Enhanced Proatherogenic Responses in Macrophages and Vascular Smooth Muscle Cells Derived From Diabetic db/db Mice

Phosphospecific antibodies for p38^MAPK (Thr-180/Tyr-182), NF-κB-p65 (Ser-536), cAMP response element–binding protein (CREB) (Ser-133), STAT-3 (Tyr-705), Akt (Ser-473), p42/44^MAPK (Thr-202/Tyr-204), Src (Tyr-416), inhibitor of NF-κB kinase (IKK) α/β (Ser-180/Ser-181), and antibodies to total p38^MAPK and Akt were from Cell Signaling (Beverly, MA). COX-2 (murine) polyclonal antibody was from Cayman Chemical (Ann Arbor, MI), and 12/15-lipoxygenase (12/15-lipoxygenase) antibody has been described earlier (23). β-Actin antibody and all other reagents were from Sigma-Aldrich (St. Louis, MO) unless otherwise indicated.

All animal studies were performed in accordance with a protocol approved by the institutional research animal care committee. Mouse peritoneal macrophages were obtained by peritoneal lavage of 9- to 10-week-old male db/db mice. (BKS.Cg-m^+/+lepr^db/J, stock no. 000642; The Jackson Laboratory, Bar Harbor, ME). Control mice (db/+) were 9- to 10-week-old male heterozygotes from the same colony. Thioglycollate-elicited peritoneal macrophages were harvested 2–3 days after intraperitoneal injection of 2 ml of sterile 3% thioglycollate broth (Difco Laboratories, Detroit, MI). The cells were washed by centrifugation at 200g for 10 min with RPMI-1640 medium without serum, and cell pellets were resuspended in RPMI-1640 supplemented with 10% FCS, 100 units/ml penicillin, 100 μg/ml streptomycin, 50 μg/ml plasmocin, and 1,000 mg/l of glucose and seeded into six well plates. After 4 h, the cells were washed three times and attached cells were used immediately or cultured overnight at 37°C in a 5% CO₂ incubator for extraction of total RNA used for gene expression analyses. Resident macrophages were isolated from peritoneum using PBS and allowed to attach to dishes for 4 h. RNA from attached cells were used for RT-PCR analysis. Macrophage purity was determined by staining the attached cells with fluorescein isothiocyanate-conjugated F4/80 and phycoerythrin conjugated CD11b followed by flow cytometry analysis. Results showed that 96–98% of the attached cells were macrophages.

For Western blotting, after the cells were allowed to attach for 4 h, dishes were washed three times and cultured overnight in RPMI-1640 medium containing 0.5% FCS and then treated with or without TNF-α (10 ng/ml).

Aortic VSMCs from 9- to 10-week-old male db/+ and db/db mice (MVSMCs) were isolated by enzymatic digestion as described earlier (34) and cultured in Dulbecco’s modified Eagle’s (DME)-F12 culture medium supplemented with 10% FCS. MVSMCs from db/+ and db/db mice were used between passages 5 and 8. To make total RNA from aortic tissues, thoracic aortas were excised from db/+ and db/db mice and immediately frozen in liquid nitrogen. RNA was extracted using an RNAeasy kit (QIAGEN, Valencia, CA) according to the manufacturer’s instructions. RT-PCR was performed as described below.

Mouse monocytic cell line WEHI78/24 (generous gift from Dr. J.A. Berliner, University of California, Los Angeles, CA) was cultured in DME-F12 containing 10% fetal bovine serum.

Preparation of whole cell lysates and Western immunoblotting were performed as described earlier (35). Finally, the proteins on membranes were visualized using Western Lightning Chemiluminescence reagent (PerkinElmer Life Science, Boston, MA). To detect multiple proteins from a single membrane, membranes were stripped with Restore Western blot Stripping Buffer (Pierce, Rockford, IL) for 25 min at 37°C before reprobing with a different antibody. The density of protein bands was quantified using an GS-800 densitometer and QuantityOne software (Bio-Rad, Hercules, CA).

Total RNA was extracted from cells with RNA-STAT, and relative RT-PCR analyses were performed as described earlier (35,36) with gene-specific primers; 18S primers (Quantum RNA 18S primers; Ambion, Austin, TX) were used as internal standards. PCR products were fractionated on 2% agarose gel and photographed using FluorChem 8900 (Alpha Inotech, San Leandro, CA), and the intensity of DNA bands was determined with Quantity One software (Bio-Rad). Real-time quantitative RT-PCR analyses were performed using SYBR Green PCR Master Mix kit (Applied Biosystems, Foster City, CA) with an ABI 7300 real-time PCR thermal cycler (Applied Biosystems). Reaction conditions were 50°C for 2 min, 95°C for 10 min, followed by 40 cycles of 95°C for 15 s, and 58°C for 1 min. All reactions were performed in triplicate in a final volume of 25 μl. Standard curves were generated using GAPDH primers. Dissociation curves were run to detect nonspecific amplification, and we confirmed that single products were amplified in each reaction. The quantities of each test gene and internal control 18S RNA were then determined from the standard curve using the Applied Biosystems software, and mRNA expression levels of test genes were normalized to 18S levels. Primers used for PCR and real-time PCR in this study were obtained from the City of Hope’s DNA synthesis facility. PCR primers were based on known gene sequences and are available on request.

Intracellular superoxide production (index of oxidant stress) was evaluated with the use of a superoxide probe dihydroethidium (Molecular Probes, Eugene, OR) as described earlier (34).

Migration of MVSMCs from db/db and control mice was assayed using a 48-well microchemotaxis modified Boyden chamber apparatus as described earlier (23). Briefly, the lower wells of the chamber were filled with 27 μl of depletion medium alone or depletion medium containing 0.1–1.0 nmol/l PDGF. The top wells were filled with 52 μl (50,000 cells) of serum-depleted MVSMCs and incubated for 4 h at 37°C under 5% CO₂. The migrated cells on the undersurface of the filter were fixed and stained with hematoxylin. The stained membranes were scanned and staining intensities quantified with Quantity One software (Bio-Rad).

Mouse monocytic cell line (WEHI78/24) cells were preincubated in DME-F12 containing 2% serum and 5 μg/ml of the fluorescent dye 2′,7′-bis(carboxyethyl)-5(6′)-carboxyfluorescein/acetoxymethyl ester at 37°C for 30 min in foil-covered tubes. Fluorescent-labeled cells were washed twice to remove unincorporated dye and then resuspended in DME-F12 containing 0.2% BSA. Fluorescently labeled WEHI cells (5 × 10⁴/well) were added to MVSMC monolayers in 24 wells and incubated at 37°C. After 30 min, unbound monocytes were aspirated, and VSMC layers with attached monocytes were gently washed twice with DME-F12. Cells were lysed with 0.1% Triton X-100 in 0.1 mol/l Tris (0.2 ml/well), and fluorescence was measured with a fMax microplate reader (Molecular Devices, Sunnyvale, CA) (excitation 485 nm and emission 535 nm). Data were analyzed with SoftMax Pro. The number of adherent cells per well was expressed as percent fluorescence of control.

The culture supernatants of macrophages from db/db and db/+ mice were collected 24 h after culturing and centrifuged at 2,000 rpm for 10 min to remove floating cells and supernatants stored at −70°C. Levels of cytokines/chemokines IL-1β, IL-8, and JE/monocyte chemoattractant protein-1 (MCP-1) in these culture supernatants were analyzed by enzyme-linked immunosorbent assay (ELISA) using SearchLight Multiplexing kits (Pierce-Endogen, Rockford, IL), and the results are expressed as fold over db/+ cells.

Data were analyzed by using PRISM software (GraphPad, San Diego, CA). Student’s t tests were used to compare two groups for mean comparisons. P < 0.05 was considered statistically significant.

We first tested our hypothesis that macrophages from db/db mice would have increased expression of proinflammatory cytokines because they are implicated in the pathogenesis of type 2 diabetes, insulin resistance, and associated cardiovascular complications. The mean blood glucose levels in db/db mice were 472.6 ± 21 mg/dl compared with 151.13 ± 10.03 mg/dl in control db/+ mice (P < 0.001). Blood cholesterol and triglycerides were also significantly elevated but to a lesser degree (137 ± 8 vs. 104 ± 11 mg/dl for db/+, P < 0.0001; and 116 ± 23 vs. 72 ± 6 mg/dl for db/+, P < 0.001, respectively). Thioglycollate-elicited macrophages were isolated from 9- to 10-week-old db/db and control db/+ mice. Total RNA extracted from these macrophages was used to perform relative RT-PCR using both gene specific and 18S primers (internal control) as described in research design and methods. As shown in Fig. 1, the expressions of IL-1β (A), IL-6 (B), and TNF-α (C) were significantly increased by 3-, 20-, and 3-fold, respectively, in macrophages from db/db mice compared with db/+ mice.

Chemokines potently induce cellular chemotaxis and play important roles in the migration and recruitment of monocytes to endothelial cells and subendothelial retention at sites of the lesion (37). We next evaluated the expression of key chemokines IL-8 (mouse KC), JE/MCP-1, and interferon (IFN)-γ–inducible protein (IP-10) by RT-PCR. Results in Fig. 1 (D, E, and F, respectively) show that the expression of these chemokines is significantly upregulated by two- to threefold in db/db macrophages compared with db/+ macrophages.

Cytokines such as IL-10, IL-4, IL-12, IL-18, and IFN-γ are associated with T-helper-type (Th) 1 and Th2 responses and are key players in the regulation of immune system (38). Previous studies showed that db/db mice are more prone to bacterial infection and more susceptible to endotoxin, suggesting that the host defense system is compromised (31). Furthermore, it is now known that abnormal immune responses may also contribute to the development of atherosclerosis. We therefore examined whether the mRNA expression levels of key immunomodulatory cytokines are altered in macrophages from db/db mice. As shown in Fig. 2, the expression of IL-10 (A) and IL-12 (C) mRNAs were significantly increased in macrophages from db/db mice compared with db/+ mice. In particular, the expression of IL-12, a potent proinflammatory cytokine, was increased by more than eightfold. However, the expressions of IL-18 and IL-4 were not significantly altered (Fig. 2B and D, respectively) nor was that of IFN-γ (not shown). Thus, the altered expression of specific cytokines may contribute to dysregulated immune and inflammatory responses in these mice. IL-10 is typically associated with protective Th2 responses, and its increased expression in db/db mice macrophages may reflect an adaptive response.

Evidence suggests that the arachidonic acid–metabolizing enzymes, 12/15-lipoxygenase and COX-2 and their products, such as 12(S)-hydroxyeicosatetraenoic acid and prostaglandin E₂, respectively, have proinflammatory properties (39,40) and are induced in monocytes and VSMCs by high glucose (18,39). These pathways are also implicated in vascular injury responses and atherosclerosis (39–43). We therefore examined the expression of these two enzymes. Both 12/15-lipoxygenase (Fig. 3A) and COX-2 (Fig. 3B) mRNA levels were significantly elevated in macrophages from db/db mice by 1.9- and 5-fold, respectively, compared with those in db/+ macrophages.

We further confirmed the differential expression of key inflammatory genes in db/db macrophages using real-time quantitative RT-PCR analysis. As shown in Fig. 4, mRNA levels of 12/15-lipoxygenase (A), IL-1β (B), IL-6 (C), and IL-8 (D) were significantly increased in db/db macrophages compared with those in db/+ cells. These results further support results obtained by relative RT-PCR. In addition, we examined the levels of secreted cytokine IL-1β and chemokines IL-8 and MCP-1 in the culture supernatants of macrophages by ELISA. Results showed that levels of these cytokines and chemokines were significantly elevated in culture supernatants from db/db mice compared with those in db/+ cells (Fig. 4E).

We next examined whether nonelicited, resident macrophages also behave similarly. Results showed that, similar to the thioglycollate-elicited macrophages, resident peritoneal macrophages isolated from db/db mice also clearly expressed markedly higher levels of key inflammatory genes including IL-6 (12.5 ± 1.5-fold), IL-8 (2.6 ± 0.15-fold), and MCP-1 (1.43 ± 0.025-fold), 12/15-lipoxygenase (1.7 ± 0.10-fold), and COX-2 (3.75 ± 0.25-fold) relative to those from db/+ mice. These results further confirm that macrophages from db/db mice exhibit enhanced inflammatory gene expression.

To determine the potential mechanisms for the increased expression of inflammatory genes in the db/db macrophages, we next examined levels of oxidant stress as well as activation of key signaling kinases and transcription factors. To detect oxidant stress, isolated macrophages were incubated with the cell-permeable dye dihydroethidium in the presence or absence of TNF-α (10 ng/ml), washed to remove excess dye, and examined under a fluorescence microscope. Dihydroethidium is converted to ethidium bromide by superoxide, which then binds to DNA and emits red fluorescence. As shown in Fig. 5A, db/db macrophages have increased basal superoxide production relative to db/+ cells. Stimulation with TNF-α increased oxidant stress in control db/+ cells. However, TNF-α did not induce further increases in fluorescence over basal in the db/db macrophages. Thus, db/db macrophages have increased basal oxidant stress that may contribute to their preactivated state.

We next examined the activation of key signaling kinases known to be involved in the activation of transcription factors such as NF-κB that regulate inflammatory gene expression. Total cell lysates were immunoblotted with phosphospecific antibodies that recognize activated forms of the MAPKs extracellular signal–regulated kinase (ERK)1/2 and p38, as well as IKKα/β, an upstream kinase known to activate NF-κB. Results showed that TNF-α (10 ng/ml)–induced activation (phosphorylation) of ERK1/2, p38^MAPK, and IKKα/β was significantly higher in db/db macrophages compared with db/+ macrophages (Fig. 5B and D). Immunoblotting with an antibody to total p38^MAPK showed that equal amounts of protein were loaded in all the lanes (bottom panel).

We then examined whether there is differential activation of downstream transcription factor targets of these kinases, such as CREB and NF-κB, which are known to be involved in the regulation of inflammatory genes. The cell lysates were immunoblotted with phosphospecific antibodies that recognize activated forms of these transcription factors. Phosphorylation (activation) of CREB at Ser-133 and p65 (active subunit of NF-κB) at Ser-536 stimulated by TNF-α was significantly enhanced in db/db cells compared with db/+ cells (Fig. 5C and D). Thus, enhanced activation of CREB and NF-κB may contribute to the upregulation of inflammatory genes in db/db cells.

We also examined the activation of another transcription factor STAT3, which has been shown to mediate expression of certain growth and inflammatory genes. Immunoblotting was performed with a phospho-STAT3 antibody that recognizes STAT3 phosphorylated at Tyr-705. Interestingly, STAT3 phosphorylation was greatly attenuated in db/db macrophages cells in the basal state, and TNF-α failed to stimulate this tyrosine phosphorylation of STAT3 in cells from both the animals (Fig. 5C). Lack of STAT3 activation in db/db macrophages (that are leptin receptor deficient) further confirms previous findings that the leptin receptor regulates STAT3 activation (32).

Activation of VSMCs plays an important role in the development of atherosclerosis. Furthermore, because monocyte to macrophage transitions take place in the subendothelial space, VSMCs and monocyte/macrophages may cooperate to augment diabetic vascular dysfunction (26,44). Hence, we next examined inflammatory gene expression in aortic VSMCs isolated from db/+ and db/db mice. Experiments were performed with cells from passages 5–8. RNA was extracted from VSMCs grown to 80% confluency and analyzed by RT-PCR as described in research design and methods. Results showed that expressions of both MCP-1 (Fig. 6A) and IL-6 (Fig. 6B) were significantly increased in db/db cells compared with db/+ cells, suggesting that the VSMCs from db/db mice are also in a preactivated state.

We also examined the expression of inflammatory arachidonic acid–metabolizing enzymes COX-2 and 12/15-lipoxygenase. Results showed that, similar to macrophages, the expression levels of 12/15-lipoxygenase and COX-2 were increased in VSMCs from db/db mice compared with those from db/+ mice (Fig. 6C and D, respectively).

To further determine whether the differences observed in isolated cells are mirrored in whole aortas, we isolated RNA from intact aortas excised from these mice and examined the expressions of IL-6 and MCP-1. Results showed that expressions of both IL-6 and MCP-1 were significantly increased in db/db mouse aortas compared with those in db/+ mice (Fig. 6E and F, respectively).

Next we examined oxidant stress levels and the activation of signaling pathways that are involved in inflammatory gene expression or VSMC migration. Staining of VSMCs with dihydroethidium showed that, similar to macrophages, VSMCs derived from db/db mice had higher basal oxidant stress (superoxide production) (Fig. 7A), and although TNF-α could increase superoxide production in db/+ mice, there was no further increase by TNF-α in db/db VSMCs. This finding confirms that both macrophages and VSMCs from these db/db mice are in a preactivated state.

Cell lysates from VSMCs were also immunoblotted with phosphospecific antibodies that recognize the activated forms of the indicated proteins. Results showed that the activation (phosphorylation) of key kinases associated with VSMC migration and growth, namely, Src kinase, ERK1/2, and Akt and p38^MAPK were significantly enhanced in db/db VSMCs compared with those in db/+ cells. (Fig. 7B and D). Immunoblotting with total p38^MAPK antibody showed that equal amounts of proteins were loaded in all the lanes. Phosphorylation (activation) of transcription factors CREB at Ser-133 and active subunit p65 of NF-κB at Ser-536 were also significantly elevated in VSMCs derived from db/db mice compared with those from db/+ cells (Fig. 7C and D). Thus, these results demonstrate that the activation of specific signaling pathways and related downstream key transcription factors associated with VSMC activation and inflammatory gene expression are enhanced in db/db VSMC.

VSMC migration plays a key role in the pathology of restenosis and atherosclerosis. Therefore, we next examined whether migratory responses were altered in VSMC derived from db/db mice. Early passage VSMCs isolated from db/db and db/+ mice were subjected to migration assays as described in research design and methods. Results are shown in Fig. 8A. The basal unstimulated level of migration of MVSMCs from db/db mice was significantly higher than those from db/+ mice. Furthermore, VSMCs from db/db mice were clearly more sensitive to the effects of PDGF-BB, a known VSMC chemotactic agent. In fact, the lower dose of PDGF (0.01 μM) induced significant migration of VSMCs derived only from db/db mice (by 3.5-fold) but not from db/+ mice. Thus these results clearly demonstrate that VSMCs from db/db mice are preactivated and also hypersensitive to extracellular chemotactic stimuli.

We next examined whether the increased inflammatory chemokine gene expression observed in the db/db MVSMCs can lead to enhanced adhesion of monocytes because evidence shows that heterotypic interactions between these cell types may enhance monocyte to macrophage differentiation (44). Adhesion assays with WEHI mouse monocytes and MVSMCs were carried out as described earlier (26,44) and in research design and methods. Results shown in Fig. 8B demonstrate that MVSMCs derived from db/db mice clearly display augmented adhesion of WEHI mouse monocytes compared with control VSMCs. Thus, these results further confirm that VSMCs in db/db mice exist in a preactivated state.

Diabetic conditions can lead to the production of several inflammatory mediators, including chemokines and cytokines, that are associated with the development of accelerated cardiovascular complications such as atherosclerosis (45). In the current study, we examined the ex vivo responses of macrophages and VSMCs derived from db/db mice, a model of obesity, insulin resistance, and type 2 diabetes, to determine whether these key cells associated with vascular complications display a preactivated phenotype and altered pathophysiological responses. We found that the expression levels of key proinflammatory cytokines IL-1β, IL-6, and TNF-α were enhanced in macrophages from db/db mice. An elevated level of inflammatory cytokines is generally a strong predictor of diabetes complications as well as insulin resistance. Furthermore, these cytokines also regulate the innate immune system and thus their dysregulated expression can impair the capacity of the immune system to appropriately regulate inflammation.

Expression levels of the chemokines JE/MCP-1, IL-8, and IP-10 were also increased in the macrophages from db/db mice. Chemokines play an important role in the development of atherosclerosis by inducing migration of leukocytes to the site of inflammation. Chemokines can also promote the migration of monocytes into the subendothelial space where they differentiate into macrophages. In addition, chemokines can induce the migration of VSMCs from the media to the intima, leading to the development of neointimal thickening and lesion formation.

Interestingly, we also noted that the levels of key immunomodulatory cytokines IL-12 and IL-10 were significantly increased in macrophages from db/db mice relative to those of control mice. However, IL-4, IL-18, or IFN-γ were not significantly altered. IL-12 is a potent inflammatory and proatherogenic cytokine (46) that also plays an important role in regulating Th1 immune responses. It can also in turn further activate T-cells, macrophages, and other vascular cells to release more cytokines and chemokines. IL-10 and IL-4 are usually associated with Th2 responses. Increased expression of IL-10 in db/db macrophages may reflect an adaptive response.

Our results also showed increased oxidant stress in both macrophages and VSMCs of db/db mice, further supporting the notion that diabetes, insulin resistance, and metabolic disorders are associated with enhanced oxidant stress. We also found an increase in the expression of arachidonic acid–metabolizing enzymes 12/15-lipoxygenase and COX-2 in both macrophages and VSMCs. These enzymes and their products have proatherogenic effects in monocytes, endothelial cells, and VSMCs (39–43).

The activities of key signaling pathways associated with inflammatory gene expression were found to be enhanced in db/db macrophages. These included TNF-α–induced ERK1/2, p38^MAPK, and IKKα/β. Upon activation, these kinases can phosphorylate various transcription factors, leading to the regulation of target genes. We observed that the activation of CREB and NF-κB transcription factors was elevated in macrophages of db/db mice compared with those of db/+ mice. Thus, the increased activation of signaling kinases and downstream oxidant-sensitive transcription factors in the basal state and in response to cellular stimuli might be key mechanisms for the augmented expression of inflammatory gene expression in these mice. Activation of STAT3 in the hypothalamus of db/db mice is defective because of the lack of the leptin receptor. Our results showed that STAT3 activation is also defective in macrophages, suggesting that JAK/STAT signaling is impaired in cells other than the hypothalamus in these mice.

Interestingly, VSMCs from db/db mice also showed enhanced expression of JE/MCP-1 and IL-6. These cytokines/chemokines are involved in VSMC proliferation and migration and can also attract monocytes to bind to VSMCs. As recently shown, monocyte adhesion to VSMCs may be an important event in leading to monocyte retention, differentiation, and foam cell formation within the subendothelial space (26,44). Our results also showed that key signals involved in VSMC proliferation and migration are enhanced in db/db VSMCs. These include oxidant stress (superoxide production), Src kinase, ERK1/2, MAPK, Akt, and transcription factors CREB and NF-κB. Furthermore, we were able to demonstrate functional relevance of the increased expression of these inflammatory markers because migratory responses were greatly enhanced in VSMCs derived from db/db mice. We have noted that MVSMCs from db/db mice also display 20% faster rates of proliferation relative to those from db/+ mice (L.L., M.A.R., and R.N., unpublished results).

We observed that resident macrophages also have increased expression of these inflammatory genes, suggesting that changes in gene expression and other signaling responses are not due to artifacts of thioglycollate treatment. Aortas from db/db mice also showed increases in inflammatory genes, further demonstrating that the observed changes in VSMCs were not due to artifacts of culturing and that the changes are maintained during the culturing process. Evidence shows that similar changes are also seen in aortas of Zucker fatty rats, a well-established rat model of insulin resistance, relative to control lean rats (47). Recently it was shown that TNF-α induced toxicity, monocyte-endothelial adhesion, lipopolysaccharide-induced inflammatory responses, and IL-1β–mediated innate immunity were amplified in db/db mice (30,31). Endothelial cells derived from db/db mice also display a similar preactivated phenotype (48). Increases in IL-12 levels have recently been reported in mouse macrophages stimulated with glucose in vitro as well in macrophages isolated from streptozotocin-induced diabetic mice and db/db mice (49). In another very recent study, we reported that VSMCs from db/db mice displayed increased AGE receptor expression, Src kinase activation, and migration (50). Thus, our new data along with those of others indicate that vascular cells including VSMCs, endothelial cells, and macrophages are in a preactivated state in db/db mice, and this could be a major underlying cause for the reported predisposition of db/db mice to develop accelerated atherosclerosis (33). The fact that some of these cellular responses, particularly in VSMCs, are evident even after a few passages lend support to the presence of a novel “hyperglycemic memory” that has been suggested to be the cause for the continued development of complications in patients with diabetes even after conventional glycemic control (51). Taken together, our results provide mechanistic insights for the augmented inflammation and accelerated cardiovascular complications associated with type 2 diabetes, insulin resistance, and atherosclerosis.

These studies were supported by grants from the National Institutes of Health (PO1 HL55798 and RO1 DK65073) and the American Diabetes Association.

The authors thank Mei Wang for excellent technical assistance, Dr. Patrick Lundberg (Department of Virology, Beckman Research Institute), and the Flow Cytometry Core at City of Hope for their help.