This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Adams, L. D. Articles by Schwartz, S. M. Search for Related Content PubMed PubMed Citation Articles by Adams, L. D. Articles by Schwartz, S. M. Related Collections Cell signalling/signal transduction Gene expression Smooth muscle proliferation and differentiation

(Arteriosclerosis, Thrombosis, and Vascular Biology. 1999;19:2600.)
© 1999 American Heart Association, Inc.

Vascular Biology

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

Lawrence D. Adams; Joan M. Lemire; Stephen M. Schwartz

From the Department of Pathology, University of Washington, Seattle, Wash.

Correspondence to Lawrence D. Adams, PhD, University of Washington, Department of Pathology, Vascular Biology/Box 357335, 1959 NE Pacific Street, Seattle WA 98195-7335. E-mail ladams{at}u.washington.edu

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Abstract—An accumulation of evidence suggests that vascular smooth muscle is composed of cell subpopulations with distinct patterns of gene expression. Much of this evidence has come from serendipitous discoveries of genes marking phenotypically distinct aortic cultures derived from 12-day-old and 3-month-old rats. To identify more systematic differences, we isolated 40 genes at random from libraries of these 2 cultures and examined message expression patterns. To determine consistency of differential expression, we measured mRNA levels in 4 sets of cultures in 6 phenotypically distinct aortic cell clones and in balloon injured rat carotid arteries to determine the relevance of these differences in vitro to in vivo biology. The following 5 consistently differentially expressed genes were identified in vitro: zonula occludens 2 (ZO-2); peroxisome proliferator-activated receptor (PPAR); secreted protein, acidic and rich in cysteine (SPARC); 1(I)collagen; and A2, an uncharacterized gene. We examined these 5 clones during carotid artery injury and an inconsistently differentially expressed clone Krox-24 because, as an early response transcription factor, it could be involved in the injury response. PPAR, A2, and Krox-24 mRNAs were upregulated during the day after injury. ZO-2 and 1(I)collagen messages were modulated for up to a month, whereas SPARC message showed no consistent change. An analysis of ZO-2 and other tight junction genes indicates that tight junctions may play a role in smooth muscle biology. These data suggest that a systematic analysis of these libraries is likely to identify a very large number of differentially expressed genes. ZO-2 is particularly intriguing both because of this tight junction gene’s pattern of prolonged over-expression after injury and because of its potential role in determining the distinctive epithelioid phenotype of smooth muscle cells identified in rat and other species.

Key Words: vascular smooth muscle • neointima • tight junction • zonula occludens 2

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Smooth muscle cultures derived from arteries of 12-day-old pup and 3-month-old adult Wistar-Kyoto rats have several unique properties suggesting that they represent subtypes of the smooth muscle cell (SMC).^{1 2 3 4} Similar differences have been seen between cultures of human arterial SMCs.⁵ Correlating with these morphological and proliferative differences are differential RNA expression patterns for several genes identified either by guesses or related differences in function or by subtraction hybridization.^{6 7 8 9 10} Several of these genes’ message levels are also upregulated during formation of rat neointima after balloon injury and in human atherosclerotic plaque, suggesting that differences between the "pup" and "adult" phenotypes may provide clues to the nature of the intima. Examples include proteases, mitogens, and matrix proteins.^{7 11 12 13 14 15 16}

In this study, we systematically examined the expression of a random set of 40 genes chosen from a pup and an adult cDNA library. Genes were first identified by single pass sequencing. Next we examined differential patterns of expression between the pup and adult SMCs using dot blot and Northern analysis to identify candidates for further study. If a gene did not have more than 1.5-fold expression difference between culture types by Northern analysis, it was discarded from further examination (with the exception of Krox-24). To determine expression consistency, we examined the expression of those selected genes in 3 additional culture sets, 1 being a set at 3 different culture growth states. Those genes with consistent differential expression in vitro were tested for modulation during balloon injury to the carotid arteries as an assay for involvement in vessel injury response.

Five genes were consistently differentially expressed in vitro, including the tight junction gene ZO-2. Four of these genes were modulated after balloon injury, as well as Krox-24. Of these genes, ZO-2 was particularly intriguing because it is tight junction specific and is a member of the family of membrane associated guanylate kinase homologues (MAGUKs),¹⁷ a class involved in organizing the presentation of transmembrane proteins^{18 19 20 21 22} and possibly involved in cell signaling,²³ that has not previously been found in smooth muscle. Further analysis of other tight junction genes in the injured wall as well as in the culture systems suggests that these proteins may have profound implications in the different morphologies and growth patterns of the SMCs analyzed and in the morphogenesis of the intima.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Cell Culture and RNA Isolation
Cultures of rat pup and adult aortic media derived from male Wistar-Kyoto strain rats^{4 9} were used between the fifteenth and twentieth passages. For immunocytochemistry, cells were plated onto Labtek 4-welled plastic chamber slides (Nalge Nunc International). RNA was isolated using TRIzol (Life Technologies GIBCO-BRL).

cDNA Clone Isolation and DNA Sequence Analysis
cDNA libraries were constructed in ZAP (pup) and ZAP II (adult) and manipulated as specified (Stratagene). Plasmid cultures and purifications were performed using Qiagen kits and protocols. Plasmids were sequenced using either the fmole cycle sequencing kit (Promega) or the ABI Dye Terminator Cycle Sequencing Ready Reaction Kit (Perkin Elmer) with primers designed from Bluescript: 5'AGCGGATAACAATTTCACACAGGA 3' (reverse) and 5' CGCCAGGGTTTTCCCAGTCACGAC 3' (forward). Clones A2, A4, and A13 were sequenced further with primers designed from first pass sequences. Clone identities were analyzed using Basic Local Alignment Tool (BLAST).²⁴ All clone sequences were deposited into the National Center for Biotechnology Information GenBank database.

Dot Blot Analysis
Fifty ng of each plasmid was applied to Zeta-probe GT nylon membranes (Biorad). cDNA probes were generated by reverse transcription and random priming as follows: 5 µg total RNA was mixed with 0.5 µL of 50 µmol/L dT15 (Operon) to a total volume of 9.5 µL. This sample was heated to 65°C for 10 minutes, chilled on ice for 3 minutes, and a reaction of 18 µL was set up in the tube with the following additional final composition: 5 U/µL Superscript II reverse transcriptase (Life Technologies Gibco-BRL), 55.5 mmol/L Tris-HCl pH 8.3, 61 mmol/L KCl, 3.3 mmol/L MgCl2, 9 mmol/L DTT, 11 µmol/L dNTPs, and 0.18 µCi/µL of 3000 Ci/mmol [-³²P] dCTP/µL (Dupont-NEN). Reverse transcription proceeded at 37°C for 1 hour. Samples were incubated in NaOH at a final concentration of 0.2 N for 30 minutes, ethanol precipitated, washed in 70% ethanol, and resuspended in 10-µL H₂O. ³²P labeled second strand cDNA was synthesized from first strand cDNA with the Multiprime reaction kit (Amersham).

Prehybridization proceeded for 2 hours at 42°C in the following solution: 50% deionized formamide, 0.75 mol/L NaCl, 50 mmol/L Tris pH 7.4, 1X Denhardt’s Reagent, 1% sodium dodecyl sulfate (SDS), 10% dextran sulfate, and 200 µg/mL salmon sperm DNA. Hybridization proceeded overnight at 42°C with a probe concentration of 1.0x10⁶ cpm/mL. The blots were washed 3 times for 5 minutes at room temperature in 2X SSC, 0.1% SDS and 2 times for 20 minutes in 0.3X SSC, 0.1% SDS at 65°C. Dot blots were scanned and analyzed with a Molecular Dynamics PhosphorImager and software.

Carotid Artery Balloon Injury Model
RNA samples were obtained as previously reported⁸ from carotid arteries of balloon injured²⁵ 3-month-old adult male Sprague-Dawley rats (Zivic-Miller, Allison Park, PA), as approved by the Animal Care Committee of the University of Washington. Specimens were taken from non-reendothelialized areas of the vessel or, for control vessels, after ex vivo removal of the endothelium. Some vessels were separated into neointimal and medial portions.

Northern Analysis
Northerns were performed as described previously.⁶ Probes were synthesized using the Multiprime kit with 50 ng of agarose gel purified cDNA (restriction digested from each cloned plasmid). The -SM actin primer was used as described.⁹ SM22 and ß-actin clones were isolated from the pup library and verified by sequence analysis. Probes for synapse associated protein (SAP90), SAP97, ZO-1, and occludin were generated by reverse transcription polymerase chain reaction (rtPCR) using BRL Superscript II reverse transcriptase and Promega Taq polymerase as specified on 0.5-µg pup SMC RNA using primers specific to the published sequences (X66474, U14950, L14837, and U49185, respectively). Claudin rtPCRs were performed as above for 40 cycles with 0.5-µg pup and adult SMC RNA with primers designed to claudin 1 (rat clone AI171286) and claudin 2 (mouse clone AF072128). Amplification products were gel purified and verified by sequence analysis. All Northern blots were stripped and reprobed with antisense 28S rRNA primer 5' GCAAGAGCGCCAGCTATCCTGAGG 3' for equalizing hybridization values for loading differences between lanes as follows: 10 pmol of 28S rRNA primer was end labeled using 3 µCi[-³²P] ATP at 3000 Ci/mmol (Dupont-New England Nuclear) and 10 U T4 polynucleotide kinase (Promega) in the buffer supplied by the manufacturer and purified by filtration through Microspin G-25 sephadex columns (Pharmacia Biotech). Blots were prehybridized at 42°C for 2 hours. Labeled 28S rRNA primer was added at 1x10⁵ cpm/mL final concentration in the presence of 0.2 µg/mL unlabeled 28S rRNA primer. Autoradiography, PhosphorImager scanning, and analysis were performed as for dot blots. To equalize for loading message levels were calculated as a ratio of message specific hybridization divided by the 28S rRNA specific hybridization value for each lane.

Immunocytochemistry
Primary antibodies to ZO-1, polyclonal Rabbit Anti-ZO-1 61-7300, and negative control primary antibody, Rabbit Anti-Chicken/Turkey IgG 61-3100 were supplied by Zymed Laboratories Inc. Rabbit Anti-ZO-1 61-7300 and Rabbit Anti-Chicken/Turkey IgG 61-3100 were used at 5 µg/mL. Biotinylated Goat Anti-Rabbit IgG (H+L) BA-1000 secondary antibody from Vector Laboratories was used at 3.75 µg/mL. Fixation and staining were as follows: slide cultures were placed on ice, washed 3 times with 1X phosphate-buffered saline (PBS) and fixed in room temperature Methyl Carnoy’s without chloroform (3 parts methanol:1 part glacial acetic acid) for 10 minutes. All further procedures were at room temperature. Slides were washed 3 times for 3 minutes in 1X PBS, incubated in 0.3% H₂ 0₂ in methanol for 20 minutes and washed again 3 times for 3 minutes in 1X PBS. The primary antibody, Rabbit Anti-ZO-1 61-7300, was applied to the slides for 1 hour followed by three 3-minute washes in 1X PBS. The secondary antibody, Goat Anti-Rabbit IgG (H+L) BA-1000 or Rabbit Anti-Chicken/Turkey IgG 61-3100, was applied to the slides for 1 hour followed by three 3-minute washes in 1X PBS. The Vectastain ABC elite kit with peroxidase (Vector Laboratories) was used with the Vector DAB substrate kit. A dark brown reaction product accumulated at the site of secondary antibody binding. The Vectastain ABC Elite reaction was applied to the slides for 30 minutes at a 1/100 dilution. The slides were then washed in 1x Tris buffered saline 3 times for 3 minutes. DAB (Dako) was applied for approximately 2 minutes and the slides were washed in water. All slides were counter stained with Evan’s blue dye, dehydrated in alcohol and cover slipped in Permount (Fischer Scientific). All photomicrographs were taken on an Olympus Vanus microscope at 200X power.

	Results

Top Abstract Introduction Methods Results Discussion References

 
Isolation and Identification of a Random Set of cDNA Clones From SMC Libraries
Twenty random clones were isolated from the pup cDNA library and 24 from the adult. For the purpose of comparing with other RNA samples, we will refer to these 2 libraries as set 1 RNA (set 1 being the RNA samples from which the cDNA libraries were made). Four clones without inserts were discarded. The 40 remaining clones were partially sequenced and analyzed by BLAST homology screening²⁴ of the GenBank, European Molecular Biology Laboratory, Data Bank of Japan, and Protein Data Bank databases and the nonredundant database of the GenBank expressed sequence tag (EST) division. This analysis revealed that approximately three-quarters of the clones (29 clones) were matched in the databases examined to a gene or EST entry (data not shown). No gene was cloned more than once.

Dot Blot Versus Northern Analysis of Cloned Genes
Dot blots containing every random clone and 2 control plasmids with the pup-specific gene osteopontin and the housekeeping-gene ß-actin were used to measure message levels. ³²P-labeled second strand cDNA probes for this assay were prepared from RNA of confluent pup or adult SMCs. A second preparation of RNAs had to be isolated to make probes because the RNA samples used to synthesize the libraries (set 1 RNA) were exhausted during library construction. This new RNA will be referred to as set 2 RNA.

Seventeen random clones were differentially expressed at ratios 1.5-fold between cultures and 10 were differentially expressed at 2-fold. The 2.2-fold value for osteopontin suggests that the assay was sensitive enough to determine differential expression for messages at least 2-fold in variance. Only one gene, SPARC (A24), was expressed at 10-fold. No genes had an "all or none" expression pattern. No ratios could be determined for 9 genes hybridizing at background levels suggesting a detection cutoff. Similar results were obtained after stripping the dot blot and reversing the probe order (data not shown).

We selected for Northern analysis the 17 genes with a dot blot ratio 1.5-fold, 7 nearly equally (<1.5-fold) expressed genes, and the 9 undetectable genes and compared the expression ratios to determine consistency between methods. There were 8 genes with message ratios measured at 1.5-fold by both detection methods (Table 1). For these genes, there was complete agreement on which culture had the highest message level. Sixteen genes had message ratios measured by one or both methods that were <1.5-fold. For these genes, 8 pairs of ratios were in agreement and 8 pairs were in disagreement for the culture with the highest message levels. Therefore, results from the 2 methods coincide only for fold levels >50% difference in expression, establishing a 1.5-fold difference as the bottom line of reproducibility between the 2 methods and the limit for dot blot usefulness in these experiments.

View this table: [in this window] [in a new window]   Table 1. Northern and Dot Blot Data for the 14 Genes With Northern Expression Ratios 1.5-Fold

Analysis of Additional Sets of RNA
Fourteen genes with 1.5-fold difference by Northern analysis alone of set 2 RNA, shown in Table 1, were studied for consistency of differential expression. In addition, we examined the following 6 genes previously reported as over expressed in pup SMCs: elastin, osteopontin, platelet-derived growth factor B chain (PDGF-B), cytochrome P450IA1,^{6 7 8} or as characteristic of all smooth muscle: smooth muscle actin and SM22.^{26 27} Message levels were analyzed in 3 additional RNA samples. These were sets 3 and 4, isolated from additional pairs of pup and adult confluent cultures and set 5, isolated from 3 different states of confluence (2 days after passage, at confluence, and at 3 days after confluence) in the same SMCs used to make set 2 RNA.

Five of the random clones were consistently differentially expressed (Table 2). Highest in the pup SMCs were A2, A4, and A13 and in the adult were 1(I)collagen and SPARC. Effects of confluence on message are shown in Figure 1. Northern hybridization of A13 in set 5 RNA is shown in Figure 2.

View this table: [in this window] [in a new window]   Table 2. Message Ratios From Northern Analysis of RNA Sets 3, 4, and 5

View larger version (15K): [in this window] [in a new window]   Figure 1. Comparison of pup and adult SMC message levels as measured by Northern blotting. Message levels are expressed as ratios of higher expressor over lower expressor and graphed for the randomly cloned differentially expressed genes ZO-2, PPAR, A2, 1(I)collagen, and SPARC, 1 to 5, respectively, and the control differentially expressed genes osteopontin and elastin, 6 and 7, respectively. Samples above the zero mark represent message levels where pup>adult; samples below the zero mark represent message levels where adult>pup. Message levels are displayed in pairs with the hatched bars representing cultures at confluency and the filled bars representing cultures at 3 days after confluency. All Northern hybridizations in this paper were equalized for loading as explained in methods.

View larger version (49K): [in this window] [in a new window]   Figure 2. Northern blot analysis of ZO-2 message levels at 3 growth states in pup and adult SMCs. Cells were 2 days after passage (lanes 1 and 4), at confluency (lanes 2 and 5), and 3 days after reaching confluency (lanes 3 and 6). The blot was stripped and reprobed with a 28S rRNA specific primer to demonstrate equality of loading between lanes.

Identification of Differentially Expressed Unknowns
We next further sequenced clones A2, A4, and A13. A13 had 84% identity in common with the ZO-2 gene sequence between bases 3367 and 3551 of the human sequence²⁸ and 86% to bases 1938 to 2224 of the canine ZO-2 sequence.¹⁷ A4 had 100% DNA sequence identities in common between bases 9 and 249 of the published rat sequence of PPAR.²⁹ A2 was shown to have 80% DNA homology to a human myeloid clone, KIAA0055, related to the tre oncogene.³⁰ With these 3 clones identified, 32 of the 40 clones were identified as known genes or ESTs. These extended sequences were deposited into GenBank as ZO-2, U75916, A2, AA107119 and AA107120, and PPAR, U75919.

Gene Expression in Response to Balloon Injury
We next examined message levels of the 5 differentially expressed genes to screen for involvement in the process of neointimal formation. RNA for each time point was pooled from carotid arteries of 5 to 15 animals that were injured on the same day and harvested at the same time after angioplasty. Although Krox-24 was inconsistently differentially expressed, we examined it here because this transcription factor is an immediate-early serum response gene.³¹ Osteopontin was also monitored because patterns of change in mRNA levels during balloon injury for this gene have been well characterized.^{8 32}

PPAR, Krox-24, and A2 mRNAs were elevated during the first 24 hours after injury (Figure 3). The PPAR message level was highest 4 hours after injury at 2.6 times the control level (Figure 3a). Message levels returned to the control level at 24 hours and remained there for at least 1 week (data not shown). Krox-24 was upregulated at least as early as 1 hour after injury, where it was 11.4 times higher than the control level, shown in Figure 3b. Message levels quickly dropped to 2.5-fold over the control by 6 hours after injury. By 24 hours, Krox-24 returned to the control level (data not shown). A2 message levels, Figure 3c, increased with a peak at 6 hours of 1.5-fold, returned to the control level by 24 hours, and remained low up to a week after injury (data not shown)

View larger version (13K): [in this window] [in a new window]   Figure 3. Carotid artery balloon injury time course quantified Northern analysis of clones showing a short-term elevation in message expression. a to c, The message levels of PPAR, A2, and Krox-24, respectively, during the first 8 hours after injury. Message levels are presented as fold values of the uninjured control hybridization values. Time was measured in hours (h). Control was noted as C.

ZO-2 exhibited sustained elevations comparable with levels of osteopontin, and like osteopontin, ZO-2 remained elevated up to at least 28 days after injury (Figure 4). Osteopontin message levels (Figure 4a) were in agreement with published results.⁸ ZO-2 message is shown in Figure 4b. Within 4 hours, ZO-2 message reached a peak of 8.8 times the control value; previous northern data showed that the message level was still rising at 1 hour after injury (data not shown). This level decreased to 2 times the control value by 24 hours and was maintained between 1.5 to 2 times the control level. To determine whether the upregulation of ZO-2 was in intimal or medial layers, injured carotids were separated into neointimal and medial portions (Figure 5). At 14 to 21 days after injury, neointimal ZO-2 message was approximately 2-fold higher than the uninjured whole carotid. Interestingly, there was a difference in ZO-2 levels between the neointima and the media at 14 days that mostly disappeared by 28 days after injury, when both layers had around twice the ZO-2 level of the control uninjured media.

View larger version (15K): [in this window] [in a new window]   Figure 4. Carotid artery balloon injury time course quantified Northern analysis of the message levels of clones displaying longer-term responses and osteopontin in carotid arteries after balloon injury over a 28-day time course. a, osteopontin; b, ZO-2; c, 1(I)collagen; and d, SPARC. Time was measured in hours (h) and days (d). Control was noted as C.

View larger version (17K): [in this window] [in a new window]   Figure 5. Northern analysis of separated neointima and media from balloon-injured carotid arteries. Lane 1, uninjured carotid; lanes 2 and 3, 14-day balloon-injured media and neointima, respectively; lanes 4 and 5, 28-day balloon-injured media and neointima, respectively.

Consistent with published results,⁷ 1(I)collagen mRNA (Figure 4c) began decreasing within 1 hour after injury and reached a low of 5.9-fold less than control at 24 hours. Message levels returned to and remained near uninjured control levels by 10 days. SPARC message, a consistent marker of adult SMCs, showed no overall trend in expression in vivo after balloon injury (Figure 4 days).

ZO-2 Expression in Pup Cell Clones
We have previously reported that clones made from pup cell cultures show either an adult or pup-like phenotype both by morphology and expression pattern. Four epithelioid pup clones⁹ (Figure 6) showed equally high levels of ZO-2 expression. One clone had the spindle-shaped morphology of adult SMC cells (clone V). This clone and clone II, with epithelioid morphology, had approximately equally low levels of ZO-2 message. Clone II is also adult-like in having very low expression of PDGF-B,⁹ a pup SMC–specific gene.⁷

View larger version (40K): [in this window] [in a new window]   Figure 6. Northern analysis of ZO-2 in 6 clonal isolates from the pup SMC at confluency. Pup I through IV and VI are pup-like and pup V is adult-like with respect to morphological phenotype, growth properties, and gene expression.9 The blot was stripped and reprobed with a 28S specific primer to demonstrate equality of loading.

Expression of Other Tight Junction Proteins
We examined the message level of several MAGUK and tight junction proteins, including ZO-1,¹⁷ the synapse associated proteins 90 and 97 (SAP90 and SAP97)^{33 34} and tight junction integral membrane proteins occludin,^{22 35} claudin 1 and 2.^{36 37} ZO-1 message levels were higher in pup SMCs at all growth states (Table 2) with the highest level at subconfluence with 2.8 times the adult and decreasing to 1.2-fold the adult at three days after confluency. There was a higher level of SAP90 in the adult SMCs at all growth states, with the highest level, a 4.9-fold difference, being found at confluence. Occludin and SAP97 messages were equally expressed between the pup and adult SMCs (data not shown). We analyzed claudin 1 and 2 by rtPCR in pup and adult SMCs. Both genes are expressed in pup SMCs but not adult SMCs (data not shown).

ZO-1 Protein Expression In Vivo
Because there are no antibodies currently available against rodent ZO-2, we next examined the protein expression pattern of ZO-1, another tight junction MAGUK gene to examine for the presence of tight junctions in smooth muscle.¹⁷ ZO-1 antibody staining of confluent pup cells showed a continuous band of ZO-1 protein localized to cell/cell interfaces (Figure 7A). A similar pattern was seen when cells grown from sparse density formed cell contacts (Figure 7B), whereas membranes at the exterior of cell islands and isolated cells exhibited only disconnected spots of ZO-1 staining. The adult cells, in contrast, exhibited cytoplasmic rather than membrane localized staining for ZO-1, whether at confluent or subconfluent growth states, Figures 7C and 7D, respectively.

View larger version (117K): [in this window] [in a new window]   Figure 7. ZO-1 protein immunocytochemistry of pup and adult SMCs. Confluent pup SMCs stained with polyclonal Rabbit Anti–ZO-1 61-7300 are shown in A, subconfluent pup SMCs are shown in B. Confluent adult SMCs are shown in C and subconfluent adult SMCs are shown in D. Control primary antibody Rabbit Anti-Chicken/Turkey IgG 61-3100 staining for background level is shown for pup SMCs in E, adult SMC staining was performed with the control antibody with the same results (data not shown). Original magnification 200X power.

Disappointingly, ZO-1 was not localized on luminal neointimal SMCs but was more generally expressed in the intima and media, as determined by immunohistochemical staining of cross sections of uninjured or injured carotid arteries at either 14 or 28 days after injury. En face preparations of 14-day balloon-injured vessels stained with ZO-1 antibodies showed that while endothelial cells at the injury boundary localized ZO-1 to sites of cell to cell contact, the adjacent neointimal SMCs had only diffuse cytoplasmic staining (data not shown).

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
Analysis of a small sample of randomly selected clones suggests that pup and adult SMCs differ in expression for a very large number of genes. Five (12.5%) of the genes were consistently differentially expressed at the same growth states. A2, PPAR, and ZO-2 messages were always higher in pup SMCs. SPARC and 1(I)collagen messages were always higher in adult SMCs. Three of these genes, ZO-2, PPAR, and 1(I)collagen were upregulated after balloon injury; ZO-2 was higher in the neointima than in the uninjured media at least up to 21 days after injury and elevated in the injured whole carotid up to 28 days after injury.

We were initially surprised that no gene was found more than once among the 40 clones. Lewin’s analysis of different tissues³⁸ led him to estimate that <100 abundant genes make up approximately 50% of a cell’s mRNA, whereas approximately 10 000 intermediate and low abundance genes make up the remainder. Even large scale studies using in depth sequencing show <25% repeat sequences from several thousand random clones.³⁹ Repeat clones, although possible, are not likely to be frequent occurrences in the sample size used in our survey.

Two genes showing consistent differential expression in the pup cells, PPAR and A2, are transcription factors. Both genes showed transient elevation in vivo shortly after balloon injury. PPAR message was higher in pup SMCs and was elevated only during the day after balloon injury. Three additional members of the PPAR family have been discovered to date.⁴⁰ These genes belong to the steroid/thyroid/vitamin super-family of nuclear receptors involved in transcription and differentiation. Overexpression studies in vitro have shown that PPAR initiates adipogenic differentiation and suppresses myogenic differentiation in myoid and fibroblastic cell lines.⁴¹ PPAR has recently been found by rtPCR in rat aortic SMCs.⁴² Activators of PPAR inhibit smooth muscle activation and inflammatory response.⁴³ A2, found in immature myeloid cells, appears to be a novel transcription factor based on an 80% homology to a human gene with homology to the tre oncogene.

Perhaps the most surprising observation made during this study was the presence of tight junction proteins in smooth muscle because tight junctions are thought to be found almost exclusively in epithelia, including the endothleium. However, in vitro pup cells⁹ form an epithelioid monolayer that resembles the appearance of tight junction containing cells cultures. The ZO-1 staining pattern for confluent and subconfluent pup cultures is similar to that seen by Li and Poznansky in confluent and subconfluent endothelial cultures.⁴⁴ In both pup and these endothelial cells, adjacent cells had a continuous band of ZO-1 staining, whereas cells in islands or by themselves had discontinuous bands or punctate staining at exposed membrane surfaces. Similar membrane structures were seen by Ehler et al who examined cell/cell junctions containing ZO-1 and cingulin in epithelioid but not spindle-shaped smooth muscle clonal lines cultured from juvenile mouse aortas.⁴⁵ An epithelioid pattern with freeze fracture evidence for tight junctions is also seen in vivo in the surface layer of neointimal cells formed after balloon injury.^{46 47} Our data suggest a broader expression of tight junction genes in SMC. In addition to ZO-1 and ZO-2, both pup and adult SMCs express occludin mRNA by Northern analysis (data not shown). Recently it has been shown by gene targeted disruption that occludin is not required for tight junction strand formation and may play an accessory role at tight junctions.³⁵ Furuse et al have shown there are at least 2 additional integral membrane proteins, claudin 1 and 2,³⁶ that are present in tight junctions and reconstitute tight junctions strands when transfected into a fibroblast line lacking these structures.³⁷ We have found claudin 1 and 2 message expressed in pup but not adult SMCs by rtPCR assay. These data show that both peripheral cytoplasmic components and integral membrane components of the tight junction are expressed in smooth muscle.

Although immunocytochemistry shows that ZO-1 is located in junctions between adjacent pup cells and pup cell monolayers, other SMCs in vivo and in vitro had ZO-1 throughout the cytoplasm. The expression of ZO-1, but not ZO-2, in cells lacking tight junctions has been documented in several additional cell types. ZO-1 is found in cardiac myocytes at fascia adherens, which have a cohesive role at intercalated discs,¹⁷ at adherens junctions in hepatocytes and cell/cell contacts in rat 3Y1 fibroblasts.⁴⁸ ZO-1 is also found in astrocytes at points of cell/cell contact, in Schwann cells, and in dermal fibroblasts.⁴⁹ The RNA expression data for ZO-2, the claudins and occludin, which are tight junction-specific genes, argues for the presence of tight junctions in pup SMCs when added to the ZO-1 protein localization pattern.

The appearance of ZO proteins in smooth muscle may reflect roles for these proteins other than tight junction formation. The above examples show that ZO-1 is more ubiquitously expressed and has different functions at the membrane in a variety of cell/cell interfaces than ZO-2. Our data on ZO-2 in smooth muscle suggests it is somewhat more widely expressed as well. Studies of related dlg family members suggest the intriguing possibility that ZO-2 and ZO-1 are part of a signaling complex involved in cell-cell interactions. Gottardi et al have shown that ZO-1 can move between the nucleus and the cell membrane during remodelling of cell-cell contacts.⁵⁰ ZO-1 binds to -catenin and possibly other catenins,^{51 52} to the cytoskeletal elements F-actin⁵³ and cortactin,⁵⁴ and the tight junction associated protein occludin²² (thereby bridging between the cytoskeleton and the tight junction) and binds to ZO-2 itself.¹⁷ Matsumine et al have found that the MAGUK human disc large protein (hDLG) binds to the carboxy terminal region of the adenomatous polyposis coli (APC) tumor suppresser gene through the second of its 3 disc-large homology region repeat regions,⁵⁵ an interaction which possibly affects the ability of APC to block cell cycle progression.⁵⁶ Both hDLG and APC colocalize at the lateral cytoplasm of colon epithelial cells and the synaptic regions of cultured hippocampal neurons.⁵⁵ Another MAGUK family member, SAP90 (an adult SMC specific gene), binds to N-methyl-D-aspartate receptor subunits, nitric oxide synthase, and 1-syntrophin and binds to and clusters shaker-type K+ channels.^{18 19 20} The presence of ZO-1, ZO-2, and SAP90 in morphologically different SMCs suggest similar signaling processes might be involved in morphogenic changes that occur after vessel injury.

In summary, mRNA expression differences and morphological data from only a small sampling of genes imply there are many more differences in gene expression between distinct types of rat vascular SMCs. Of the new differentially expressed genes identified in this study, ZO-2 may provide clues to morphogenic signaling pathways that could have implications for other phenotypic differences between smooth muscle subtypes.

	Acknowledgments

 
We would like to thank Dr Cecilia Giachelli for her helpful advice and critical reading of this manuscript. We are indebted to Isa Werny and Robin Pruit for their excellent technical assistance in the molecular studies and cell culture portions of this project and to the superb technical assistance of Donna Lombardi and Patti Polinsky with the animal injury model and immunocytochemistry studies. This project was supported by NIH grants HL03174 and HL26405.

Received September 15, 1998;

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Walker L, Bowen-Pope DF, Ross R, Reidy MA. Production of PDGF-like molecules by cultured arterial smooth muscle cells accompanies proliferation after arterial injury. Proc Natl Acad Sci U S A. 1986;83:7311–7315.[Abstract/Free Full Text]
2. Majesky MW, Schwartz SM. Smooth muscle diversity in arterial wound repair. Toxicol Pathol. 1990;18:554–559.[Medline] [Order article via Infotrieve]
3. Bochaton-Piallat M-L, Ropraz P, Gabbiani F, Gabbiani G. Phenotypic heterogeneity of rat arterial smooth muscle cell clones: Implications for the development of experimental intimal thickening. Arterioscler Thromb Vasc Biol. 1996;16:815–820.[Abstract/Free Full Text]
4. Gordon D, Mohai LG, Schwartz SM. Induction of polyploidy in cultures of neonatal rat aortic smooth muscle cells. Circ Res. 1986;59:633–644.[Abstract/Free Full Text]
5. Dartsch PC, Voisard R, Bauriedel G, Hofling B, Betz E. Growth characteristics and cytoskeletal organization of cultured smooth muscle cells from human primary stenosing and restenosing lesions. Arteriosclerosis. 1990;10:62–75.[Abstract/Free Full Text]
6. 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.[Abstract/Free Full Text]
7. 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.[Abstract/Free Full Text]
8. 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.[Medline] [Order article via Infotrieve]
9. Lemire JM, Covin CW, White S, Giachelli CM, Schwartz SM. Characterization of cloned aortic smooth muscle cells from young rats. Am J Pathol. 1994;144:1068–1081.[Abstract]
10. Lemire JM, Potter-Perigo S, Hall KL, Wight TN, Schwartz SM. Distinct rat aortic smooth muscle cells differ in versican/PG-M expression. Arterioscler Thromb Vasc Biol. 1996;16:821–829.[Abstract/Free Full Text]
11. Giachelli CM, Bae N, Almeida M, Denhardt DT, Alpers CE, Schwartz SM. Osteopontin is elevated during neointima formation in rat arteries and is a novel component of human atherosclerotic plaques. J Clin Invest. 1993;92:1686–1696.
12. O’Brien ER, Garvin MR, Stewart DK, Hinohara T, Simpson JB, Schwartz SM, Giachelli CM. Osteopontin is synthesized by macrophage, smooth muscle and endothelial cells in primary and restenotic human coronary atherosclerotic plaques. Arterioscler Thromb. 1994;14:1648–1656.[Abstract/Free Full Text]
13. Wilcox JN, Smith KM, Williams LT, Schwartz SM, Gordon D. Platelet-derived growth factor mRNA detection in human atherosclerotic plaques by in situ hybridization. J Clin Invest. 1988;82:1134–1143.
14. Lindner V, Giachelli CM, Schwartz SM, Reidy MA. A subpopulation of smooth muscle cells in injured rat arteries expresses PDGF-B chain mRNA. Circ Res. 1995;76:951–957.[Abstract/Free Full Text]
15. Murry CE, Bartosek T, Giachelli CM, Alpers CE, Schwartz SM. Platelet-derived growth factor-A mRNA expression in fetal, normal adult, and atherosclerotic human aortas. Analysis by competitive polymerase chain reaction. Circulation. 1996;93:1095–1106.[Abstract/Free Full Text]
16. Tanaka H, Sukhova G, Schwartz D, Libby P. Proliferating arterial smooth muscle cells after balloon injury express TNF-alpha but not interleukin-1 or basic fibroblast growth factor. Arterioscler Thromb Vasc Biol. 1996;16:12–18.[Abstract/Free Full Text]
17. Jesaitis DA, Goodenough DA. Molecular characterization and tissue distribution of ZO-2, a tight junction protein homologous to ZO-1 and the Drosophila discs-large tumor suppressor protein. J Cell Biol. 1994;124:949–961.[Abstract/Free Full Text]
18. Niethammer M, Kim E, Sheng M. Interaction between the C terminus of NMDA receptor subunits and multiple members of the PSD-95 family of membrane-associated gyanylate kinases. J Neurosci. 1996;16:2157–2163.[Abstract/Free Full Text]
19. Kim E, Niethammer M, Rothschild A, Jan YN, Sheng M. Clustering of Shaker-type K+ channels by interaction with a family of membrane-associated guanylate kinases. Nature. 1995;378:85.[Medline] [Order article via Infotrieve]
20. Brenman JE, Chao DS, Gee SH, McGee AW, Craven SE, Santillano DR, Wu Z, Huang F, Xia H, Peters MF, Froehner SC, Bredt DS. Interaction of nitric oxide synthase with the postsynaptic density protein PSD-95 and 1-syntrophin mediated by PDZ domains. Cell. 1996;84:757–767.[Medline] [Order article via Infotrieve]
21. Marfatia SM, Leu RA, Branton D, Chishti AH. Identification of the protein 4.1 binding interface on glycophorin C and p55, a homologue of the Drosophila discs-large tumor suppressor protein. J Biol Chem. 1995;270:715–719.[Abstract/Free Full Text]
22. Furuse M, Itoh M, Hirase T, Nagafuchi A, Yonemura S, Tsukita S. Direct association of occludin with ZO-1 and its possible involvement in the localization of occludin at tight junctions. J Cell Biol. 1994;127:1617–1626.[Abstract/Free Full Text]
23. Zapata JM, Takahashi R, Salvesen GS, Reed JC. Granzyme release and caspase activation in activated human T-lymphocytes. J Biol Chem. 1998;273:6916–6920.[Abstract/Free Full Text]
24. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Molec Biol. 1990;215:403–410.[Medline] [Order article via Infotrieve]
25. Reidy MA, Clowes AW, Schwartz SM. Endothelial regeneration V. Inhibition of endothelial regrowth in arteries of rat and rabbit. Lab Invest. 1983;49:569–575.[Medline] [Order article via Infotrieve]
26. McHugh KM, Lessard JL. The development expression of the rat -vascular and gamma-enteric smooth muscle isoactins: Isolation and characterization of a rat gamma-enteric actin cDNA. Mol Cell Biol. 1988;8:5224–5231.[Abstract/Free Full Text]
27. Solway J, Seltzer J, Samaha FF, Kim S, Alger LE, Niu Q, Morrisey EE, Ip HS, Parmacek MS. Structure and expression of a smooth muscle cell-specific gene, SM22 . J Biol Chem. 1995;270:13460–13469.[Abstract/Free Full Text]
28. Duclos F, Rodius F, Wrogemann K, Mandel J-L, Koenig M. The Friedrich ataxia region: characterization of two novel genes and reduction of the critical region to 300 kb. Hum Mol Genet. 1994;3:909–914.[Abstract/Free Full Text]
29. Xing G, Zhang L, Heynen T, Yoshikawa T, Smith M, Weiss S, Detera-Wadleigh S. Rat PPAR-gamma contains a CGG triplet repeat and is prominently expressed in the thalamic nuclei. Biochem Biophys Res Commun. 1995;217:1015–1025.[Medline] [Order article via Infotrieve]
30. Nakamura T, Hillova J, Mariage-Samson R, Onno M, Huebner K, Cannizzaro LA, Boghosian-Sell L, Croce CM, Hill M. A novel transcriptional unit of the tre oncogene widely expressed in human cancer cells. Oncogene. 1992;7:733–741.[Medline] [Order article via Infotrieve]
31. Lemaire P, Revelant O, Bravo R, Charnay P. Two mouse genes encoding potential transcription factors with identical DNA-binding domains are activated by growth factors in cultured cells. Proc Natl Acad Sci U S A. 1988;85:4691–4695.[Abstract/Free Full Text]
32. Shanahan CM, Weissberg PL, Metcalfe JC. Isolation of gene markers of differentiated and proliferating vascular smooth muscle cells. Circ Res. 1993;73:193–204.[Abstract]
33. Cho K-O, Hunt CA, Kennedy MB. The rat brain postsynaptic density fraction contains a homolog of the drosophila discs-large tumor suppressor protein. Neuron. 1992;9:929–942.[Medline] [Order article via Infotrieve]
34. Müller BM, Kistner U, Veh RW, Cases-Langhoff C, Becker B, Gundelfinger ED, Garner CC. Molecular characterization and spatial distribution of SAP97, a novel presynaptic protein homologous to SAP90 and the drosophila discs-large tumor suppressor protein. J Neurosci. 1995;15:2354–2366.[Abstract]
35. Saitou M, Fujimoto K, Doi Y, Itoh M, Fujimoto T, Furuse M, Takano H, Noda T, Tsukita S. Occludin-deficient embryonic stem cells can differentiate into polarized epithelial cells bearing tight junctions. J Cell Biol. 1998;141:397–408.[Abstract/Free Full Text]
36. Furuse M, Fujita K, Takashi HFKTS. Claudin-1 and -2: novel integral membrane protein localizing at tight junctions with no sequence similarity to occludin. J Cell Biol. 1998;141:1539–1550.[Abstract/Free Full Text]
37. Lee E, Vaughan DE, Parikh SH, Grodzinsky AJ, Libby P, Lark MW, Lee RT. Regulation of matrix metalloproteinases and plasminogen activator inhibitor-1 synthesis by plasminogen in cultured human vascular smooth muscle cells. Circ Res. 1996;78:44–49.[Abstract/Free Full Text]
38. Lewin B. Genes V. 1st Ed. New York, NY: Oxford University Press; 1994:1–1272.
39. Yang Y, Peterson KR, Stamatoyannopoulos G, Papayannopoulou T. Human CD34+ cell EST database: single-pass sequencing of 402 clones from a directional cDNA library. Exp Hematol. 1996;24:605–612.[Medline] [Order article via Infotrieve]
40. Motojima K. Peroxisome proliferator-activated receptor (PPAR): Structure, mechanisms of activation and diverse functions. Cell Struct Funct. 1993;18:267–277.[Medline] [Order article via Infotrieve]
41. Wu Z, Xie Y, Bucher NLR, Farmer SR. Conditional ectopic expression of C/EBP ß eta in NIH-3T3 cells induces PPAR-gamma and stimulates adipogenesis. Genes Dev. 1995;9:2350–2363.[Abstract/Free Full Text]
42. Iijima K, Yoshizumi M, Ako J, Eto M, Kim S, Hashimoto M, Sugimoto N, Liang YQ, Sudoh N, Toba K, Ouchi Y. Expression of peroxisome proliferator-activated receptor gamma (PPARgamma) in rat aortic smooth muscle cells. Biochem Biophys Res Commun. 1998;247:353–356.[Medline] [Order article via Infotrieve]
43. Staels B, Koenig W, Habib A, Merval R, Lebret M, Torra IP, Delerive P, Fadel A, Chinetti G, Fruchart J-C, Najib J, Maclouf J, Tedgui A. Activation of human aortic smooth-muscle cells is inhibited by PPAR but not by PPARy activators. Nature. 1998;393:790–793.[Medline] [Order article via Infotrieve]
44. Li C, Poznansky MJ. Characterization of the ZO-1 protein in endothelial and other cell lines. J Cell Sci. 1990;97:231–237.[Abstract/Free Full Text]
45. Ehler E, Parmjit SJ, Noble MD, Citi S, Draeger A. Vascular smooth muscle cells of H-2Kb-tsA58 transgenic mice: characterization of cell lines with distinct properties. Circulation. 1995;92:3289–3296.[Abstract/Free Full Text]
46. Schwartz SM, Stemerman MB, Benditt EP. The aortic intima. II. Repair of the aortic lining after mechanical denudation. Am J Pathol. 1975;81:15–42.[Abstract]
47. Kocher O, Gabbiani F, Gabbiani G, Reidy MA, Cokay MS, Peters H, Huttner I. Phenotypic features of smooth muscle cells during the evolution of experimental carotid artery intimal thickening: biochemical and morphologic studies. Lab Invest. 1991;65:459–470.[Medline] [Order article via Infotrieve]
48. Itoh M, Nagafuchi A, Yonemura S, Kitani-Yasuda T, Tsukita S. The 220-kD protein colocalizing with cadherins in non-epithelial cells is identical to ZO-1, a tight junction-associated protein in epithelial cells: cDNA cloning and immunoelectron microscopy. J Cell Biol. 1993;121:491–502.[Abstract/Free Full Text]
49. Howarth AG, Hughes MR, Stevenson BR. Detection of the tight junction-associated protein ZO-1 in astrocytes and other nonepithelial cell types. Am J Physiol. 1992;262:C461–C469.[Abstract/Free Full Text]
50. Gottardi CJ, Arpin M, Fanning AS, Louvard D. The junction-associated protein, zonula occludens-1, localizes to the nucleus before the maturation and during the remodeling of cell-cell contacts. Proc Natl Acad Sci U S A. 1996;93:10779–10784.[Abstract/Free Full Text]
51. Rajasekaran AK, Hojo M, Huima T, Rodriguez-Boulan E. Catenins and zonula occludens-1 form a complex during early stages in the assembly of tight junctions. J Cell Biol. 1996;132:451–463.[Abstract/Free Full Text]
52. Itoh M, Nagafuchi A, Monroi S, Tsukita S. Involvement of ZO-1in cadherin-based cell adhesion through its direct binding to a catenin and actin filaments. J Cell Biol. 1997;138:181–192.[Abstract/Free Full Text]
53. Fanning AS, Jameson BJ, Jesaitis LA, Anderson JM. The tight junction protein ZO-1 establishes a link between the transmembrane protein occludin and the actin cytoskeleton. J Biol Chem. 1998;273:29745–29753.[Abstract/Free Full Text]
54. Katsube T, Takahisa M, Ueda R, Hashimoto N, Kobayashi M, Togashi S. Cortactin associates with the cell-cell junction protein ZO-1 in both drosophila and mouse. J Biol Chem. 1998;273:29672–29677.[Abstract/Free Full Text]
55. Matsumine A, Ogai A, Senda T, Okumura N, Satoh K, Baeg GH, Kawahara T, Kobayashi S, Okada M, Toyoshima K, Akiyama T. Binding of APC to the human homolog of the Drosophila discs large tumor suppressor protein. Science. 1996;272:1020–1023.[Abstract]
56. Baeg GH, Matsumine A, Kuroda T, Bhattacharjee RN, Miyashiro I, Toyoshima K, Akiyama T. The tumour suppressor gene product APC blocks cell cycle progression from G0/G1 to S phase. EMBO J. 1995;14:5618–5625.[Medline] [Order article via Infotrieve]

This article has been cited by other articles:

				H. Hao, G. Gabbiani, and M.-L. Bochaton-Piallat Arterial Smooth Muscle Cell Heterogeneity: Implications for Atherosclerosis and Restenosis Development Arterioscler. Thromb. Vasc. Biol., September 1, 2003; 23(9): 1510 - 1520. [Abstract] [Full Text] [PDF]

				R. L. Geary, J. M. Wong, A. Rossini, S. M. Schwartz, and L. D. Adams Expression Profiling Identifies 147 Genes Contributing to a Unique Primate Neointimal Smooth Muscle Cell Phenotype Arterioscler. Thromb. Vasc. Biol., December 1, 2002; 22(12): 2010 - 2016. [Abstract] [Full Text] [PDF]

				J. Zhang, M. Fu, X. Zhu, Y. Xiao, Y. Mou, H. Zheng, M. A. Akinbami, Q. Wang, and Y. E. Chen Peroxisome Proliferator-activated Receptor delta Is Up-regulated during Vascular Lesion Formation and Promotes Post-confluent Cell Proliferation in Vascular Smooth Muscle Cells J. Biol. Chem., March 22, 2002; 277(13): 11505 - 11512. [Abstract] [Full Text] [PDF]

				J. T. N. Tai, E. E. Brooks, S. Liang, R. Somogyi, J. D. Rosete, R. M. Lawn, and D. Shiffman Determination of Temporal Expression Patterns for Multiple Genes in the Rat Carotid Artery Injury Model Arterioscler. Thromb. Vasc. Biol., October 1, 2000; 20(10): 2184 - 2191. [Abstract] [Full Text] [PDF]

				L. D. Adams, R. L. Geary, B. McManus, and S. M. Schwartz 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 Circ. Res., September 29, 2000; 87(7): 623 - 631. [Abstract] [Full Text] [PDF]

				S. Li, Y.-S. Fan, L. H. Chow, C. Van Den Diepstraten, E. van der Veer, S. M. Sims, and J. G. Pickering Innate Diversity of Adult Human Arterial Smooth Muscle Cells: Cloning of Distinct Subtypes From the Internal Thoracic Artery Circ. Res., September 14, 2001; 89(6): 517 - 525. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend