Origins of Serum Semicarbazide-Sensitive Amine Oxidase Craig M. Stolen, Gennady G. Yegutkin, Riikka Kurkijärvi, Petri Bono, Kari Alitalo and Sirpa Jalkanen Circ. Res. 2004;95;50-57; originally published online Jun 3, 2004; DOI: 10.1161/01.RES.0000134630.68877.2F Circulation Research is published by the American Heart Association. 7272 Greenville Avenue, Dallas, TX 72514 Copyright © 2004 American Heart Association. 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/cgi/content/full/95/1/50 Data Supplement (unedited) at: http://circres.ahajournals.org/cgi/content/full/95/1/50/DC1
Subscriptions: Information about subscribing to Circulation Research is online at http://circres.ahajournals.org/subscriptions/ Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, a division of Wolters Kluwer Health, 351 West Camden Street, Baltimore, MD 21202-2436. Phone: 410-528-4050. Fax: 410-528-8550. E-mail:
[email protected] Reprints: Information about reprints can be found online at http://www.lww.com/reprints
Downloaded from circres.ahajournals.org by on May 23, 2011
Origins of Serum Semicarbazide-Sensitive Amine Oxidase Craig M. Stolen, Gennady G. Yegutkin, Riikka Kurkija¨rvi, Petri Bono, Kari Alitalo, Sirpa Jalkanen Abstract—Semicarbazide-sensitive amine oxidases (SSAO) are enzymes that are capable of deaminating primary amines to produce aldehyde, ammonia, and hydrogen peroxide. This activity has been associated with vascular adhesion protein-1 (VAP-1) and is found in the serum, endothelium, adipose, and smooth muscle of mammals. Circulating SSAO activity is increased in congestive heart failure, diabetes, and inflammatory liver diseases. To investigate the origin of circulating SSAO activity, two transgenic mouse models were created with full-length human VAP-1 (hVAP-1) expressed on either endothelial (mTIEhVAP-1) or adipose tissues (aP2hVAP-1), with tie-1 and adipocyte P2 promoters, respectively. Under normal conditions a circulating form of hVAP-1 was found at high levels in the serum of mice with endothelium-specific expression and at low levels in the serum of mice with adipose specific expression. The level of circulating hVAP-1 in the transgenic mice varied with gender, transgene zygosity, diabetes, and fasting. Serum SSAO activity was absent from VAP-1 knockout mice and endothelial cell–specific expression of human VAP-1 restored SSAO activity to the serum of VAP-1 knockout mice. Together, these experiments show that in the mouse VAP-1 is the only source of serum SSAO, that under physiological conditions vascular endothelial cells can be a major source of circulating VAP-1 protein and SSAO, and that serum VAP-1 can originate from both endothelial cells and adipocytes during experimental diabetes. An increased endothelial cell capacity for lymphocyte binding and altered expression of redox-sensitive proteins was also associated with the mTIEhVAP-1 transgene. (Circ Res. 2004;95:50-57.) Key Words: diabetes 䡲 hydrogen peroxide 䡲 transgenic mice 䡲 vascular disease 䡲 vascular endothelium
V
ascular adhesion protein-1 (VAP-1) was first identified as a novel adhesion molecule that is critically involved in the process of leukocyte trafficking to sites of inflammation in man.1–3 Subsequent cloning and sequencing of the cDNAs encoding murine and human VAP-1 revealed the marked homology of VAP-1 with enzymes called semicarbazide-sensitive amine oxidases (SSAO).4 In fact, VAP-1 has dual functions: besides its adhesive properties, it also possesses SSAO enzyme activity.5 SSAOs belong to the widely distributed copper-containing monoamine oxidase class (EC 1.4.3.6) and catalyze the oxidative deamination of primary amines to produce aldehydes, ammonium, and hydrogen peroxide (RCH2NH2⫹O2⫹H2O3 RCHO⫹NH3⫹H2O2).6 The VAP-1 molecule is expressed on the surface of endothelial cells and adipocytes, where its enzyme activity can directly regulate lymphocyte rolling7 and stimulate glucose transport,8 respectively. The products of the SSAO reaction may also have other biological consequences. For example, the oxidative deamination of endogenous methylamine results in the formation of formaldehyde, which is an extremely reactive chemical that is capable of causing protein crosslinking9 and increasing advanced glycation end product formation.10,11 In addition, hydrogen peroxide is a major reactive oxygen species that can cause cell damage and function as a signaling molecule.12–14
In addition to the membrane-associated VAP-1, a catalytically active form of this protein circulates in the blood stream as a soluble molecule.15 A number of clinical reports have suggested a role for the circulating SSAO activity in specific pathological conditions, including diabetes mellitus,16 –19 congestive heart failure,20,21 and inflammatory liver diseases.22 The cellular source of this circulating enzyme is unknown. Whether it is derived from shedding of the transmembrane spanning VAP-1 protein or is a product of a different gene is also unknown. In this study, we used transgenic technology to specifically express the full-length human VAP-1 protein either on the surface of endothelial cells or on adipocytes. Using these mice together with VAP-1 knockout mice, we show that VAP-1 is the only source of circulating SSAO in the mouse, that circulating VAP-1/SSAO can be derived from the human VAP-1 gene and that, under physiological conditions, the endothelium can be major source of circulating, catalytically active VAP-1/SSAO, whereas in times of biological stress, such as diabetes, both adipocytes and endothelial cells can release serum hVAP-1. We also show VAP-1 expression increases the capacity of endothelial cells to bind lymphocytes and that its over-expression promotes the expression of redox sensitive proteins.
Original received January 14, 2004; revision received May 13, 2004; accepted May 21, 2004. From the MediCity Research Laboratory (C.M.S., G.G.Y., R.K., S.J.), University of Turku and National Public Health Institute, Turku, Finland; the Department of Oncology (P.B.), Helsinki University Central Hospital, Helsinki, Finland; and the Molecular/Cancer Biology Laboratory (K.A.), University of Helsinki, Helsinki, Finland. Correspondence to Craig M. Stolen, PhD, MediCity Research Laboratory, University of Turku, Tykisto¨katu 6A, FIN-20520, Turku, Finland. E-mail
[email protected] © 2004 American Heart Association, Inc. Circulation Research is available at http://www.circresaha.org
DOI: 10.1161/01.RES.0000134630.68877.2F
50 Downloaded from circres.ahajournals.org by on May 23, 2011
Stolen et al
Origins of Serum SSAO
51
Transgenic Lines Expressing Human VAP-1 Protein in Serum and Tissues Zygosity
Gender
Copy No.
sVAP-1, g/L⫾SD**
n
Non-tg
⫺/⫺
female
0
0.7⫾1.5
25
⫺
Non-tg
⫺/⫺
male
0
1.6⫾2.0
13
⫺
E25
⫹/⫺
female*
4
E25
⫹/⫺
male*
E35
⫹/⫺㛳
female†
E35
⫹/⫺¶
male†
E35
⫹/⫹㛳
female‡
E35
⫹/⫹¶
male‡
⫹/⫺
female
Line
hVAP-1 Immunostaining
mTIEhVAP-1
4 8
9.9⫾5.6
16
⫹
15.5⫾6.0
12
⫹
12.5⫾2.8
23
⫹
25.1⫾14.4
32
⫹
28.0⫾3.7
6
⫹
47.9⫾21.8
20
⫹
4.9⫾2.2
2
⫹
aP2hVAP-1 A3
ND
A31
⫹/⫺
male
ND
6.5⫾10.8
4
⫹
A33
⫹/⫺
female§
ND
0.2⫾2.6
3
⫹
A33
⫹/⫺
male§
ND
9.6⫾7.1
12
⫹
Non-tg, nontransgenic; ND, not determined; SD, standard deviation; n, No. of serum samples analyzed. Immunostaining: ⫺, no specific staining; ⫹, mTIEhVAP-1 endothelial staining or aP2hVAP-1 adipose staining. Student t test comparing groups with matching symbols: *, ‡, §P⬍0.05; †, 㛳, ¶Pⱕ0.0001. **Values of controls without serum⫽⫺0.3⫾2.6 g/L⫾SD.
Materials and Methods Human VAP-1 ELISA Transgenic human VAP-1 levels were analyzed using an ELISA exactly as described.15 Briefly, the serum and tissue samples were diluted in blocking solution and tested in triplicate. The VAP-1 content was quantified by absorbing anti–VAP-1 mAb, TK8 –18, onto the bottom of 96-well microplates, incubating with the diluted samples, and detecting the bound VAP-1 with another anti–VAP-1 mAb, biotinylated TK8 –14, followed by peroxidase-conjugated streptavidin and BM Chemiluminescence ELISA Reagent (Boehringer-Mannheim). The VAP-1 concentration was determined for each sample by subtracting the mean value of luminescence (Tecan Ultra Luminometer) obtained with an irrelevant detecting mAb, biotinylated Hermes-3, from the VAP-1–specific intensity followed by conversion to ng/mL using linear regression and known standards included on each plate.
were injected daily for 5 days during the first week with STZ (60 mg/kg body weight) or vehicle only and then similarly for 5 consecutive days during week 7.
Stamper-Woodruff Adhesion Assay Nonstatic ex vivo lymphocyte binding assays were performed as described earlier.23 For a brief description, see the expanded Materials and Methods section, found in the online data supplement available at http://circres.ahajournals.org.
Two-Dimensional PAGE and Peptide Sequencing Isoelectric focusing and mass-spectrometry were performed at the University of Turku Center for Biotechnology. A brief description of the protocol can be found with the expanded Materials and Methods section. Additional details of the materials and methods used can be found in the online data supplement.
Measurement of SSAO Activity Total amine oxidase activity, derived from both endogenous and transgenic sources, was assayed either radiochemically using [7-14C]benzylamine hydrochloride (specific activity 57 mCi/mmol, Amersham) as a substrate22 or by fluorometric determination of the reaction product H2O2 using Amplex Red reagent.7
Induction of Diabetes Streptozotocin (STZ) was dissolved in 0.05 mol/L sodium citrate (pH 4.5) and immediately injected intraperitoneally. The glucose levels of arteriovenous blood collected from tail vessels was measured at regular intervals using a MediSense Precision Xtra Plus sensor (Abbot Oy). An optimal streptozotocin “high dose” of 240 mg/kg body weight was selected during a pilot study for its capacity to consistently induce stable hyperglycemia in this strain of mice (FVB/NTac; Taconic M&B, Denmark). All experiments were conducted in accordance with the University of Turku and Finnish National guidelines for animal care and use. Mice that failed to become hyperglycemic (⬍13 mmol/mL blood) after a single high dose of STZ were given a second STZ injection at an equal or lower concentration (240 or 120 mg/kg). Littermate controls, were injected with vehicle only. In the “multiple low dose” STZ experiment, mice
Results Generation of Transgenic Mice That Express Human VAP-1 in Specific Tissues To determine the source and regulation of serum VAP-1 (sVAP), transgenic mice were created with either a mouse tie-1 promoter or a murine aP2 enhancer/promoter driving the expression of full-length human VAP-1 (online Figure 1 in the online data supplement), including the transmembrane domain, specifically to endothelial and adipose cells respectively. Transgenic founder mice were identified by PCR and confirmed by Southern blot analysis using transgene-specific primers and probes. Two independent lines of mice with each transgene were selected for further analyses (Table).
Transgene Expression Is Targeted to Specific Cell Populations In human tissue, VAP-1 exists as a 170- to 180-kDa homodimeric sialoglycoprotein24 and in the transgenic tissues a
Downloaded from circres.ahajournals.org by on May 23, 2011
52
Circulation Research
July 9, 2004
Figure 1. Tissue-specific transgene expression. Epididymal (A and B) and brown fat (C) sections from aP2hVAP-1 (A and C) and nontransgenic mice (B) stained with human VAP-1–specific antibody, TK8 –14-FITC. D through F, Fluorescent micrographs of frozen liver sections made from mTIEhVAP-1 (D and F) and nontransgenic mice (E) 5 minutes after the intravenous injection of FITC conjugated anti-human VAP-1 (D and E) and anti-chicken T cell control (F) antibodies. G through I, Representative fluorescent micrographs of liver frozen sections from mTIEhVAP-1 mice double stained with a CD36 specific antibody, sc-13572 PE (red stain), and a human VAP-1 specific antibody, JG2.10-FITC (green stain). Sections were photographed under UV light with a PE filter (G) and FITC filter (H), and then digitally superimposed (I, double staining is yellow). Asterisks indicate hepatic portal veins; arrows, sinusoidal endothelium. Bar⫽100 m for all micrographs.
VAP-1 protein with the appropriate sialic acid decorations and the expected molecular mass was detected (online Figure 2). To characterize the expression pattern of human VAP-1 in the mTIEhVAP-1 and aP2hVAP-1 transgenic lines, immunohistochemistry and ELISA were performed using antibodies specific for human VAP-1. Detectable human VAP-1 protein expression was found in 6 of 18 mTIEhVAP-1 lines. In lines E25 and E35, strong protein expression was restricted to the endothelial cells of all analyzed organs (liver, heart, lung, spleen, kidney, small intestine, mucosal lymph node, peripheral lymph node, eye, skin, muscle, and adipose; data not shown) and was not found in other cell types. Conversely, strong expression was found only in the adipocytes of white (Figure 1A) and brown (Figure 1C) adipose tissue from aP2hVAP-1 mice (lines A31 and A33). No VAP-1 protein expression was found in tissues other than adipose (liver, heart, lung, pancreas, spleen, kidney, small intestine, thymus, testis, muscle, and brain; data not shown). No staining of endogenous mouse VAP-1 was seen in any tissue, including smooth muscle, thus confirming the specificity of the antibodies for human VAP-1 (Figure 1B; data not shown). To determine if the human VAP-1 protein was on the endothelial cell surface, frozen tissue sections were prepared after the intravenous injection of a FITC-conjugated anti– human VAP-1 antibody. Human VAP-1–specific staining was found on the endothelial cell surface of vessels in the mTIEhVAP-1 liver (Figure 1 D), kidney, pancreas, and lymph node, but not in nontransgenic tissues (Figure 1 E). Liver sections were also double stained with an anti-human VAP-1 antibody and a marker for sinusoidal endothelium, anti-CD36 (Figure 1G).25 The anti– human VAP-1 antibodies stained the terminal hepatic venules and sinusoidal endothelium of mTIEhVAP-1 sections (Figure 1H) but not in the nontransgenic liver (not shown). Clear double staining of the sinusoidal endothelium was also evident in the mTIEhVAP-1
sections (Figure 1I). Thus, as predicted, the transgene expression patterns in the mTIEhVAP-1 and aP2hVAP-1 transgenic animals are different in that expression driven by the tie-1 promoter is directed to endothelial cells, whereas protein expression driven by the aP2 promoter is only directed to adipocytes.
Human VAP-1 Overexpression Increases Total SSAO Activity We next tested whether overexpression of hVAP-1 leads to increased SSAO activity in the transgenic tissues. First, adipose tissue lysates were incubated with saturating concentrations of either benzylamine or methylamine, as appropriate enzyme substrates, and clorgyline, as an inhibitor of intracellular MAO. The generation of the SSAO reaction product H2O2 was determined by using a sensitive fluorometric assay. With this technique, increased SSAO activity was found in the adipose tissue lysates of the aP2hVAP-1 mice compared with the nontransgenic mice (Figure 2A). Second, an independent, radiochemical assay was used to measure the rate of [14C]benzylamine deamination in liver lysates. Because the application of the fluorometric assay to liver lysates is complicated by very low SSAO activity and high background from the massive release of endogenous H2O2, during the solubilization of hepatocytes, it was more suitable to use the radiochemical assay with subsaturating concentrations of [14C]benzylamine. With this assay, a significant increase in SSAO activity was also found in liver lysates from mTIEhVAP-1 mice (Figure 2B).
Soluble Human VAP-1 Is Found in the Serum of mTIEhVAP-1 Transgenic Mice Although SSAO activity has been found in the serum of human patients and in experimental animals, the molecular and cellular source of this activity has not been defined. To
Downloaded from circres.ahajournals.org by on May 23, 2011
Stolen et al
Origins of Serum SSAO
53
Figure 3. mTIEhVAP-1 transgene restores SSAO activity to mouse VAP-1 knockout serum. A, SSAO activity is absent from the serum of nontransgenic VAP-1 knockout mice (NULL/NULL NT) but is present in mTIEhVAP-1 transgenic/VAP-1 knockout mice (NULL/NULL TG) and in nontransgenic heterozygous null mice (WT/NULL NT). B, Human VAP-1 is present in the serum of transgenic (NULL/NULL TG) but not in nontransgenic (NULL/ NULL NT) knockout mice or nontransgenic heterozygous null (WT/NULL NT) mice. All results are from female mice on a mixed FVB/N and 129S6 background
antibody, JG2.10 (data not shown). Furthermore SSAO activity is restored to the serum of VAP-1 knockout mice by the mTIEhVAP-1 transgene (Figure 3). Thus the transmembrane form of human VAP-1 expressed on endothelial cells can act as a source of sVAP-1 and SSAO activity. Figure 2. Transgenic expression of human VAP-1 increases total SSAO activity in both tissue lysates and serum. A, Rate of SSAO-mediated hydrogen peroxide formation in adipose tissue lysates was determined in a fluorometric assay using the MAO inhibitor clorgyline and 1 mmol/L of the indicated substrates: methylamine (MA) and benzylamine (BA). B, Rate of benzylamine oxidation by liver tissue lysates and serum samples was determined radiochemically. Number of samples is indicated at the base of each bar. All results are from male mice. White bars indicate nontransgenic; gray bars, aP2hVAP-1; black bars, mTIEhVAP-1.
determine if full-length VAP-1 expressed on endothelial cells or adipocytes could act as a source of this activity, the serum of transgenic mice was examined. By using a quantitative sandwich ELISA, with anti-VAP-1 antibodies that are specific for the human VAP-1 protein, a soluble form of VAP-1 was found at high levels in the serum of mTIEhVAP-1 transgenic mice and at very low levels in the serum of aP2hVAP-1 mice but not in nontransgenic controls (Table). Whereas the average concentration of soluble human VAP-1 in the mTIEhVAP-1 mice (10 to 48 ng/mL; Table) was lower than that found in human AB serum samples (92.6⫾13.4 ng/mL, n⫽61), it was of a similar magnitude, suggesting normal regulation of its release. Increased SSAO activity, as determined by the rate of [14C]benzylamine deamination, was also found in the serum of the mTIEhVAP-1 mice compared with nontransgenic controls (Figure 2B) and could be diminished by immunodepletion with the human VAP-1–specific
Human VAP-1 Is Released From Adipocytes and Endothelial Cells During Biological Stress Having determined that the transmembrane form of VAP-1 can act as a source of circulating SSAO, we next investigated whether the levels present in mouse serum could vary. A higher level of human sVAP-1 was found in mTIEhVAP-1 homozygous versus heterozygous mice, indicating that this level could be influenced by gene dose (Table). The age of the mTIEhVAP-1 mice did not correlate with sVAP-1 levels (heterozygous males r⫽⫺0.09, P⫽0.6; females r⫽⫺0.20, P⫽0.4); however, higher human sVAP-1 levels were found in the serum of male versus female mice (Table). The zygosity and gender-specific differences were found in two independent lines of mTIEhVAP-1 mice (E25 and E35) and in one line of aP2hVAP-1 mice (A33), indicating that the differences are not due to the site of transgene integration, but are more likely due to differential promoter activity or protein shedding. No gender-specific difference in serum SSAO activity was found in nontransgenic samples (male, 22.7⫾5.6 pmol/hr per mL, n⫽11 versus female, 26.7⫾2.2 pmol/hr per mL, n⫽11). Previous studies have found increased levels of sVAP-1 and SSAO in the plasma of patients with diabetes and in animal models of diabetes. Thus, we decided to test whether STZ treatment would lead to increased SSAO activity and sVAP-1 levels in the mTIEhVAP-1 and aP2hVAP-1 trans-
Downloaded from circres.ahajournals.org by on May 23, 2011
54
Circulation Research
July 9, 2004
Figure 5. Fasting increases circulating human VAP-1 levels in mTIEhVAP-1 mice. Mean values for fed vs fasted female mice of the same litter are connected by lines. Duration of food withdrawal and the number of mice (fed/fasted) from each litter is shown next to its symbol. Bars represent the mean⫾SEM for all fed (n⫽23) and fasted (n⫽13) heterozygous female mice, including mice not paired with fed or fasted litter mates.
Figure 4. Serum SSAO and VAP-1 increase in response to experimental diabetes. A, Effect of STZ treatment on serum SSAO activity. B, Effect of STZ treatment on circulating human VAP-1 levels. Non-Tg indicates nontransgenic; E35, mTIEhVAP-1 line E35; A33, aP2hVAP-1 line A33; buffer, vehicleonly injected mice; H.STZ, high-dose STZ–injected; L.STZ, lowdose STZ–injected.
genic mice. We used two different STZ regimens to induce experimental diabetes. First, high-dose STZ exposure was used to induce -cell death and acute hyperglycemia. After 2 to 5 days of hyperglycemia, serum and tissue samples were collected. Using a radiochemical assay, increased SSAO activity was found in the serum of the STZ injected nontransgenic, mTIEhVAP-1, and aP2hVAP-1 mice compared with buffer injected controls (Figure 4A). A marked increase in hVAP-1 was also found in the serum of both mTIEhVAP-1 and in aP2hVAP-1 mice (Figure 4B), thus indicating that sVAP-1 can be released from both endothelial cells and adipocytes in diabetes. No change was found in the total hVAP-1 content of heart, liver, kidney,
and pancreatic lysates of the same mTIEhVAP-1 mice (data not shown). Second, a multiple low-dose regimen of STZ was used to induce pancreatic insulitis and thus avoid any immediate effects from STZ itself.26,27 In this model, the onset of hyperglycemia was delayed but then maintained for several weeks before sample collection at 13 weeks. Using this model, dramatically elevated sVAP-1 levels were again found in the serum of the hyperglycemic mTIEhVAP-1 mice several weeks after exposure to STZ (Figure 4B). Finally, we postulated that ketone bodies might regulate sVAP-1 release. To elevate the blood level of ketones, we fasted the mice for 44 to 48 hours and then compared their sVAP-1 levels to littermate controls that had been similarly handled and fed ad libitum. Although there was an increase in sVAP-1 levels in the fasted mTIEhVAP-1 mice compared with their fed transgenic littermates (paired t test of mean litter values, P⫽0.03; Figure 5) and compared with all fed female mTIEhVAP-1 mice (unpaired t test P⫽0.001; Figure 5), there was no correlation with ketone body levels (r⫽0.20, P⫽0.5). To determine if fasting could also trigger sVAP-1 release from adipose cells, we measured the sVAP-1 level in plasma taken from aP2hVAP-1 mice before and after a 40-hour fasting period. There was no detectable increase in sVAP-1 found (data not shown).
Human VAP-1 Overexpression Increases Lymphocyte Adhesion To assess the functional significance of increased VAP-1 expression, a Stamper-Woodruff assay was used to evaluate the effect of VAP-1 on lymphocyte binding to lymph node high endothelial venules. Initially, in the absence of blocking antibodies, both mouse and human lymphocytes preferen-
Downloaded from circres.ahajournals.org by on May 23, 2011
Stolen et al
Figure 6. Lymphocyte binding to transgenic high endothelial venules is increased. Stamper-Woodruff binding assays were performed using lymph node frozen sections from transgenic or nontransgenic mice, anti– human VAP-1 mAbs (TK8 –14 and 1B2), or isotype matched control mAbs (HB116 and 7C7) and human (black bars, n⫽4) or mouse (white bars, n⫽6) lymphocytes. Asterisks indicate one sample t tests comparing to the control relative adherence ratio of 1.0, P⬍0.05; cross, unpaired Student t tests comparing transgenic samples, P⬍0.05.
tially adhered to the transgenic venules when compared with nontransgenic sections (Figure 6). By pretreating the transgenic lymph node sections with antibodies specific to human VAP-1 (TK8 –14 and 1B2), the binding of both mouse and human lymphocytes was inhibited back to the nontransgenic level.
Human VAP-1 Overexpression Increases Hepatic Expression of Redox-Sensitive Proteins A transgene-specific increase in total liver mass was found in the mTIEhVAP-1 mice after a 15-month challenge with a high-fat diet, oral methylamine, or a control diet (ANOVA transgene effect, P⫽0.0005; for the study design, see Stolen et al)11 In particular, there was a 22% transgene-specific increase in liver mass when the mice were fed a high fat diet (Student t test, P⫽0.04). Although histological examination of liver sections taken from the mice fed a high-fat diet revealed fatty liver changes (ie, globular intracytoplasmic
Origins of Serum SSAO
55
clear spaces and Oil red-O lipid staining), these changes were not specific to the transgenic mice and no other pathological changes were identified in any group of mice. To further investigate the molecular basis of this phenotype, hepatic proteins were separated by two-dimensional gel electrophoresis and differentially expressed proteins were identified by peptide sequencing (Figure 7). A large number of differentially expressed redox-sensitive proteins were found, including peroxiredoxin 1, glutathione S-transferase P, glyoxalase I, superoxide dismutase-Mn, and aldolase 2-B isoform, when comparing the transgenic to nontransgenic samples.
Discussion Human VAP-1 is an important cell adhesion molecule with high homology to copper-dependent SSAOs4 and abundant expression on endothelial cells, adipocytes, smooth muscle cells, and the dendritic cells of germinal centers. A form of human VAP-1 is also found circulating in the blood stream.15 Until now the exact cellular origin of this circulating molecule and the biological triggers that regulate its release have remained elusive. In the present study, we used transgenic mouse technology to specifically direct expression of human VAP-1 to either endothelial cells or adipocytes. This approach allowed us to determine the following things: (1) all detectable serum SSAO in the mouse is derived from VAP-1; (2) serum SSAO activity can be derived from expression of the full-length VAP-1 gene; (3) vascular endothelial cells are a major source of circulating VAP-1 protein and SSAO activity under normal physiological conditions; (4) under conditions of biological stress, the release of additional sVAP-1/SSAO is differentially regulated so that it can originate from both endothelial cells and adipocytes, as found in experimental diabetes, or solely from endothelial cells, as was found with fasting; (5) overexpression of human VAP-1
Figure 7. Redox-sensitive protein expression is altered in mTIEhVAP-1 mice. Liver extracts prepared from mTIEhVAP-1 transgenic (A) and nontransgenic (B) mice were analyzed by 2-D electrophoresis. Pictured silver stained gels are representative of 3 transgenic and 3 nontransgenic gels with similar results. Spot pairs with altered intensity in the transgenic samples are circled or boxed. Protein spots were characterized by mass spectrometry and compared with NCBInr protein databases (a total identity scores ⬎37 indicates identity or extensive homology; P⬍0.05). 1, Diazepam binding inhibitor (223); 2, Cofilin (283) and Destrin (203); 3, Peroxiredoxin 1 (667) and Fatty acid binding protein 1 (114); 4, Peroxiredoxin 1 (492); 5, Peroxiredoxin 1 (577) and ATP synthase (364); 6, RIKEN cDNA (340) and Glyoxylase I (294); 7, Superoxide Dismutase–Mn (371) and Glutathione S-transferase P (96); 8, no identity; 9, Arginase 1 (411), Aldolase 2–B isoform (376), heterogenous nuclear ribonucleoprotein A2/B1/B0 (187), coproporphyrinogen oxidase (186), and peroxiredoxin 1 (182). Note the change in a large number of redox-sensitive proteins italicized in this list.
Downloaded from circres.ahajournals.org by on May 23, 2011
56
Circulation Research
July 9, 2004
increases the capacity of mouse endothelial cells to bind both human and mouse lymphocytes; and (6) overexpression of human VAP-1 alters hepatic expression of redox-sensitive proteins. In human diabetes and in animal models of diabetes, the plasma levels of SSAO are increased, and a significant positive correlation between SSAO activity and the development of late diabetic complications has been found.16,17 Because the deamination of endogenous substrates by SSAO produces substances that can directly damage endothelial cells, potentiate oxidative glycation, and increase oxidative stress,28,29 it is possible that this activity creates or exacerbates the damage leading to many of the various diabetic complications, such as retinopathy, nephropathy, neuropathy, and accelerated atherosclerosis. These toxic effects may be mitigated in part by the differential expression of redoxsensitive genes. In this study, we found that the expression levels of several redox-regulating proteins were changed in mice overexpressing VAP-1/SSAO (Figure 7). These changes may result from a direct signaling action of the VAP-1 molecule or the reaction byproducts: aldehyde, ammonia, and H2O2. Hydrogen peroxide in particular is a known regulator of gene expression.12–14 Alternatively, because these mice have an increased propensity for H2O2 production (Figure 2) and because H2O2 is a major reactive oxygen species that can be converted to hydroxyl-free radicals via the Fenton reaction the changes in protein expression may be secondary to SSAO-induced oxidative stress. We also show an increase in lymphocyte binding to the high endothelial venules of the transgenic lymph nodes (Figure 6). The adhesion of mouse lymphocytes to transgenic human VAP-1 illustrates the conservation and importance of the extravasation cascade during evolution. These studies when combined with other transgenic studies11,30 verify the multifunctional nature of VAP-1/SSAO and indicate that increased SSAO is not just a benign byproducts of specific disease states but that it has real physiological and pathological consequences. Although the many functions of VAP-1/ SSAO appear unrelated, they are not mutually exclusive. We propose that the SSAO enzymatic activity normally acts to form transient but covalent crosslinks between endothelial cells and amines presented on the leukocyte cell surface,7 and that in times of crisis (ie, diabetes), its ability to produce H2O2 is utilized in an attempt to stimulate the insulin signaling pathway.8 Finally, when this activity is chronically elevated, in inflammation or diabetes,31 it alters redox sensitive protein expression and promotes vascular complications.11 Previous studies have suggested several possible sources of serum SSAO activity. In cattle, a gene encoding a soluble form of SSAO has been cloned (BSAO),32 and in sheep, kinetic data suggest the existence of two different soluble SSAO enzymes.33,34 On the other hand, in humans, the N-terminal sequence of isolated serum SSAO is identical to the membrane distal sequence of VAP-1,35 and when VAP-1 is depleted from the serum of patients with diabetes all detectable SSAO activity simultaneously disappears,31 suggesting that a VAP-1 cleavage product may be the sole source. In transgenic mice that have the entire human VAP-1 coding sequence expressed with a smooth muscle ␣-actin
promoter, an increase in serum SSAO activity was found.30 In the experiments presented in this study, we show that circulating VAP-1 and serum SSAO activity may be derived from endothelial cell expression of the full-length form of the VAP-1 gene (Figure 2). This SSAO activity is absent from the serum of VAP-1 knockout mice but is restored by the mTIEhVAP-1 transgene (Figure 3), suggesting that VAP-1 is responsible for all serum SSAO activity. In this study, we have also found that with the induction of experimental diabetes both endothelial cells and adipocytes could release substantial amounts of VAP-1 into the serum (Figure 4), suggesting the existence of supplementary release triggering mechanisms during biological stress. By having determined the source of serum SSAO, work can now be done to inhibit or regulate its release and thus attenuate its action in inflammation and vascular pathology. The differences in circulating VAP-1 levels (Table) that were found between the male and female mice may reflect genderspecific differences in human VAP-1 protein cleavage. Because the differences were found in both mTIEhVAP-1 and aP2hVAP-1 transgenic mice, it is unlikely to be a result of gender-specific promoter activity. Although similar differences in sVAP-1 or SSAO have not been detected in nontransgenic mice or in human patients,15 VAP-1 overexpression may accentuate a real gender difference. Interestingly, increased SSAO production in males could be hypothesized to contribute to the increased rate of atherosclerosis formation found in males. Although ketone bodies were not found to directly regulate sVAP-1 levels, there does appear to be similarities between diabetes and fasting in that both lead to VAP-1 release from endothelial cells. Recent clinical studies have found that circulating human VAP-1/SSAO levels rapidly respond, in an inverse manner, to changes in plasma insulin,31 thus suggesting that insulin could be the common regulator in both diabetes and fasting. In conclusion, the use of transgenic mouse technology for these experiments has enabled us to circumvent the lack of good antibodies for directly assessing endogenous VAP-1 in animal models. In addition, by using the human VAP-1 transgene, we were allowed to use the species specificity of our antibodies to identify the cellular source of circulating VAP-1, something that would not be possible using human VAP-1 antibodies alone. This approach has also provided physiologically relevant information that verifies the role of VAP-1 in lymphocyte binding and indicates that VAP-1/ SSAO activity can regulate redox-sensitive protein expression.
Acknowledgments This work was supported by grants from NIH, the Juvenal Diabetes Foundation International, the European Union (QLG1-CT-199900295), the Finnish Academy, the Sigrid Juselius Foundation, and the Technology Development Center of Finland. The authors wish to acknowledge the expert technical assistance of Maritta Pohjansalo and Suvi Nevalainen. Jussi Niemela¨ is also thanked for sharing his tie-1 promoter sequence data.
References 1. Salmi M, Tohka S, Berg EL, Butcher EC, Jalkanen S. Vascular adhesion protein 1 (VAP-1) mediates lymphocyte subtype-specific, selectin-
Downloaded from circres.ahajournals.org by on May 23, 2011
Stolen et al
2.
3. 4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14. 15.
16.
17.
18.
independent recognition of vascular endothelium in human lymph nodes. J Exp Med. 1997;186:589 – 600. Salmi M, Kalimo K, Jalkanen S. Induction and function of vascular adhesion protein-1 at sites of inflammation. J Exp Med. 1993;178: 2255–2260. Salmi M, Jalkanen S. A 90-kilodalton endothelial cell molecule mediating lymphocyte binding in humans. Science. 1992;257:1407–1409. Smith DJ, Salmi M, Bono P, Hellman J, Leu T, Jalkanen S. Cloning of vascular adhesion protein-1 reveals a novel multifunctional adhesion molecule. J Exp Med. 1998;188:17–27. Bono P, Salmi M, Smith DJ, Jalkanen S. Cloning and characterization of mouse vascular adhesion protein-1 reveals a novel molecule with enzymatic activity. J Immunol. 1998;160:5563–5571. Lyles GA. Mammalian plasma and tissue-bound semicarbazide-sensitive amine oxidases: biochemical, pharmacological and toxicological aspects. Int J Biochem Cell Biol. 1996;28:259 –274. Salmi M, Yegutkin GG, Lehvonen R, Koskinen K, Salminen T, Jalkanen S. A cell surface amine oxidase directly controls lymphocyte migration. Immunity. 2001;14:265–276. Enrique-Taranco´n G, Marti L, Morin N, Lizcano JM, Unzeta M, Sevilla L, Camps M, Palacı´n M, Testar X, Carpe´ne´ C, Zorzano A. Role of semicarbazide-sensitive amine oxidase on glucose transport and GLUT4 recruitment to the cell surface in adipose cells. J Biol Chem. 1998;273: 8025– 8032. Yu PH, Zuo DM. Formaldehyde produced endogenously via deamination of methylamine: a potential risk factor for initiation of endothelial injury. Atherosclerosis. 1996;120:189 –197. Yu PH, Zuo DM. Aminoguanidine inhibits semicarbazide-sensitive amine oxidase activity: implications for advanced glycation and diabetic complications. Diabetologia. 1997;40:1243–1250. Stolen CM, Madanat R, Marti L, Kari S, Yegutkin GG, Sariola H, Zorzano A, Jalkanen S. Semicarbazide sensitive amine oxidase overexpression has dual consequences: insulin mimicry and diabetes-like complications. FASEB J. 2004;18:702–704. Griendling KK, Sorescu D, Lassegue B, Ushio-Fukai M. Modulation of protein kinase activity and gene expression by reactive oxygen species and their role in vascular physiology and pathophysiology. Arterioscler Thromb Vasc Biol. 2000;20:2175–2183. Bradley JR, Johnson DR, Pober JS. Endothelial activation by hydrogen peroxide. Selective increases of intercellular adhesion molecule-1 and major histocompatibility complex class I. Am J Pathol. 1993;142: 1598 –1609. Suzuki YJ, Forman HJ, Sevanian A. Oxidants as stimulators of signal transduction. Free Radic Biol Med. 1997;22:269 –285. Kurkija¨rvi R, Adams DH, Leino R, Mo¨tto¨nen T, Jalkanen S, Salmi M. Circulating form of human vascular adhesion protein-1 (VAP-1): increased serum levels in inflammatory liver diseases. J Immunol. 1998; 161:1549 –1557. Gronvall-Nordquist JL, Backlund LB, Garpenstrand H, Ekblom J, Landin B, Yu PH, Oreland L, Rosenqvist U. Follow-up of plasma semicarbazidesensitive amine oxidase activity and retinopathy in Type 2 diabetes mellitus. J Diabetes Complications. 2001;15:250 –256. Boomsma F, Derkx FHM, van den Meiracker AH, Man in’t Veld AJ, Schalekamp MADH. Plasma semicarbazide-sensitive amine oxidase activity is elevated in diabetes mellitus and correlates with glycosylated haemoglobin. Clinical Science (Colchester). 1995;88:675– 679. Nilsson SE, Tryding N, Tufvesson G. Serum monoamine oxidase (MAO) in diabetes mellitus and some other internal diseases. Acta Med Scand. 1968;184:105–108.
Origins of Serum SSAO
57
19. Tryding N, Nilsson SE, Tufvesson G, Berg R, Carlstro¨m S, Elmfors B, Nilsson JE. Physiological and pathological influences on serum monoamine oxidase level: effect of age, sex, contraceptive steroids and diabetes mellitus. Scand J Clin Lab Invest. 1969;23:79 – 84. 20. McEwen CM, Jr., Harrison DC. Abnormalities of serum monoamine oxidase in chronic congestive heart failure. J Lab Clin Med. 1965;65: 546 –559. 21. Boomsma F, van Veldhuisen DJ, de Kam PJ, Man in’t Veld AJM, Mosterd A, Lie KI, Schalekamp MADH. Plasma semicarbazide-sensitive amine oxidase is elevated in patients with congestive heart failure. Cardiovasc Res. 1997;33:387–391. 22. Kurkijarvi R, Yegutkin GG, Gunson BK, Jalkanen S, Salmi M, Adams DH. Circulating soluble vascular adhesion protein 1 accounts for the increased serum monoamine oxidase activity in chronic liver disease. Gastroenterology. 2000;119:1096 –1103. 23. Jalkanen ST, Butcher EC. In vitro analysis of the homing properties of human lymphocytes: developmental regulation of functional receptors for high endothelial venules. Blood. 1985;66:577–582. 24. Salmi M, Jalkanen S. Human vascular adhesion protein-1 (VAP-1) is a unique sialoglycoprotein that mediates carbohydrate-dependent binding of lymphocytes to endothelial cells. J Exp Med. 1996;183:569 –579. 25. Terpstra V, van Amersfoort ES, van Velzen AG, Kuiper J, van Berkel TJ. Hepatic and extrahepatic scavenger receptors: function in relation to disease. Arterioscler Thromb Vasc Biol. 2000;20:1860 –1872. 26. Kondo S, Iwata I, Anzai K, Akashi T, Wakana S, Ohkubo K, Katsuta H, Ono J, Watanabe T, Niho Y, Nagafuchi S. Suppression of insulitis and diabetes in B cell-deficient mice treated with streptozocin: B cells are essential for the TCR clonotype spreading of islet-infiltrating T cells. Int Immunol. 2000;12:1075–1083. 27. Kolb-Bachofen V, Epstein S, Kiesel U, Kolb H. Low-dose streptozocininduced diabetes in mice: electron microscopy reveals single-cell insulitis before diabetes onset. Diabetes. 1988;37:21–27. 28. Yu PH. Deamination of methylamine and angiopathy; toxicity of formaldehyde, oxidative stress and relevance to protein glycoxidation in diabetes. J Neural Transm Suppl. 1998;52:201–216. 29. Exner M, Hermann M, Hofbauer R, Kapiotis S, Quehenberger P, Speiser W, Held I, Gmeiner BM. Semicarbazide-sensitive amine oxidase catalyzes endothelial cell-mediated low density lipoprotein oxidation. Cardiovasc Res. 2001;50:583–588. 30. Gokturk C, Nilsson J, Nordquist J, Kristensson M, Svensson K, Soderberg C, Israelson M, Garpenstrand H, Sjoquist M, Oreland L, Forsberg-Nilsson K. Overexpression of semicarbazide-sensitive amine oxidase in smooth muscle cells leads to an abnormal structure of the aortic elastic laminas. Am J Pathol. 2003;163:1921–1928. 31. Salmi M, Stolen C, Jousilahti P, Yegutkin GG, Tapanainen P, Janatuinen T, Knip M, Jalkanen S, Salomaa V. Insulin-regulated increase of soluble vascular adhesion protein-1 in diabetes. Am J Pathol. 2002;161: 2255–2262. 32. Mu D, Medzihradszky KF, Adams GW, Mayer P, Hines WM, Burlingame AL, Smith AJ, Cai D, Klinman JP. Primary structures for a mammalian cellular and serum copper amine oxidase. J Biol Chem. 1994;269:9926 –9932. 33. Boomsma F, van Dijk J, Bhaggoe UM, Bouhuizen AM, van den Meiracker AH. Variation in semicarbazide-sensitive amine oxidase activity in plasma and tissues of mammals. Comp Biochem Physiol C Toxicol Pharmacol. 2000;126:69 –78. 34. Elliott J, Callingham BA, Sharman DF. Amine oxidase enzymes of sheep blood vessels and blood plasma: a comparison of their properties. Comp Biochem Physiol C. 1992;102:83– 89. 35. Jalkanen S, Salmi M. Cell surface monoamine oxidases: enzymes in search of a function. EMBO J. 2001;20:3893–3901.
Downloaded from circres.ahajournals.org by on May 23, 2011
1
Online Data Supplements Expanded Materials and Methods Generation of the Transgene Constructs To generate the mTIEhVAP-1 and aP2hVAP-1 transgenes (Additional Figure 1) a plasmid, pVAP-1, containing the complete 4040 base pair human VAP-1 cDNA (genebank: NM_003734) was used1. This VAP-1 construct contained the entire open reading frame as well as the 3' polyadenylation signals inserted into the EcoRI and NotI sites of pcDNA3 (Invitrogene). For construction of the mTIEhVAP-1 construct, the 1.0 kb HindIII-ApaI fragment of the mouse tie-1 promoter containing the AflII-ApaI fragment, which was previously shown to be endothelial cell specific in transgenic mice 2, was cloned after klenow treatment into the HindIII-BamHI sites of pVAP-1. For construction of aP2hVAP-1, the KpnI site of the pVAP-1 polylinker was first changed to SacI with a cohesive end adapter duplex (5'GATCCCCGCGGGACCCA/ 5'AGCTTGGGTCCCGCGGG). The adipose tissue specific, 5.4 kb, adipocyte P2 (aP2) enhancer/promoter3 was then moved from plasmid pA22B (obtained from Philippe Rouet, INSERM, Toulouse, France) into the SacI-HindIII site of pVAP-1 after removal of the NotINotI polylinker sequences from pA22B. Sequencing of the tie-1 promoter construct identified an insertion and a base pair difference at nucleotides +455 and +458 of the published tie-1 promoter sequence (AGCC vs A-CG; GI:1086920). This most likely reflects an error in the original promoter sequence because the new sequence was additionally verified by comparison to the mouse genome (RP23-263B20 & RP23-145E1). Hence, these differences are not predicted to affect the spatiotemporal specificity of expression of the promoter.
Downloaded from circres.ahajournals.org by on May 23, 2011
2 Generation and Identification of Transgenic Mice Transgenic mice were generated by microinjection on the FVB/n inbred genetic background at the University of Helsinki (mTIEVAP-1 line E35) and at the Turku University Transgenic Mouse Facility (all other lines). Each founder transgenic mouse was bred to non-transgenic FVB/n mice to establish and maintain independent lines. All studies were performed in accordance with the local guidelines for animal care. Transgenic mice were identified by PCR screening of purified genomic DNA (GentraSystems, Inc.). CS-1 and CS-18 are sense primers that hybridize to the tie-1 promoter domain of the mTIEhVAP-1 transgene and CS-27 is a sense primer that hybridizes to the linker sequence between the aP2 and hVAP-1 domains of the aP2hVAP-1 transgene. When paired with the human VAP-1 specific antisense primer, CS-2, each of the sense primers amplifies a product that is specific for its respective transgene (CS-1, 303 bp; CS-18, 507bp; CS-27, 296 bp). For the screening of transgenic mice by Southern blot analysis genomic DNA was digested with selected restriction enzymes (SpeI or BglII), separated by electrophoresis on 1% agarose gels, transferred to Hybond-N nylon membranes (Amersham Pharmacia Biotech), and hybridized with α[32P] dCTP random-prime labeled probes. Two human VAP-1 specific probes were used to screen the mTIEhVAP-1 mice: a 0.8 Kb DraIII-EcoRI fragment isolated from the plasmid pTIEhVAP-1 and a 2.1 Kb KpnI-KpnI fragment from the plasmid pVAP-1. The copy number was determined by comparison of genomic DNA samples with 10 pg of pTIEhVAP-1, representing approximately one copy per haploid genome.
Downloaded from circres.ahajournals.org by on May 23, 2011
3
Generation of mTIEhVAP-1 Transgenic/VAP-1 Knockout Mice The mTIEhVAP-1 line E35 mice were crossed to VAP-1 knockout mice that were previously created by using conventional gene targeting techniques to replace the mouse VAP-1 gene with a nonfunctional mutant-allele (C. Stolen, unpublished data). The mTIEhVAP-1 transgene, mouse VAP-1 mutant-allele (null) and endogenous mouse VAP-1 allele (wt) were all identified by PCR screening of purified genomic DNA with specific primers and verified immunohistochemically with human and mouse VAP-1 antibodies.
Analysis of Tissue Specific Transgene Expression After euthanasia, by cervical dislocation or carbon dioxide inhalation, organs were collected, immediately placed in OCT compound (Sakura Finetek Europe B.V., The Netherlands), snap frozen with liquid nitrogen, and stored for a short period at -70°C. Frozen sections (5 µm) were cut, acetone fixed, and used for immunohistochemical analyses. Warmed sections were incubated for 30 minutes with fluorescein isothiocyanate (FITC) conjugated monoclonal anti-human VAP-1 antibodies, TK8-14 (20µg/ml) or JG2.10 (20µg/ml, kind gift of Eugene Butcher), and/or phycoerythrin (PE) conjugated monoclonal anti-mouse CD36 (CD36 (ME542) PE: sc-13572 PE, Santa Cruz Biotechnology), and as a negative control FITC conjugated anti-chicken T cell antibody, 3G6 (20µg/ml). The sections were then rinsed twice with PBS, mounted with fluoromount-G (Southern Biotechnology Associated, Birmingham, AL), and viewed with a fluorescence microscope. Human tonsil frozen sections were used as positive controls for human VAP-1 staining. Digital photomicrographs were created using AnalySIS version 3.00 software (Soft Imaging System GmbH).
Downloaded from circres.ahajournals.org by on May 23, 2011
4 Analysis of Endothelial Cell Surface Expression Organs were collected and frozen into OCT blocks five minutes following an intravenous injection (tail vein) of FITC conjugated anti-human VAP-1 (TK8-14-FITC, 100µg/mouse) or anti-chicken T cell (3G6-FITC, 100µg/mouse) antibodies. Frozen sections (5 µm) were then cut, air dried, acetone fixed, rinsed twice with PBS, mounted with flouromount-G, and viewed with a fluorescence microscope. Identical results were also obtained without acetone fixation.
Immunoblotting Analyses Tissue samples from heart, liver and adipose tissue were lysed in buffer containing 10 mM Tris-Base (pH 7.4), 150 mM NaCl, 1.5 mM MgCl2, 1% NP-40, 1% Aprotinin and 1 mM PMSF. Insoluble material was removed by centrifugation and supernatants were depleted of contaminating immunoglobulins by incubating with Protein G-Sepharose beads (Pharmacia, Biotech AB, Uppsala, Sweden). Thereafter, 50-µl aliquots of the samples were treated with 20 mU Vibrio cholerae neuraminidase (Dade-Behring, Marburg, Germany) for 4 h, with 5 mU N-glycanase (Glyco, CA, USA) overnight or first with 20 mU neuraminidase for 4 h and then with ≥1 mU O-glycanase (Glyko) overnight. All glycosidase digestions were performed at 37°C. For gel electrophoresis, the samples were mixed with non-reducing Laemmli’s sample buffer, incubated at 37°C for 20 min, loaded in protein concentrations that were equal or justified by VAP-1 band intensities, and run on 5-12.5% SDS-PAGE gels. The proteins were transferred onto nitrocellulose membranes (Hybond-ECL, Amersham, Buckinghamshire, UK) and detected with primary antibodies JG-2.10 (5 µg/ml), 2D10 (5
µg/ml), 3G6 (5 µg/ml), and 9B5 (5 µg/ml), HRP-conjugated secondary antibody, and an enhanced chemiluminescence detection system (ECL kit for Western blotting, Amersham).
Downloaded from circres.ahajournals.org by on May 23, 2011
5 Plasma, Serum and Tissue Lysate Preparation Blood was collected from the tail vein with heparin coated syringes, and centrifuged 10 minutes at 2083 g for plasma collection. Immediately after euthanasia blood was collected from the right ventricle of the heart, allowed to clot at room temperature for 30 to 60 minutes, and centrifuged 10 minutes at 2083 g for serum separation. Tissue samples were cut into small pieces, lysed in an equal volume of lysis buffer (PBS and 0.2 % Triton-X 100), rocked overnight at 4°C, and centrifuged 30 minutes at 17383 g prior to collection of supernatants. All samples were stored at -70°C. Protein concentration was determined using the BCA protein assay according to manufacture instructions (Pierce, Rockford, IL).
Fasting Five litters of female, heterozygous mTIEhVAP-1 mice, were separated into individual cages and fed ad libitum or were fasted for 44 to 48 hours before euthanasia and serum collection. Plasma was collected from the tail vein of 8 heterozygous aP2hVAP-1 line 33 mice before and after 40 fours of fasting. The amount of ketone bodies was determined by testing the serum and plasma samples for βhydroxybutyrate with a MediSense Precision Xtra Plus sensor (Abbot Oy, Espoo, Finland).
Stamper-Woodruff Adhesion Assay Briefly, 8 µM frozen sections were freshly cut from pools of transgenic and non-transgenic peripheral lymph nodes (each pool was from at least 5 mice). The sections were encircled with a wax-pen rings and preincubated with 100 µl of the appropriate antibodies diluted in RPMI 1640 supplemented with 10% fetal calf serum and 10 mM Hepes while rotating (60 rpm) for 30 min. at +7 °C. Meanwhile, mesenteric lymph node lymphocytes were isolated from non-transgenic mice and peripheral blood lymphocytes from healthy human volunteers using Ficoll-centrifugation. Thereafter, 2 x 106 freshly
Downloaded from circres.ahajournals.org by on May 23, 2011
6 isolated lymphocytes were laid over the sections and incubated under constant rotation for 30 min at +7°C. The adherent cells were then fixed with 1% glutaraldehyde in PBS. To facilitate direct comparison between the different experiments and between the mouse and human lymphocytes, the relative adherence ratios (RAR) were calculated. An RAR value of 1.0 was arbitrarily set for the average number of lymphocytes bound to the venules of wildtype lymph nodes in the presence of the control antibodies. A minimum of 100 high endothelial venules/sample were counted in each experiment.
Two-dimensional PAGE and peptide sequencing. Liver samples were isolated, minced in 3 volumes of rehydration buffer (7M urea, 2M thiourea, 2% NP-40, 0.5% IPG-buffer 3-10 NL, 2 mM TBP, trace bromophenol blue), and rocked at 4°C. After removal of insoluble materials by centrifugation at 17383 g for 30 min the liver lysates were loaded onto immobilized pH gradient strips (pH 3-10, non-linear, BioRad Laboratories, Hercules, CA), focused isoelectrically for 43500 Vh, submitted to SDS/PAGE on 12% polyacrylamide gels, and silver stained. Protein spots that varied consistently between transgenic and non-transgenic samples were selected for in-gel trypsin digestion, extraction, and mass spectrometry analysis. LC-MS/MS data anlysis was performed with ProID software and Mascot (www.matrixscience.con) searches of NCBInr databases (taxonomy mouse and rodent) were used to identify protein spots with a high number of peptide matches.
Statistical methods Unless otherwise indicated all data are presented as mean±SEM. Statistical comparisons were performed using Microsoft Excel, GraphPad Prism 3.0, and SPSS software. Student's t-tests (paired
Downloaded from circres.ahajournals.org by on May 23, 2011
7 and unpaired) and one sample t-tests were used to compare groups of data and correlations were calculated using Pearson's correlation coefficient.
Additional Figures and Supporting Information Additional Figure 1. Diagram of the DNA expression constructs. The vectors contain the human VAP-1 cDNA (filled boxes) including the open reading frame (black), transmembrane domain (tm), and polyadenylation signal (pA). The endothelial cell specific, tie-1 promoter and adipocyte specific, aP2 enhancer/promoter were used to create the transcription units. The 4.7 kb, mTIEhVAP-1 and the 9.4 kb, aP2hVAP-1 transgenes were isolated with AflII and NotI or HindIII and NotI respectively.
Additional Figure 2. Transgenic human VAP-1 maintains a characteristic glycosylation pattern. Human and mTIEhVAP-1 tissue lysates were subjected to digestion before separation with SDS-PAGE and immunoblotted with human VAP-1 specific mAb, JG2.10, or isotype matched negative control, 9B5, under non-reducing conditions. A: Human VAP-1 expressed from endothelial cells and adipocytes displays characteristic electrophoretic mobility patterns4, when treated with V. cholerae sialidase (Sial), sialidase plus O-glycanase (Sial + O-glyc), or N-glycanase (N-glyc). The decreased electrophoretic mobility after sialidase treatment is consistent with the removal of a large number of negatively charged sialic acids residues5 B: VAP-1 expressed in human liver displays a faster electrophoretic mobility in both untreated and sialidase treated samples than similarly treated human VAP-1 expressed in transgenic mouse liver. This difference between human and mouse expression of the same coding sequence, human VAP-1, may reflect additional posttranslational modifications that could subsequently contribute to the broader SSAO substrate specificity in mice than in men6. Molecular weight standards are listed on the right. These Figures are representative of several similar experiments.
Downloaded from circres.ahajournals.org by on May 23, 2011
8
References 1.
Smith DJ, Salmi M, Bono P, Hellman J, Leu T, Jalkanen S. Cloning of vascular adhesion protein-1 reveals a novel multifunctional adhesion molecule. J Exp Med. 1998;188:17-27.
2.
Korhonen J, Lahtinen I, Halmekytö M, Alhonen L, Jänne J, Dumont D, Alitalo K. Endothelialspecific gene expression directed by thetie gene promoter in vivo. Blood. 1995;86:1828-35.
3.
Ross SR, Graves RA, Greenstein A, Platt KA, Shyu HL, Mellovitz B, Spiegelman BM. A fatspecific enhancer is the primary determinant of gene expression for adipocyte P2 in vivo. Proc Natl Acad Sci U S A. 1990;87:9590-4.
4.
Salmi M, Jalkanen S. Different forms of human vascular adhesion protein-1 (VAP-1) in blood vessels in vivo and in cultured endothelial cells: implications for lympohocyte-endothelial cell adhesion models. Eur J Immunol. 1995;25:2803-12.
5.
Segrest JP, Jackson RL. Molecular Weight Determination of Glycoproteins by Polyacrylamide Gel Electrophoresis in Sodium Dodecyl Sulfate. Methods Enzymol. 1972;28:54-63.
6.
Bono P, Jalkanen S, Salmi M. Mouse vascular adhesion protein-1 (VAP-1) is a sialoglycoprotein with enzymatic activity and is induced in diabetic insulitis. Am J Pathol. 1999;155:1613-24.
Downloaded from circres.ahajournals.org by on May 23, 2011
mTIEhVAP-1
tie-1
tm
hVAP-1
4.7 kb
Afl II
pA
Not I
aP2hVAP-1 aP2 enhancer/promoter Hind III
tm
hVAP-1
9.4 kb
Downloaded from circres.ahajournals.org by on May 23, 2011
pA
Not I
Additional Figure 1
Downloaded from circres.ahajournals.org by on May 23, 2011
