KLF2 provokes a gene expression pattern that establishes functional quiescent differentiation of the endothelium.

The flow-responsive transcription factor KLF2 is acquiring a leading role in the regulation of endothelial cell gene expression. A genome-wide microarray expression profiling is described employing lentivirus-mediated, 7-day overexpression of human KLF2 at levels observed under prolonged flow. KLF2 is not involved in lineage typing, as 42 endothelial-specific markers were unaffected. Rather, KLF2 generates a gene transcription profile (> 1000 genes) affecting key functional pathways such as cell migration, vasomotor function, inflammation, and hemostasis and induces a morphology change typical for shear exposure including stress fiber formation. Protein levels for thrombomodulin, endothelial nitric oxide synthase, and plasminogen activator inhibitor type-1 are altered to atheroprotective levels, even in the presence of the inflammatory cytokine TNF-alpha. KLF2 attenuates cell migration by affecting multiple genes including VEGFR2 and the potent antimigratory SEMA3F. The distribution of Weibel-Palade bodies in cultured cell populations is normalized at the single-cell level without interfering with their regulated, RalA-dependent release. In contrast, thrombin-induced release of Weibel-Palade bodies is significantly attenuated, consistent with the proposed role of VWF release at low-shear stress regions of the vasculature in atherosclerosis. These results establish that KLF2 acts as a central transcriptional switch point between the quiescent and activated states of the adult endothelial cell.

Endothelial KLF2 links local arterial shear stress levels to the expression of vascular tone-regulating genes.

Lung Krüppel-like factor (LKLF/KLF2) is an endothelial transcription factor that is crucially involved in murine vasculogenesis and is specifically regulated by flow in vitro. We now show a relation to local flow variations in the adult human vasculature: decreased LKLF expression was noted at the aorta bifurcations to the iliac and carotid arteries, coinciding with neointima formation. The direct involvement of shear stress in the in vivo expression of LKLF was determined independently by in situ hybridization and laser microbeam microdissection/reverse transcriptase-polymerase chain reaction in a murine carotid artery collar model, in which a 4- to 30-fold induction of LKLF occurred at the high-shear sites. Dissection of the biomechanics of LKLF regulation in vitro demonstrated that steady flow and pulsatile flow induced basal LKLF expression 15- and 36-fold at shear stresses greater than approximately 5 dyne/cm2, whereas cyclic stretch had no effect. Prolonged LKLF induction in the absence of flow changed the expression of angiotensin-converting enzyme, endothelin-1, adrenomedullin, and endothelial nitric oxide synthase to levels similar to those observed under prolonged flow. LKLF repression by siRNA suppressed the flow response of endothelin-1, adrenomedullin, and endothelial nitric oxide synthase (P < 0.05). Thus, we demonstrate that endothelial LKLF is regulated by flow in vivo and is a transcriptional regulator of several endothelial genes that control vascular tone in response to flow.

Aminopeptidase inhibitor bestatin stimulates microvascular endothelial cell invasion in a fibrin matrix.

The aminopeptidase inhibitor bestatin has been shown to have anti-angiogenic effects in a number of model systems. These effects are thought to result from inhibition of CD13 activity. Because tumor angiogenesis can evolve in a fibrin-rich stroma matrix we have studied for the first time the effects of bestatin on microvascular endothelial capillary-like tube formation in a fibrin matrix. Bestatin enhanced the formation of capillary-like tubes dose-dependently. Its effects were apparent at 8 micro M; the increase was 3.7-fold at 125 micro M; while high concentrations (>250 micro M), that were shown to have anti-angiogenic effects in other systems, caused extensive matrix degradation. Specific CD13-blocking antibodies WM15 and MY-7, and the aminopeptidase inhibitors amastatin and actinonin also enhanced capillary-like tube formation (maximally 1.5-fold), but these effects did not reach statistical significance. The effect of bestatin was not due to a change in uPAR availability because the relative involvement of the u-PA/u-PAR activity was not altered by bestatin. In view of the present findings we hypothesize that aminopeptidases other than CD13 predominantly contribute to the observed pro-angiogenic effect of bestatin in a fibrin matrix. The identification of this novel effect of bestatin is important in the light of the proposed use of bestatin as anti-angiogenic and/or anti-tumor agent.

Design, synthesis, and biological evaluation of a dual tumor-specific motive containing integrin-targeted plasmin-cleavable doxorubicin prodrug.

The design, synthesis, and initial biological evaluation of a doxorubicin prodrug that contains a dual tumor specific moiety, which allows enhanced tumor recognition potential, is reported. Both a tumor-specific recognition site and a tumor selective enzymatic activation sequence are incorporated in the prodrug. The first tumor-specific sequence is the bicyclic CDCRGDCFC (RGD-4C) peptide that selectively binds alpha v beta 3 and alpha v beta 5 integrins. These integrins are highly overexpressed on invading tumor endothelial cells. The second tumor-specific sequence is a D-Ala-Phe-Lys tripeptide that is selectively recognized by the tumor-associated protease plasmin, which is involved in tumor invasion and metastasis. An aminocaproyl residue was incorporated as a spacer between the two peptide sequences, whereas a self-eliminating 4-aminobenzyl alcohol spacer was inserted between the plasmin substrate and doxorubicin. Although the prodrug showed a decreased binding affinity as compared with the unconjugated reference peptide, it was still a potent ligand for alpha v beta 3 and alpha v beta 5 integrin receptors. The synthesized construct also possessed plasmin substrate properties as demonstrated by doxorubicin release from 1 upon incubation with plasmin. The release of doxorubicin from 1 was not complete, possibly related to low prodrug solubility. In vitro prodrug 1 showed plasmin-dependent cytotoxicity for endothelial cells and HT1080 fibrosarcoma cells. On the basis of these in vitro results, derivatives of 1 with improved water solubility are considered good candidates for additional development and in vivo evaluation of this dual targeting concept.

A doxorubicin-CNGRC-peptide conjugate with prodrug properties.

There is increasing interest in the exploitation of molecular addresses for the targeting of tumor imaging or therapeutic agents. A recent study demonstrated anticancer activity in human xenografts of doxorubicin (DOX)-peptide conjugates targeted to the tumor vascular endothelium, among them DOX coupled to the cyclic pentapeptide CNGRC [Science 279 (1998) 377]. In order to learn more about the mechanism of action of this type of DOX-peptide conjugates, we have studied the interaction of DOX-CNGRC with primary human umbilical cord vein endothelial cells (HUVEC) and tumor cells under defined in vitro conditions. We used a DOX conjugate, in which the cyclic CNGRC peptide, for which an in vivo endothelial address has recently been identified as aminopeptidase N (APN)/CD13, has been coupled via a hydrolysable spacer to the C-14 anthracycline-side chain. First we determined that the t(1/2) of DOX-CNGRC conjugate in human blood was 442 min (at 37 degrees ) allowing sufficient time for endothelial targeting when administered i.v. When cultured cells were exposed for 30 min to DOX-CNGRC a more cytoplasmic localization of fluorescent drug was seen when compared to DOX exposure and intracellular DOX-CNGRC was identified after extraction from the cells. This revealed differences in the cellular uptake process of the conjugate compared to DOX. The antiproliferative effect of DOX-CNGRC was determined by 30 min exposure in medium with a high protein content in order to mimick the in vivo targeting situation. In this medium, the IC(50) was 1.1 microM for highly CD13 expressing HT-1080, 1.45 microM for CD13 negative SK-UT-1 sarcoma cells and 6.5 microM for CD13 positive HUVEC. The IC(50) of DOX for these cells were 1.0, 2.0 and 7.3 microM, respectively. Although DOX-CNGRC inhibited the peptidase activity of CD13 up to 50%, our data do not favor an important role for the enzyme inhibition in the cytotoxic effect of the conjugate. The antitumor activity was tested in nude mice bearing human ovarian cancer xenografts (OVCAR-3). A weekly i.v. administration (3mg/kg DOX-equivalent, 3x) showed a minor (40%) growth delay, which does not indicate efficacy better than that expected for free DOX. In conclusion, this study indicates that the antiproliferative and anti-angiogenic effects of DOX-CNGRC as reported before, are likely caused by the cytostatic effects of intracellularly released parent drug DOX, independent of CD13 expression/activity. More research is needed to identify the optimal specific chemical configuration of DOX-peptide conjugates for in vivo targeting and receptor-mediated cellular uptake.