res. 1992;71:759 –768.

Expression Profiles of Aorta and Vena Cava

631

22. Watson N, Linder ME, Druey KM, Kehrl JH, Blumer KJ. RGS family members: GTPase-activating proteins for heterotrimeric G-protein ␣-subunits. Nature. 1996;383:172–175. 23. Druey KM, Blumer KJ, Kang VH, Kehrl JH. Inhibition of G-proteinmediated MAP kinase activation by a new mammalian gene family. Nature. 1996;379:742–746. 24. Nekrasova ER, Berman DM, Rustandi RR, Hamm HE, Gilman AG, Arshavsky VY. Activation of transducin guanosine triphosphatase by two proteins of the RGS family. Biochemistry. 1997;36:7638 –7643. 25. Doupnik CA, Davidson N, Lester HA, Kofuji P. RGS proteins reconstitute the rapid gating kinetics of G␤␥-activated inwardly rectifying K⫹ channels. Proc Natl Acad Sci U S A. 1997;94:10461–10466. 26. Snow BE, Betts L, Mangion J, Sondek J, Siderovski DP. Fidelity of G protein ␤-subunit association by the G protein ␥-subunit-like domains of RGS6, RGS7, and RGS11. Proc Natl Acad Sci U S A. 1999;96: 6489 – 6494. 27. Peehl DM, Sellers RG. Induction of smooth muscle cell phenotype in cultured human prostatic stromal cells. Exp Cell Res. 1997;232:208 –215. 28. Pomie`s P, Louis HA, Beckerle MC. CRP1, a LIM domain protein implicated in muscle differentiation, interacts with ␣-actinin. J Cell Biol. 1997;139:157–168. 29. Qian J, Hendrix M, Larsen WJ, Dorn GW, Lessard JL. Establishment and characterization of a conditionally immortalized smooth muscle/ myometrial-like cell line. Mol Reprod Dev. 1997;47:284 –294. 30. Panetta R, Guo Y, Magder S, Greenwood MT. Regulators of G-protein signaling (RGS) 1 and 16 are induced in response to bacterial lipopolysaccharide and stimulate c-fos promoter expression. Biochem Biophys Res Commun. 1999;259:550 –556. 31. Hepler JR. Emerging roles for RGS proteins in cell signalling. Trends Pharmacol Sci. 1999;20:376 –382. 32. Chen C, Seow KT, Guo K, Yaw LP, Lin SC. The membrane association domain of RGS16 contains unique amphipathic features that are conserved in RGS4 and RGS5. J Biol Chem. 1999;274:19799 –19806. 33. Srinivasa SP, Bernstein LS, Blumer KJ, Linder ME. Plasma membrane localization is required for RGS4 function in Saccharomyces cerevisiae. Proc Natl Acad Sci U S A. 1998;95:5584 –5589. 34. Rogers JH, Tamirisa P, Kovacs A, Weinheimer C, Courtois M, Blumer KJ, Kelly DP, Muslin AJ. RGS4 causes increased mortality and reduced cardiac hypertrophy in response to pressure overload. J Clin Invest. 1999;104:567–576. 35. D’Angelo DD, Sakata Y, Lorenz JN, Boivin GP, Walsh RA, Liggett SB, Dorn GW. Transgenic G␣q overexpression induces cardiac contractile failure in mice. Proc Natl Acad Sci U S A. 1997;94:8121– 8126. 36. Adams JW, Sakata Y, Davis MG, Sah VP, Wang Y, Liggett SB, Chien KR, Brown JH, Dorn GW. Enhanced G␣q signaling: a common pathway mediates cardiac hypertrophy and apoptotic heart failure. Proc Natl Acad Sci U S A. 1998;95:10140 –10145. 37. Mulvany MJ, Hansen PK, Aalkjaer C. Direct evidence that the greater contractility of resistance vessels in spontaneously hypertensive rats is associated with a narrowed lumen, a thickened media, and an increased number of smooth muscle cell layers. Circ Res. 1978;43:854 – 863. 38. Lewin B. Genes V. New York, NY: Oxford University Press; 1994.

Downloaded from http://circres.ahajournals.org/ by guest on July 26, 2015

A Comparison of Aorta and Vena Cava Medial Message Expression by cDNA Array Analysis Identifies a Set of 68 Consistently Differentially Expressed Genes, All in Aortic Media Lawrence D. Adams, Randolph L. Geary, Bruce McManus and Stephen M. Schwartz Circ Res. 2000;87:623-631 doi: 10.1161/01.RES.87.7.623 Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 2000 American Heart Association, Inc. All rights reserved. Print ISSN: 0009-7330. Online ISSN: 1524-4571

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circres.ahajournals.org/content/87/7/623

Data Supplement (unedited) at: http://circres.ahajournals.org/content/suppl/2000/09/26/87.7.623.DC1.html

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation Research is online at: http://circres.ahajournals.org//subscriptions/

Downloaded from http://circres.ahajournals.org/ by guest on July 26, 2015