Long-Circulating Immunoliposome Targeting in Animal Models

Abstract

The current status of newly developed PEG-immunoliposomes (Type C), carrying monoclonal antibodies or their fragments (Fab') at the distal ends of the PEG chains, were described in this review. In terms of target binding of Type C, two different anatomical compartments were considered. They are the mouse lung endothelial for a readily accessible site in intravascular and the mouse implanted solid tumor for a much less accessible target site located in extravascular through leaky vascular. Distearoyl phosphatidylethanolamine derivatives of PEG with carboxyl group (DSPE-PEG-COOH) and dipalmitoyl phosphatidylethanolamine derivatives of PEG with maleimidyl group (DPPE-PEG-Mal) at the PEG's terminus were newly synthesized. Small unilamellar liposomes (90 - 130 nm in diameter) were prepared from phosphatidylcholine and cholesterol (2:1, m/m) containing 6 mol% of DSPE-PEG-COOH or DPPE-PEG-Mal. For targeting to the vascular endothelial surface in the lung, 34A antibody, which is highly specific to mouse pulmonary endothelial cells, was conjugated to PEG-liposome (34A-Type C). The degree of lung binding of 34A-Type C in BALB/c mouse was significantly higher than that of 34A-Type A which is an ordinary type of immunoliposome (without PEG derivatives). For targeting to the solid tumor tissue, 21B2 antibody which is anti-human CEA and its Fab' fragment were used. The targeting ability of Fab'-Type C was examined by using CEA-positive human gastric cancer strain MKN-45 cells inoculated into BALB/c nu/nu mice. Fab'-Type C showed the low RES uptake and the long circulation time, and resulted in enhanced accumulation of the liposomes in the solid tumor. The small Fab'-Type C could predominantly pass through the leaky tumor endothelium by passive convective transport. These studies offer some important insights into the potential of target-specific drug delivery.     

Modulation of the IgE immune response to BSA by fragments of the antigen. I. Suppression by free fragments and by fragments conjugated to homologous gamma-globulin

Nonimmunogenic peptic fragments of bovine serum albumin (BSA), Fraction Ia, suppressed immune response to BSA in mice. Splenic T lymphocytes from mice treated with these fragments suppressed the anti-DNP response in irradiated mice reconstituted with DNP-BSA-primed cells, indicating carrier-specific suppression. The conjugate of Fraction Ia with mouse γ-globulin (MGG) was found to be an effective suppressive substance but it did not induce suppressor T cells. B cells from mice given Ia-MGG were unresponsive to BSA when transferred to irradiated recipients along with either normal or BSA-primed T cells. Thus, unresponsiveness to BSA was mediated by either T or B lymphocytes, depending whether the inducing substance was a free fragment of the antigen or fragments conjugated to homologous γ-globulin.     

Evaluation of cell disruption effects on primary recovery of antibody fragments using microscale bioprocessing techniques

Intracellular antibody Fab' fragments periplasmically expressed in Escherichia coli require the release of Fab' from the cells before initial product recovery. This work demonstrates the utility of microscale bioprocessing techniques to evaluate the influence of different cell disruption operations on subsequent solid–liquid separation and product recovery. Initially, the industrial method of Fab' release by thermochemical extraction was established experimentally at the microwell scale and was observed to yield Fab' release consistent with the larger scale process. The influence of two further cell disruption operations, homogenization and sonication, on subsequent Fab' recovery by microfiltration was also examined. The results showed that the heat‐extracted cells give better dead‐end microfiltration performance in terms of permeate flux and specific cake resistance. In contrast, the cell suspensions prepared by homogenization and sonication showed more efficient product release but with lower product purity and poorer microfiltration performance. Having established the various microscale methods the linked sequence was automated on the deck of a laboratory robotic platform and used to show how different conditions during thermochemical extraction impacted on the optimal performance of the linked unit operations. The results illustrate the power of microscale techniques to evaluate crucial unit operation interactions in a bioprocess sequence using only microliter volumes of feed. © 2010 American Institute of Chemical Engineers Biotechnol. Prog., 2010     

Characterization of antibodies to chicken riboflavin carrier protein: Antigenicity of the tryptic fragments

The antigenecity of tryptic fragments of reduced and carboxymethylated chicken riboflavin carrier protein were studied. The tryptic sites of the native riboflavin carrier protein bound to riboflavin were inaccessible. The molecular weight and the elution profile on high performance liquid chromatography (TSK 545 DEAE) were unaltered at an enzyme to substrate ratio of 1:31. However, carboxymethylated riboflavin carrier protein could be cleaved into 3 or 4 fragments at an enzyme to substrate ratio of 1:250 or 1:125. Chromatographic separation of the tryptic fragments on high pressure liquid chromatography (TSK 545 DEAE) revealed the presence of two fragments with different elution profiles but similar molecular weight 26 ±2 kDa. Only one fragment (associated with peak 2) had the ability to displace chicken riboflavin carrier protein in an homologous chicken riboflavin carrier protein radioimmunoassay. Thus, carboxymethylated ribotlavin carrier protein which does not compete with chicken riboflavin carrier protein in the radioimmunoassay, on mild trypsinization generates a fragment which interacts with chicken riboflavin carrier protein in radioimmunoassay.     

Understanding the role of internal lysine residues of serum albumins in conformational stability and bilirubin binding

The role of internal lysine residues of different serum albumins, viz. from human, rabbit, goat, sheep and buffalo (HSA, RbSA, GSA, SSA and BuSA), in conformational stability and bilirubin binding was investigated after blocking them using acetylation, succinylation and guanidination reactions. No significant change in the secondary structure was noticed whereas the tertiary structure of these proteins was slightly altered upon acetylation or succinylation as revealed by circular dichroism (CD), fluorescence and gel filtration results. Guanidination did not affect the native protein conformation to a measurable extent. Scatchard analysis, CD and absorption spectroscopic results showed marked reductions (5-21-fold decrease in K(a) and approximately 50% decrease in the CD Cotton effect intensity) in the affinity of albumins for bilirubin upon acetylation or succinylation whereas guanidination produced a small change. Interestingly, monosignate CD spectra of bilirubin complexed with GSA, SSA and BuSA were transformed to bisignate CD spectra upon acetylation or succinylation of internal lysine residues whereas spectra remained bisignate in the case of bilirubin bound to acetylated or succinylated derivatives of HSA and RbSA. When probed by CD spectroscopy, bilirubin bound to acetylated or succinylated derivatives of GSA and SSA rapidly switched over to native albumins and not vice versa. These results suggested that salt linkage(s) contributed by internal lysine residue(s) play an important role in the high-affinity binding of bilirubin to albumin and provide stability to the native three-dimensional conformation of the bound pigment. Chloroform severely decreased the intensity of both positive and negative CD Cotton effects of bilirubin complexed with acetylated or succinylated derivatives of all albumins which otherwise increased significantly in the case of bilirubin complexed with native and guanidinated albumin derivatives, except the bilirubin-RbSA complex which showed a small decrease in intensity. These results suggest that the presence of salt linkage(s) in bilirubin-albumin complexation is(are) crucial to bring about effective and efficient stereochemical changes in the bound pigment by co-binding of chloroform which seems to have at least one conserved binding site on these albumins that is shared with bilirubin.     

Preparation of monospecific antibodies against isoform 2 of translation elongation factor 1A (eEF1A2)

Amino acid sequences of eukaryotic translation elongation factor isoform 1 (eEF1A1) and 2 (eEF1A2) were compared and two peptide fragments of eEF1A2 were chosen as linear antigenic determinants for generation of monospecific antipeptide antibodies. Synthesized peptides corresponded to the selected peptide fragments were conjugated to bovine serum albumin (BSA) and used for immunizations of mice. Antibodies, produced against the eEF1A2 fragment 330–343 conjugated to BSA, specifically recognized this isoform in the native and partially denatured states but did not interact with the eEF1A1 isoform. It was shown that these monospecific anti-eEF1A2 antibodies could be employed for eEF1A2 detection both by enzyme-linked immunosorbent assay and by immunoblotting.     

Electrophoretic and immunologic characterization of proteins of merozoites of Eimeria acervulina, E. maxima, E. necatrix, and E. tenella.

Merozoites of Eimeria acervulina, Eimeria maxima, Eimeria necatrix, and Eimeria tenella were compared by gel electrophoresis, western-blotting with chicken antiserum, indirect fluorescent antibody reactions, and antiserum neutralization. Merozoites from the 4 species had dissimilar patterns of proteins and antigens in soluble and membrane fractions. Coomassie blue staining of SDS-PAGE gels revealed 16-22 protein bands depending on the species of merozoite but only 3 bands per species in the membrane fractions. Homologous and heterologous antisera recognized 5-12 soluble fraction bands and 3-7 membrane fraction bands on immunoperoxidase-stained western blots, depending on the species. When antisera from infected chickens were used in an indirect fluorescent antibody reaction, the merozoites of E. tenella and E. necatrix had a strong reaction with homologous and heterologous antisera. Merozoites of E. acervulina and E. maxima reacted with homologous antisera but had a weak or no reaction with heterologous antisera. Chicken antiserum against E. tenella had no effect on the viability of E. tenella merozoites when they were inoculated into chicken embryos.     

Immune response of rabbits to native G-actins

Actins are highly conserved proteins and are therefore claimed to be not very immunogenic without prior denaturation or chemical modification. We have obtained in rabbits high-titered antibodies to "native" G-actins from chicken and man, and assayed their cross-reaction using an enzyme immunoassay, Western blotting and immunohistochemistry. The antigens differ in their ability to induce antibody formation (chicken gizzard actin [(beta), gamma] greater than chicken skeletal actin [alpha] = human platelet actin [beta, (gamma)]). Antibodies to skeletal actin [alpha] are muscle-specific and mainly directed against the homologous region comprising the N-terminus (residues 1-226). Antibodies to gizzard actin [(beta), gamma] cross-react, to a lesser extent, with the alpha and beta, (gamma) isoforms. They show no regional specificity within the homologous antigen. Antibodies to the tryptic core fragment (residues 69-374) of skeletal actin react with fragments comprising the C-terminal part of muscular actins. Antibodies to platelet actin [beta, (gamma)] cross-react with muscular actins, recognizing not the native, but slightly degraded molecules. Platelet actin induces the formation of high-titered albumin antibodies for hitherto unknown reasons.     

Inhibition of immune precipitation by complement

Normal human complement serum (NHS) inhibited precipitin reactions between tetanus toxoid and human or rabbit anti-tetanus toxoid IgG antibody, between bovine serum albumin (BSA) and rabbit anti-BSA IgG antibody, and between hen egg albumin and rabbit anti-egg albumin IgG antibody. Ethylene-diaminetetraacetic acid (EDTA) prevented this inhibition. Mg-ethyleneglycol-bis(aminoethyl)-tetra-acetic acid-(EGTA) also prevented the inhibition except with lower concentrations of antibody and antigen. Therefore, the inhibition of immune precipitation seemed to occur mainly through the classical pathway of complement activation. The alternative pathway was usually dispensable, but it augmented the inhibition. Guinea pig complement serum (NGS) was less effective than NHS in inhibiting immune precipitation. Guinea pig serum deficient in C4 (C4DGS) did not inhibit the immune precipitation. Mouse complement serum was effective for inhibiting precipitation, and C5-deficient serum was as effective as normal serum. Therefore, the inhibition of immune precipitation is considered to occur by activation of complement up to the step of C3. The size of the soluble immune complexes formed in the presence of NHS varied depending on the concentrations of antibody and antigen, even when the ratio of antigen to antibody was constant. On incubation at 37 degrees C immune precipitation was inhibited by 1/2 dilution of NHS for 2 to 3 hr and then gradually increased to the level in the absence of complement. When the immune complexes were formed in the presence of serum containing complement, fragments of C4 and C3 were incorporated into the soluble immune complexes. The C3 fragments incorporated into the soluble complexes were C3b, iC3b, C3c, and C3d, some of which were bound covalently with heavy chains of IgG antibody molecules. Some of the covalent linkages between C3 fragments and IgG seemed to be destroyed by alkali treatment, but not by hydroxylamine treatment. The formation of covalent bonds between IgG and C3 and probably C4 was essential for inhibition of immune precipitation, because inhibitors of their formation, such as putrescine, cadaverine, and salicylhydroxamic acid, effectively prevented the inhibition of precipitation. When antigen and antibody reacted in the presence of mixtures of various combinations of isolated complement components, C1, C4, C2, and C3 showed maximal inhibition of immune precipitation, whereas factors I and H had little effect.     

Differentiation Between Herpes Simplex Virus Type 1 and Type 2 Strains by Immunoelectroosmophoresis

A method has been elaborated to differentiate between herpes simplex type 1 and type 2 viruses by immunoelectroosmophoresis. With rabbit immune sera cross-absorbed with heterologous virus antigen, a distinct difference was shown between the two virus types. Herpes simplex type 1 virus tested against cross-absorbed type 1 antiserum gave two precipitin lines. Herpes simplex type 2 virus gave one precipitin line when tested against cross-absorbed homologous serum. When the viral antigens were tested against cross-absorbed heterologous immune sera, no or only very weak precipitin reactions were observed. The test is easy and rapid, requires relatively small quantities of antigen and antibody, and is suitable for typing of herpes simplex virus in diagnostic routine work.     

Adjuvant Action of Capsular Polysaccharide of Klebsiella pneumoniae on Antibody Response

Study was made to clarify the experimental conditions for the capsular polysaccharide of Klebsiella pneumoniae (CPS-K) to exhibit maximum adjuvant effect on antibody production to bovine serum albumin (BSA) in mice. To obtain the maximum primary antibody response and also the strongest priming for a secondary response to BSA, 1000 μg of CPS-K had to be injected intramuscularly into the same or adjacent site of BSA injection within the period of 1 hr before to 6 hr after the BSA injection. The optimum amount of BSA giving the maximum antibody response and also the strongest priming under these experimental conditions was 15 mg. In mice thus primed, an extremely high secondary response was induced by injecting 0.5 mg of BSA 30 days after the initial injection. The minimum amount of CPS-K, to exhibit a strong adjuvant action, was 100 μg, which was equal to the minimum amount to induce immunologic paralysis to a homologous antigen. Extremely large amounts, such as 100 to 300 mg per mouse of BSA, were also strongly immunogenic when injected together with paralyzing doses of CPS-K. In vitro admixture of BSA and CPS-K before injection did not strengthen adjuvant action of CPS-K. Alum-precipitated BSA mixed with CPS-K was not more immunogenic than native BSA mixed with CPS-K. Addition of Freund's complete adjuvant to an injection of BSA and CPS-K mixture did not enhance the adjuvant effect of CPS-K.     

Unwanted Interactions of Maleimidophenylbutyrate-Phosphatidylethanolamine Containing (Immuno) Liposomes with Cells In Vitro

Abstract

The interaction between (immuno)liposomes and different (target and nontarget) cells was investigated in vitro. Maleimidophenylbutyrate-phosphatidylethanolamine (MPB-PE)-containing reverse-phase evaporation vesicles (REV-MPB-PE) were used; Fab' (polyclonal) fragments against mouse red blood cells (RBC) were selected as the homing device. Unwanted, nonspecific interactions were observed. these could be overcome by blocking the free reactive maleimide group of MPB-PE after Fab' coupling with dithiothreitol (DTT), or by storing the immunoliposomes for a period of 1 week before use. the specific interaction between immunoliposomes and target cells was maintained during storage. Storage of MPB-PE liposomes before Fab' coupling to REV-MBP-PE, however, reduced the coupling capacity considerably.     

LPS-based conjugate vaccines composed of O-polysaccharide from Pseudomonas aeruginosa IATS 6 and 11 bound to a carrier protein

Lipopolysaccharide (LPS) is the major surface antigen of Pseudomonas aeruginosa and elicits protective antibodies in animals. No cross reaction was observed between LPSs of P. aeruginosa International Antigenic Typing Scheme (IATS) 6 and 11 strains using rabbit polyclonal antibodies raised against the whole cells. The O-polysaccharides (O-PSs) from IATS 6 and 11, the antigenic determinant of LPS, were directly coupled to bovine serum albumin (BSA) by carbodiimide mediated condensation reaction. The molar ratios of saccharide repeating units to BSA in the prepared conjugates were 15:1 and 26:1 for IATS 6 and 11 conjugates, respectively. Mice were immunized with the conjugates emulsified with monophosphoryl lipid A (MPL), Freund, and Alum adjuvants. The conjugates emulsified with MPL adjuvant elicited the highest IgM antibody, followed by Freund. While both MPL and Freund adjuvants elicited high IgG antibody. Good correlation was observed between the IgG and IgM levels with the bactericidal activities of the sera against homologous strains. In addition, immunization of mice with the prepared conjugates emulsified with MPL and Freund adjuvants provided high protection against ten times the LD₅₀ of P. aeruginsoa IATS 6 and 11, which showed a good correlations with the IgG titer.     

Localization within the heavy chain sequence of tyrosyl peptides from the combining sites in a rabbit anti-3-azopyridine antibody preparation

A relatively homogeneous rabbit heavy chain was cleaved by CNBr. Fragment C-1 (the N-terminal half of the heavy chain) was isolated. Reduction and alkylation of C-1 liberated three fragments and partial sequence analysis of these isolated fragments showed that C-1 had been split on the carboxyl-side of Met 84. Similar results were obtained with another anti-hapten antibody preparation in which tyrosyl residues in the combining sites had been labeled. The labeled tyrosyl residues were found in the fragment representing residues 85–253. Since the constant region begins at about residue 120 and the sequences of the tyrosyl peptides from the combining sites are not present in published constant region sequences, these peptides appear to be derived from a variable region between residues 85 and 120.     

Studies on the antibody composition and neutralizing activity of tetanus antitoxin sera from various species of animals in relation to the antigenic substructure of the tetanus toxin molecule

On the basis of the antigenic substructure of tetanus neurotoxin, the antitoxin compositions of horse, rabbit and human tetanus antitoxin sera, in terms of their contents of antibodies against four antigenic determinant groups (alpha, beta-1, beta-2 and the "topographic" determinant group gamma) so far known for the toxin were studied by quantitative precipitation reactions using purified toxin, complementary fragments alpha, beta and fragment beta-1 (a subfragment of fragment beta) of the toxin. The antitoxin antibody composition varied slightly depending on the antiserum preparation. In addition, different patterns of antitoxin antibody composition and toxin-neutralizing ability, characteristic of horse, rabbit and man were found: horse antitoxin sera contained all four kinds of antibodies and horse anti-gamma showed low toxin-neutralizing ability, while human antisera lacked anti-alpha and had anti-gamma with high neutralizing activity but contained anti-beta-1 with no detectable neutralizing activity. Rabbit sera showed an intermediate pattern between those of horse and human sera. In all antisera, antibodies against determinants on the isolated fragment beta account for approximately 80-50 percent of the total precipitable antibodies and anti-beta-2 antibody was invariably present. Immunodiffusion analyses showed that the antitoxin compositions of mouse and guinea pig antisera resembled those of human antisera. In mice, fragment beta was almost as efficient as whole toxin toxoid in eliciting a protective immune response on an equal weight basis, whereas fragments beta-1 and alpha were both relatively poor antigens.     

Binding of pyridoxal 5'-phosphate to bovine serum albumin and albumin fragments obtained after proteolytic hydrolysis. Localization and nature of the primary PLP binding site.

The binding of pyridoxal 5'-phosphate (PLP) to bovine serum albumin (BSA), and to large BSA fragments obtained after proteolytic hydrolysis, was investigated in order to study the structure of these fragments in relation to the albumin structure itself, and to get information about the PLP binding sites on albumin. From absorbance and circular dichroism spectra, combined with peptide mapping of the tryptic digests of the reduced PLP-protein complexes, it could be concluded that the primary binding site is localized with the NH2-terminal part of the albumin molecule. The COOH-terminal part contains one or more secondary sites. It appeared that in albumin and in the largest NH2-terminal fragment, the environment of the primary binding site is rather apolar in character. However, in the smallest NH2-terminal fragment this site is more exposed to the solvent. This suggests that the part of the peptide chain which is not common in both fragments has a stabilizing effect on the structure around the primary binding site.     

Purification and characterization of peptic fragments derived from the carboxyl-terminal half of human albumin

Utilizing a combination of conventional and affinity-chromatographic procedures, we have purified four fragments of human albumin that were generated by controlled limited proteolysis with pepsin [0.3 mM albumin; 37°C; 10 min; pH 3.51; 4.2 mM octanoate; pepsin/albumin, 1:1000 (w/w)]. These fragments have a molecular weight range of 9200-17,000 Da. Amino acid compositions, N- and C-terminal sequences, molecular weights, and other internal markers were used to determine the location of these fragments within the parent molecule. All of the fragments were shown to be derived from the C-terminal half of human albumin. The presence of multiple pepsin-sensitive bonds near the C terminus of each fragment complicated the assignment of specific residue numbers to each fragment. Two pairs of similar peptides were identified: (A) those corresponding to a single-loop structure (residues 309–380 and 309–387) and (B) those containing multiple loops and intraloop cleavages [residues 309–(491–495) with 408–423 deleted]. Purification of these fragments without disulfide bond reduction confirms portions of the loop structure of human albumin and demonstrates increased susceptibility of two specific regions of the C-terminal half of the molecule to peptic digestion.     

Optical detection of triplet-state magnetic resonance studies on individual tryptophan residues of serum albumin: correlation between phosphorescence and zero-field splittings

Cyanogen bromide cleavage of bovine serum albumin (BSA) yields two fragments, N (1-183) and C (184-582), containing 183 and 399 amino acid residues, respectively. Each in each fragment are characterized in this study by phosphorescence and optically detected magnetic resonance spectroscopy, and the results are compared with those of the intact albumin. Trp-134 in fragment N is located in a hydrophobic environment in the interior of the protein, as reflected by its red-shifted phosphorescence and characteristic zero-field splittings. The spectral properties of Trp-212 in fragment C suggest its location in a partially buried, inhomogeneous environment. They show great similarity to those of human serum albumin, which contains a single Trp at position 214. The Trp phosphorescence 0,0-bands of fragments C and N are fitted with Gaussian functions by computer, and their relative contributions to the phosphoresence 0,0-band of BSA are adjusted to fit the observed BSA 0,0-band. The wavelength dependence of the [D[-[E[ transition frequencies of fragments N and C is then weighted by their 0,0-band intensity, taking into account differences in spin alignment, and summed to predict the peak frequency of the [D[-[E[ band profile as a function of phosphorescence wavelength for the intact BSA. Good agreement between predicted and observed behavior of [D[-[E[ vs. wavelength for the intact protein provides strong evidence for the additivity of the phosphorescence and ODMR spectra of the individual Trp sites in BSA. We find that Trp-134 and Trp-212 have wavelength-independent and wavelength-dependent zero-field splittings, respectively.     

Immunosuppressive properties of a peptic fragment of BSA.

The immunogenic properties of a peptic fragment of BSA were investigated. BSA was subjected to limited proteolysis by pepsin and the resulting fragments were separated on DEAE cellulose. The fragment under consideration, Fraction Ia (m.w. 8000 to 10,000), did not precipitate with anti-BSA serum but did inhibi, the binding of specific antibody to labeled BSA, indicating the presence of determinants found on the native antigen. BDF1 mice immunized with Fraction Ia in A1 (OH)3 gel or in complete Freund's adjuvant produced no significant antibody response as measured by passive cutaneous anaphylited a (PCA) or by a modified Farr assay. The fragment elicited a PCA reaction in mouse skin sensitized with anti-BSA serum. Treatment of mice with single doses of Fraction Ia at various time intervals before immunization with BSA resulted in significant suppression of the formation of anti-BSA antibody. The conditions of suppresion of the IgE response by the peptic fragment were studied in greater detail. Evidence is presented that such suppression can be attributed to the presence of specific T suppressor cells in our system.     

Amadori-glycated albumin-induced vascular smooth muscle cell proliferation and expression of inhibitor of apoptosis protein-1 and nerve growth factor-gamma

We investigated the effects of Amadori-glycated serum albumin (GSA) on cell proliferation as well as expressions of antioxidant enzyme genes and marker genes associated with signal transduction pathways in rat aortic vascular smooth muscle cells (VSMCs). Quiescent VSMCs treated with GSA (0-500 microg/mL, 48 h) exhibited a dose-dependent increase in proliferation that was prevented by PD98059 (25 microM), suggesting a MAPK-dependent signaling pathway. Compared with bovine serum albumin (BSA)-treated cells, the GSA (500 microg/mL, 24~h)-treated VSMCs showed a higher superoxide dismutase 2 gene expression in quantitative RT-PCR, suggesting the involvement of oxidative stress. In a focused oligonucleotide array containing 96 signal transduction-related genes, expression of inhibitor of apoptosis protein-1 (IAP-1), nerve growth factor-gamma (NGF-gamma), and c-jun genes was significantly higher in the GSA-treated VSMCs. These results suggest that induction of antiapoptotic proteins like IAP-1 and strong mitogens like NGF-gamma by GSA might further contribute to the VSMC proliferation and accelerated vascular remodeling in diabetes.