First published September 5, 2001; 10.1152/ ajpcell.00256.2001.—The expression and function of the endogenous inhibitor of cAMP-dependent protein kinase (PKI) in endothelial cells are unknown. In this study, overexpression of rabbit muscle PKI gene into endothelial cells inhibited the cAMP-mediated increase and exacerbated thrombin-induced decrease in endothelial barrier function. We investigated PKI expression in human pulmonary artery (HPAECs), foreskin microvessel (HMECs), and brain microvessel endothelial cells (HBMECs). RT-PCR using specific primers for human PKIα, human PKIγ, and mouse PKIβ sequences detected PKIα and PKIγ mRNA in all three cell types. Sequencing and BLAST analysis indicated that forward and reverse DNA strands for PKIα and PKIγ were of >96% identity with database sequences. RNase protection assays showed protection of the 542 nucleotides in HBMEC and HPAEC PKIα mRNA and 240 nucleotides in HBMEC, HPAEC, and HMEC PKIγ mRNA. Western blot analysis indicated that PKIγ protein was detected in all three cell types, whereas PKIα was found in HBMECs. In summary, endothelial cells from three different vascular beds express PKIα and PKIγ, which may be physiologically important in endothelial barrier function.
- protein kinase inhibitor
- endothelial resistance
- barrier function
vascular-based disorders such as ischemia-reperfusion injury, atherosclerosis, diabetes, and stroke are associated with significant endothelial cell barrier dysfunction. One signaling system, the cAMP-dependent serine/threonine protein kinase (PKA), has a significant and profound effect in the protection against endothelial barrier dysfunction. There is substantial evidence documenting the protective effects of PKA against endothelial barrier dysfunction in a variety of in vitro and in vivo experimental systems in which intracellular cAMP levels were elevated with either cAMP analogs, reagents that stimulate cAMP production (i.e., forskolin), or agents that inhibit its metabolism (i.e., phosphodiesterase inhibitors) (2-4, 14,16, 18, 20, 21, 24). We showed that overexpression of the highly specific PKA inhibitor (PKI) gene in endothelial cells was effective in fully reversing the barrier-enhancing effects of increased cAMP (16), providing direct evidence for the protective role of PKA. However, the cellular and molecular mechanisms by which PKA inhibits endothelial barrier dysfunction remain undefined. Further, the upstream regulation of endothelial PKA in this function is entirely not known.
Several observations indicate that full activation of PKA is under multiple modulatory mechanisms. In the absence of cAMP, PKA exists as an inactive tetramer consisting of two regulatory and two catalytic subunits. Binding of cAMP to the regulatory subunit lowers the affinity by four orders of magnitude, which in turn causes dissociation of the tetramer into a dimer of regulatory subunit and two active monomers of the catalytic subunit (27). Further, full activation of PKA appears to require phosphorylation of its catalytic subunit by a phosphoinositide-dependent protein kinase (5, 6). This latter observation raises the interesting possibility that PKA is regulated by its phosphorylation state, which in turn determines the cell's responsiveness to cAMP.
Another level of modulation of PKA is by endogenous PKI. PKI is a family of isoforms of distinct genes and has widespread tissue distribution (7, 15). This heat-stable protein has high affinity (0.2 nM) and specific binding for the substrate binding site (as pseudosubstrate inhibitor) on the catalytic subunit of PKA (23, 27, 30). The physiological functions of PKI have not been fully established, but PKI has been shown to contain a nuclear export signal for the PKA catalytic subunit (29). Because of its differential isoform distribution in adult tissues and its distinct interactions with the PKA catalytic subunit, PKI is hypothesized to be critical in the modulation of basal PKA activity in general cell activities (7, 9). Although endogenous PKI has not been studied in endothelial cells, it is possible that PKI may be a key modulator of PKA in endothelial cell activities, such as the permeability response. In support of this thesis, we observed that inhibition of PKA resulted in alterations in endothelial cell-cell junctions and actin filament organization, which were accompanied by impairment of the barrier (22).
The goal of this study was to investigate the involvement of PKI in regulation of endothelial permeability and PKI isoform expression in endothelial cells. Results indicated that 1) overexpression of rabbit muscle PKI decreased endothelial barrier restrictiveness, and2) endothelial cells derived from the human brain microvessel, pulmonary artery, and foreskin microvessel expressed predominantly PKIα and PKIγ isoforms. We conclude that PKIα and PKIγ may be important in modulating PKA in the regulation of vascular endothelial barrier function.
MATERIALS AND METHODS
Human dermal (foreskin) microvascular endothelial cells (HMECs) (1) were maintained in culture in MCDB 131 medium, supplemented with 5% fetal bovine serum (FBS; HyClone, Logan, UT), 10 ng/ml human epidermal growth factor (EGF), 1 μg/ml hydrocortisone, 5% penicillin-streptomycin, and 5% l-glutamine. HMECs are an immortalized cell line transformed by simian virus 40 (SV40) large T antigen and have been shown to retain endothelial cell phenotypic and functional characteristics. They exhibit the expected morphological and functional endothelial phenotypes: express and secrete von Willebrand factor, take up acetylated low density lipoprotein (LDL), form tubes when grown in Matrigel, and express CD31 (platelet endothelial cell adhesion molecule-1), CD36, intercellular adhesion molecule-1 (ICAM-1), and CD44 (1). They also bind purified T cells in a regulatable manner and respond to cytokines in a similar way to nontransformed endothelial cells. HMECs were passaged 5–7 days when confluent and were used for studies at population doublings between 25 and 40.
Human brain microvascular endothelial cells (HBMECs) were provided by Dr. K. S. Kim (Children's Hospital, Los Angeles, CA). HBMECs were cultured in RPMI 1640 supplemented with 10% FBS, 10% NuSerum (Becton Dickinson, Bedford, MA), endothelial cell growth supplement (30 μg/ml), heparin (5 U/ml), 1 mM sodium pyruvate, 1 mM MEM nonessential amino acids, 1 mM MEM vitamins, 5% l-glutamine, and 5% penicillin-streptomycin. HBMECs were grown and used at population doublings between 30 and 40. These cells were immortalized by SV40 large T antigen and have been shown to retain endothelial cell phenotypic and functional characteristics (25, 26). The cells were positive for von Willebrand factor, carbonic anhydrase IV, and Ulex Europeus agglutinin I, took up fluorescently labeled acetylated LDL, and expressed γ-glutamyl transpeptidase, demonstrating their brain endothelial cell characteristics. When treated with tumor necrosis factor-α, vascular cell adhesion molecule and ICAM-1 were expressed.
Human pulmonary artery endothelial cells (HPAECs) were purchased from Clonetics (San Diego, CA). These cells have been characterized to be endothelial in origin by the uptake of acetylated LDL and by positive staining for von Willebrand factor. The HPAECs were grown in basal medium containing EGM-2 Bulletkit growth supplement (Clonetics), 10% FBS, and 5% penicillin-streptomycin. HPAECs were cultured to 10–15 population doubling for use in studies.
The transformed renal embryonic kidney cell line, 293 cells (CRL 1573; American Type Culture Collection, Manassas, VA) (13), were maintained in DMEM supplemented with 10% FBS, 2 mMl-glutamine, 100 μg/ml penicillin, 100 U/ml streptomycin, and 10 mg/ml amphotericin B.
The endothelial expression of PKI mRNA was determined from HMECs, HBMECs, and HPAECs by RT-PCR. Total RNA was extracted using an RNA STAT-60 isolation kit (Tel-Test, Friendswood, TX) according to the manufacturer's protocol. RNA concentration was determined by spectrophotometry at an absorbance of 260 nm. RNA integrity was assessed by agarose gel electrophoresis and ethidium bromide staining. The RT-PCR was made using the GeneAmp RNA PCR kit (Perkin Elmer, Branchburg, NJ), and thermal cycling reactions were run in the GeneAmp 2400 PCR System (PE Biosystems, Foster City, CA). Isolated total RNA was subjected to reverse transcription with oligo(dT) primers, generating cDNA copies of the RNA sequence. The subsequent PCR was performed using specific primer sets based on sequences for PKI isoforms, human PKIα, human PKIγ, and mouse PKIβ (Table1). Reaction included a positive control, pAW109 RNA (provided by the kit) and negative control, which was in the absence of murine leukemia virus RT. RT-PCR products were analyzed by agarose gel electrophoresis. The products were subsequently purified, and both forward and reverse DNA strands were sequenced (Research Resources Center, DNA Core Facility, University of Illinois, Chicago, IL) and analyzed using Basic Local Alignment Search Tool (BLAST 2.0, National Center for Biotechnology Information).
RNase Protection Assay
RNase protection assay (RPA) was performed as previously described (10). Probes were allowed to hybridize to target RNA in reactions containing 10.0 μg of total cell RNA and 2.5 × 104 cpm of PKIα or PKIγ antisense probes (seeRiboprobe templates) in 30 μl of hybridization buffer [80% formamide, 0.4 M NaCl, 1 mM EDTA, and 40 mM PIPES (pH 6.4)]. Reactions were heated to 85°C for 5 min and then incubated at 48°C for 12–18 h. After hybridization, 280 μl of RNase digestion buffer [50 mM sodium acetate (pH 4.5) and 2 mM EDTA] were added with 30 units of T1 RNase for all assays, followed by incubation at 30°C for 60 min. Reaction was terminated and RNA precipitated by the addition of 70 μg of yeast transfer RNA and 700 μl of 7% Tri-Reagent (diluted in 100% ethanol). RNA was dissolved in loading buffer [80% formamide, 2 mM EDTA (pH 7.4) containing 0.05% bromophenol blue and 0.05% xylene cyanol], denatured at 85°C for 5 min, and resolved on a 5% acrylamide/8 M urea gel using 89 mM Tris (pH 8.0), 89 mM boric acid, and 2.7 mM EDTA (TBE buffer). Gels were used to prepare autoradiograms for analysis.
Purified and sequenced RT-PCR products for human PKIα and human PKIγ (see RT-PCR) were ligated into pGEM-T-Easy vector, chemically transformed into DH5α competent cells, and plated onto Luria-Bertani (LB)/ampicillin agar plates at 37°C. A successfully transformed colony (screen by restriction digests) was grown in LB overnight at 37°C, and the plasmid-DNA insert was isolated using the Wizard Plus Minipreps DNA purification system. The plasmid-DNA insert was linearized with PstI (PKIα) orBamHI (PKIγ) and transcribed with T7 RNA polymerase to produce radiolabeled antisense probes.
In vitro transcription of radiolabeled probes.
Probe synthesis was conducted with 1 mM each of CTP, ATP, and UTP, respectively, 9.38 μM of [32P]GTP (800 Ci/mmol), and 25 μM of unlabeled GTP. One microgram of template DNA with appropriate RNA polymerase (T7 RNA polymerase) was added to each reaction and was allowed to transcribe at 37°C for 40 min, and template was digested with DNase (Ambion) at 37°C for 15 min. The [32P]GTP not incorporated into the probe was removed using Centri-Spin 40 purification columns (Princeton Separation, Adelphia, NJ). Radiolabeled probes were used in RNase protection assays to detect mRNA levels of PKIα and PKIγ from total cell RNA isolated from HMECs, HBMECs, and HPAECs.
Protein Expression of PKI Isoforms
Cell collection and preparation.
HMECs, HBMECs, and HPAECs were grown to confluency in 60-mm culture dishes. The cells were placed on ice, washed two times with ice-cold Ca2+- and Mg2+-free PBS, and collected by scraping in RIPA buffer [in mM: 150 NaCl, 1 EDTA, 1 EGTA, 50 Tris · HCl (pH 7.4), 1% Nonidet P-40, 1 NaF, and 1 sodium vanadate, as well as 0.25% sodium deoxycholate, 1 pepstatin A, 1 phenylmethylsulfonyl fluoride, 25 μg/ml leupeptin, and 25 μg/ml aprotinin]. The cell extract was sonicated with 10 pulses using the sonifier (Branson Ultrasonics, Danbury, CT) and was heated to 92°C for 10 min. After being cooled to 4°C, the endothelial cell lysate was centrifuged at 14,000 rpm for 15 min, and the supernatant was collected for protein concentration, which was determined using the bichinchoninic acid protein assay kit with BSA as standard (Pierce, Rockford, IL).
Affinity-purified polyclonal antibody to full-length murine PKIγ was prepared as described previously (7). In brief, purified histidine-tagged PKIγ protein was conjugated to keyhole limpet hemocyanin and used for immunization of rabbits for antibody production. The antiserum raised against PKIγ was affinity purified on nitrocellulose blots bound with the purified fusion protein. In addition, polyclonal anti-peptide antibodies were prepared against peptides for rat PKIα-(5–22) (TTYADFIASGRTGRRNAI) and rat PKIβ-(5–22) (SVISSFASSARAGRRNAL) as previously described (12, 15).
Western blot analysis.
The heat-treated cell lysates were loaded at constant protein concentrations, and the proteins were separated by SDS-polyacrylamide gradient gel electrophoresis. The separated proteins were electrotransferred to nitrocellulose or polyvinylidene difluoride membranes. Nonspecific binding of antibody to membrane was blocked with 5% nonfat dry milk in Tris-buffered saline with 0.05% Tween 20 (TBST). The blocked membrane was then incubated with primary antibodies, diluted in TBST with 1% nonfat dry milk, overnight at 4°C in a rocker. The blot was washed five times with TBST and incubated with anti-rabbit IgG secondary antibody conjugated with horseradish peroxidase, and the bands were detected using the enhanced chemiluminescence kit.
Transendothelial Electrical Resistance
The transendothelial electrical resistance, an index of endothelial barrier function, was determined in real time using the electric cell-substrate impedance sensor (ECIS) system (Applied BioPhysics, Troy, NY) (11, 19). The system consists of one large gold-plated electrode (10−1cm2), eight smaller gold-plated electrodes (10−4 cm2), and a 500-μl well fitted above each small electrode. The small and large electrodes were connected to a phase-sensitive lock-in amplifier, and an alternating current was supplied through the 1 MΩ resistor. The measured electrical impedance or calculated resistance indicates the restriction of current flow through the cell monolayer and thus provides an index of the endothelial barrier function.
For resistance measurement, endothelial cells (105 cells) were plated onto a sterile, fibronectin-coated gold-plated electrode and grown to confluency. The electrode was mounted onto the ECIS system housed within an incubator (maintained at 37°C, 5% CO2, and 100% humidity) and connected to the lock-in amplifier. After a period of 15 min of equilibration, the cells were challenged with reagents according to experimental protocol and resistances were recorded continuously in real time.
Construction of Rabbit Muscle PKI in Endothelial Cells
An E1−, E3− replication-deficient adenovirus containing PKI (AdPKI) was prepared as described previously (16). The 251-bp DNA fragment encoding the complete amino acid sequence of rabbit muscle PKI (8) was subcloned into the shuttle vector pACCMV.pLpA, creating pACCMV-PKI. Equimolar amounts of pACCMV-PKI (0.2 μg) were cotransfected with the plasmid pJM17 (0.8 μg), which contains the full-length adenovirus (Ad 5) genome sequences (with a mutant E3 region) as well as ampicillin and tetracycline resistance sequences and a bacterial origin of replication (17), into 293 cells, a transformed renal embryonic kidney cell line (13). Homologous recombination between the two plasmids resulted in an E1−, E3− adenovirus genome that can replicate and packaged into virions only in 293 cells in which E1 function is supplied in trans by integrated, constitutively expressed adenovirus E1 sequences. The vector AdPKI was amplified in 293 cells and its genome confirmed by PCR amplication of contiguous adenovirus/expression cassette sequences. The vector was purified by double cesium chloride ultracentrifugation and exhaustive dialysis against virus suspension buffer (10 mM Tris, 10 mM MgCl2, and 10% glycerol) and was titered and stored at −80°C. For control, the vector AdNull, expressing no transgene, was constructed in a similar manner but without subcloned gene sequences between the cytomegalovirus promoter and the polyadenylation signal.
The following were purchased from GIBCO BRL (Gaithersburg, MD): DH5α competent cells, MCDB 131 medium, DMEM, penicillin-streptomycin,l-glutamine, sodium pyruvate, MEM nonessential amino acids, MEM vitamins; from Sigma Chemical (St. Louis, MO): EGF, hydrocortisone, endothelial cell growth supplement, heparin, T1 RNase, ΦX174 RF DNA/HaeIII fragments; from Promega (Madison, WI): pGEM-T-Easy vector, Wizard Plus Minipreps DNA Purification System, T7 RNA polymerase. All other reagents were obtained as indicated in the text.
Overexpression of PKI Decreases Endothelial Barrier Restrictiveness
The role of PKI in the regulation of endogenous PKA in endothelial barrier function was investigated by overexpression of endothelial cells with AdPKI. HMECs were infected at 100 multiplicity of infection (MOI) of AdPKI for 2 days for the study. We have demonstrated that the use of this protocol for adenovirus infection of HMECs was highly efficient (>95%) for gene transfer (16). The cells were treated with 10 μM forskolin (direct activator of adenylyl cyclase) plus 1 μM IBMX (phosphodiesterase inhibitor) to increase intracellular cAMP levels, and the transendothelial electrical resistance was measured in real time. In control noninfected HMECs, the combination of forskolin plus IBMX significantly increased resistance above baseline (Fig. 1), indicating increased restrictiveness of the endothelial monolayer. However, this increased resistance was abolished in the PKI-overexpressing HMECs (Fig. 1).
We also determined the effects of PKI on mediator-induced changes in endothelial barrier function. HMEC were infected at 100 MOI with AdPKI or control AdNull for 2 days, and the transendothelial electrical resistance, an index of barrier function, was determined in response to 1.0 nM thrombin. Stimulation of HMECs with thrombin caused transient decreases in resistance in which the AdPKI-infected HMECs showed greater maximal decreases in resistance relative to the control AdNull-infected HMECs (Fig.2 A). Figure 2 Bsummarizes the maximal thrombin-mediated resistance decrease from baseline in the controls (noninfected wild-type HMECs and HMECs infected with AdNull) and HMECs infected with AdPKI. The results indicated that, in HMECs infected with AdPKI, the thrombin-mediated resistance decrease was significantly greater (−6,526 ± 494 Ω) than in either control wild-type HMECs or AdNull groups (−4,137 ± 123 and −3,837 ± 163 Ω, respectively).
PKIα, PKIβ, and PKIγ mRNA Expression in Endothelial Cells
For RT-PCR analysis, in HBMECs (Fig.3 A), HPAECs (Fig.3 B), and HMECs (Fig. 3 C), specific primers for PKIα and PKIγ yielded PCR products comparable to the predicted size (542 and 240 bp, respectively; Table 1). In HMECs, primers for PKIβ produced a product greater than the predicted 569 bp (Fig.3 C). The positive control (pAW109 RNA) generated the predicted 302-bp band. All negative controls (absence of RT) showed lack of RT-PCR products, indicating lack of contaminating genomic DNA. The RT-PCR products were purified, and the forward and reverse DNA strands were sequenced. The sequences were subjected to BLAST 2.0 analysis and results are summarized in Table2. The DNA sequences for PKIγ from HBMECs, HPAECs, and HMECs were of >96% identity with the human PKIγ gene database sequences, and showed an expectation value (E) of < 10−101, confirming the expression of PKIγ in these three endothelial cell types. Similarly, the DNA sequences for PKIα from HBMECs were of 100% identity with human PKIα gene database sequences, and the E value was 0.0. However, insufficient purified RT-PCR product was obtained from HPAECs and HMECs for sequencing analysis of PKIα. The sequencing results obtained from PCR products using primers for PKIβ indicated that the DNA sequences corresponded to three genes (human prolyl 4-hydroxylase β-subunit, human thyroid hormone binding protein, and human glutathione transhydrogenase).
We next used RNase protection assay to confirm the expression of PKIα and PKIγ in endothelial cells. Riboprobe templates were generated from purified RT-PCR products for PKIα and PKIγ (seematerials and methods). Total RNA isolated from HBMECs, HPAECs, and HMECs was reacted with the antisense probes. Results indicated that the PKIα antisense probe protected the 542 nucleotides of PKIα mRNA from HPAECs and HBMECs (Fig.4, top). The PKIγ antisense probe protected the 240 nt of PKIγ mRNA from HBMECs, HPAECs, and HMECs (Fig. 4, bottom).
PKI protein expression was investigated using Western blot analyses. Anti-peptide polyclonal antibodies directed against PKIα and PKIβ (15) and affinity-purified polyclonal anti-PKIγ antibody (7) were used for Western blot analysis of HBMEC, HPAEC, and HMEC cell lysates. A strong band at ∼20 kDa was observed for PKIγ from the three endothelial cell types (Fig.5 A). PKIα was detected as a band of ∼12 kDa in HBMECs only; PKIβ was not detected in any of the three cell types (Fig. 5 B). Positive control using testis tissue lysate (15) is shown for PKIα and PKIβ (Fig.5 B).
Expression of the exogenous rabbit muscle PKI gene into endothelial cells inhibited the increase in basal endothelial resistance induced by the cAMP-elevating agents, forskolin and IBMX, providing strong evidence that PKA was responsible for mediating the increase in barrier function. This observation was consistent with our earlier finding that PKI overexpression abolished the protective effects of cAMP-elevating agents (i.e., forskolin, IBMX, and cholera toxin) against a thrombin-induced increase in endothelial permeability (16, 22). Interestingly, PKI overexpression also exacerbated the decrease in thrombin-induced transendothelial resistance. This result suggests that PKA normally functioned to maintain the extent of basal endothelial barrier restrictiveness; thus PKI may be a physiological modulator of PKA activity in the endothelial barrier response. Inhibition of PKA also promotes actin stress fiber formation and loss of catenin (22), supporting the notion that endogenous PKA may be important in regulation of basal cell barrier restrictiveness.
The results from the current study indicate that endothelial cells derived from human brain, pulmonary artery, and foreskin expressed predominantly PKIα and PKIγ. PKIβ mRNA and protein were not detected in any of the endothelial cells. PKIγ mRNA was expressed by all three endothelial cell types as determined by both RT-PCR and RNase protection assays. This message expression corresponded with PKIγ protein expression from the three endothelial cell types. Further, PKIα mRNA expression was observed in the three endothelial cell types when determined by RT-PCR; however, RNase protection assay detected PKIα mRNA for only HPAECs and HBMECs, but not HMECs. This finding may possibly be attributed to relatively low levels of the PKIα message expressed by HMECs, which was detectable with amplification by RT-PCR. PKIα protein expression was detected in HBMECs, but not from HMECs or HPAECs, suggesting low levels or lack of the protein expression by these cells. PKIα has been reported to be predominantly expressed in the cerebral cortex and muscle, whereas PKIβ is more limited to the testis (28). Overall, the results are consistent with observations made by Collins and Uhler (7) who reported that PKIγ was abundant and more widely expressed than either PKIα and PKIβ in tissues, such as heart, skeletal muscle, testis, spleen, lung, liver, and kidney. The more prevalent expression of PKIγ in endothelial cells suggests that this isoform may be important in modulating a broad range of basic cellular activities, including the regulation of basal and mediator-induced changes in endothelial barrier function.
A comparison of the sequences between human PKIα and PKIγ coding regions indicated ∼53% homology, whereas the rabbit muscle PKI used for the functional studies showed 74% and 53% homology with human PKIα and PKIγ, respectively. The amino acid sequences in PKI important for binding to and inhibition of the PKA catalytic subunit have been shown to be relatively conserved among PKI isoforms (7). PKI inhibits PKA through interactions within the substrate binding site of the PKA catalytic subunit, functioning as a competitive inhibitor. Although the overexpression of rabbit muscle PKI into endothelial cells likely inhibited PKA in a similar way to PKIα and PKIγ, the relative potency and efficacy may be different. PKIα is known to have greater binding affinity to the PKA catalytic subunit than PKIγ (inhibition constant K i = 0.073 and 0.44 nm, respectively) (7, 9). The sixfold difference in affinity may be physiologically significant, since the basal level of the catalytic subunit is normally maintained quite low in cells.
It is now known that, in addition to its inhibition potential of PKA, PKI functions to shuttle the PKA catalytic subunit out of the nucleus via a nuclear export signal (NES) (29, 31). All members of the PKI family possess an amino-terminal inhibitory region and a central region containing the NES (7). In general, the least conserved region occurs in the carboxy terminus of the molecule (7). These differences among the isoforms and their differential expressions suggest possible isoform-specific physiological function(s). It remains to be determined whether PKIα and PKIγ have differential functions in endothelial cells.
PKIβ was not detected in the endothelial cells by RT-PCR. One reason may be the use of primer-based mouse DNA sequences, which may not be highly conserved with human PKIβ. However, the human PKIβ gene has not been cloned. Nonetheless, the result was not surprising, since PKIβ expression is mostly limited to testis (28).
In summary, the main findings are that 1) overexpression of rabbit muscle PKI decreased endothelial barrier restrictiveness, and2) endothelial cells derived from human brain microvessel, pulmonary artery, and foreskin microvessel expressed predominantly PKIα and PKIγ isoforms. We conclude that PKIα and PKIγ may be important in modulating PKA in the regulation of vascular endothelial barrier function.
This work was supported by National Heart, Lung, and Blood Institute Grant HL-62649 (H. Lum), the American Heart Association, National (H. Lum), and a Veterans Affairs Grant (C. E. Patterson).
Address for reprint requests and other correspondence: H. Lum, Dept. of Pharmacology, Rush Presbyterian St. Luke's Medical Center, 2242 W. Harrison St., Suite 260, Chicago, IL 60612 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2002 the American Physiological Society