Hypertension is associated with chronic vascular inflammation. We tested the hypothesis that the sensitivity to develop hypertension and vascular remodeling depends on the immunological background. Blood pressure, vascular remodeling, endothelial function, vascular architecture (number of collateral arteries), and expression of inflammatory cytokines were determined in mice that received NG-nitro-l-arginine methyl ester (l-NAME) to inhibit nitric oxide synthesis. We studied C57BL/6, BALB/c, and BALB.B6-Cmv1r mice, a congenic strain where the natural killer (NK) gene complex of C57BL/6 mice is introduced in the BALB/c background. During a 4-wk treatment with l-NAME, blood pressure initially increased in both C57BL/6 and BALB/C mice, but after 4 wk, only C57BL/6 mice showed a significant increase in mean arterial blood pressure (+53 mmHg; P < 0.001) and small artery inward remodeling. Endothelial function and vascular design were significantly different between C57BL/6 mice and BALB/C mice. The inflammatory response was similar in C57BL/6 and BALB/C mice, except for the leukocyte marker CD11b. Cellular colocalization of CD11b with NK1.1 indicated the recruitment of NK cells in C57BL/6 mice. Congenic BALB.B6-Cmv1r mice showed the same endothelial response and vascular architecture as BALB/c mice. However, BALB.B6-Cmv1r mice displayed a similar sensitivity to hypertension and vascular remodeling as C57BL/6 mice. In conclusion, we have identified the NK gene complex as an important determinant in the genetically determined sensitivity to develop l-NAME-induced hypertension in mice.
- vascular inflammation
- vascular remodeling
- natural killer cells
- nitric oxide
an increased tissue expression and plasma concentration of inflammatory markers are found in hypertensive patients and experimental animal studies (22). One of these markers, C-reactive protein, predicts the future development of hypertension in humans (24, 27). Mice that lack the proinflammatory cytokine IL-6 develop less hypertension (15). Furthermore, target organ damage in hypertension is attenuated by immunosuppression (19) and depends on endothelial NF-κB-regulated genes VCAM-1 and ICAM-1 and subsequent inflammation (12). Hypertension is therefore increasingly viewed as a disease with a strong inflammatory component. Recent work indicates that both innate and adaptive immunities play important roles in the genesis of hypertension and target organ damage. In mice, upon the infusion of angiotensin II, T lymphocytes infiltrate the perivascular adipose tissue, increase the local formation of oxygen radicals, and induce the hypertension and inward remodeling of small arteries (10). Evidence for a role of macrophages in hypertension comes from studies in granulocyte-macrophage colony-stimulating factor-deficient mice. Thus these animals lack proper macrophage function and develop less endothelial dysfunction, small artery remodeling, and oxidative stress after the infusion of angiotensin II (6).
The inflammatory response is directed by, among others, cytokines derived from T lymphocytes (9). In particular, proinflammatory cytokines such as IFN-γ derived from T-helper 1 (TH1) lymphocytes and natural killer (NK) cells promote inflammation, extracellular matrix destruction, and apoptosis. On the other hand, IL-4 and IL-13 from TH2 lymphocytes support extracellular matrix construction, cell proliferation, and angiogenesis. Several lines of evidence relate differences in immune bias to cardiovascular disease. Thus it is well known that TH1-biased mice such as the C57BL/6 strain (18) are sensitive to the development of atherosclerosis, whereas the TH2-biased BALB/c strain is resistant. On the other hand, C57BL/6 mice show a strong arteriogenic response that alleviates critical hindlimb ischemia, which is lacking in BALB/c mice (26). Whether such differences in the immune background also predispose to the development of hypertension is not known. Therefore, in this study we tested the hypothesis that the development of hypertension depends on the immune background. We first compared the C57BL/6 mouse strain (TH1 biased) and BALB/c mice (TH2 biased) with respect to blood pressure, small artery remodeling, endothelial function, and expression of inflammatory markers after treatment with the nitric oxide inhibitor NG-nitro-l-arginine methyl ester (l-NAME). The results show a high susceptibility for hypertension in C57BL/6 mice compared with BALB/c mice. This difference, however, could not be related to a clear TH1 versus TH2 response. In addition, the treatment of BALB/c mice with T-cell receptor V (TCR-V) peptides to promote a TH1 response increased blood pressure only slightly. We did find a selective increase in the expression of the leukocyte integrin CD11b in the arteries of C57BL/6 mice, which represented the influx of NK cells. Therefore, we next studied BALB.B6-Cmv1r mice, which is a congenic mouse strain where the NK gene complex of C57BL/6 mice is introduced in the BALB/c background (23). The results indicate that this gene complex plays an important role in the sensitivity to develop hypertension.
Twelve-week-old male mice, C57BL/6 and BALB/c, were obtained from Harlan (Horst, The Netherlands). BALB.B6-Cmv1r mice were originally obtained from Prof. W. M. Yokoyama (St. Louis, MO) and bred in our own facilities. Mice were randomly assigned as untreated controls or received the nitric oxide inhibitor l-NAME for 3 days or 4 wk. Some BALB/c mice were additionally treated with a mixture of TCR-Vβ5.2 and TCR-Vβ8.1 peptides (Caslo, Denmark). These peptides modulate the TH1/TH2 balance toward TH1 (17). The peptides (200 μg of each peptide per injection for each mouse) were injected intraperitoneally on days 0, 9, and 18 after the start of l-NAME treatment. l-NAME (Sigma) was added to the drinking water at a dose of 1 g/l. During the study, water intake was determined every 2 days and found to be similar in C57BL/6 and BALB/c mice (3.0 and 3.1 ml·day−1·mouse−1, respectively). The weight of the animals ranged from 26 to 29 g. The investigation conforms with the Guide for the Care and Use of Laboratory Animals, published by the National Institutes of Health (NIH Publication No. 85-23, Revised 1996). The local ethics committee for animal experiments approved this study.
Untreated mice and mice treated for 3 days or 4 wk with l
Second-order branches from the superior mesenteric artery from controls and treated mice were harvested. To determine the passive diameter and wall thickness, the segments were mounted in a pressure myograph filled with calcium-free MOPS buffer containing (in mol/l) 145 NaCl, 4.7 KCl, 1.2 NaH2PO4, 1.2 MgSO4, 3 3-(N-morpholino)-propane sulfonic acid, 5 glucose, and 2 pyruvate (pH 7.4). The isolated vessels were tied to glass cannulas on both ends. The vessels were equilibrated to 37°C. The organ bath was mounted on top of a microscope and equipped with a digital camera that was connected to a computer. The diameter and wall thickness of the vessels were measured using MatLab software and recorded continuously. The passive pressure-diameter relationship (5 to 120 mmHg) of the arteries was measured after full dilation with papaverine (10−4 mol/l). From each mouse, two vessels were measured and the data were averaged. The wall-to-lumen ratio was determined at 80 mmHg. We chose this pressure level since we expect this to be close to the in vivo pressure in the mesenteric circulation of untreated mice. To allow a comparison of wall properties with l-NAME-treated mice, all vessels were tested at the same pressure level, according to suggestions by Bund and Lee (4).
To determine endothelial function, segments of the carotid artery were isolated and mounted in a Mulvany wire myograph (Danish Myo Technology). After normalization and precontraction with the thromboxane analog U-46619 (10−7 M), a dose-response relationship was determined for methacholine (10−8−3 × 10−6 M).
Design of the mesenteric vascular bed.
Collateral connections between the perfusion territories of second-order mesenteric arteries were counted under a dissection microscope. Data are given as percentages of the total number of perfusion territories.
Quantitative real-time PCR.
From control mice and mice treated with l-NAME for 3 days and 4 wk, total RNA was extracted from mesenteric arteries and spleen using TriReagent (Sigma). Reverse transcription (RT)-PCRs were performed using the Omniscript RT kit (Qiagen) and oligo (dT)-15 (Promega). Real-time PCR was performed with IQ SYBR Green Supermix buffer (Bio-Rad) using a MyIQ thermal cycler (Bio-Rad). In each experiment, the relative amounts of mRNA for candidate genes were calculated from the threshold cycle numbers and normalized to the relative amounts of reference gene RNA. Real-time PCR primers are listed in Table 1.
For confocal microscopy, fresh mesenteric arteries were washed in PBS and incubated overnight at 4°C with antibodies. For macrophage staining (CD68), the vessels were fixed in 4% formalin and membrane permeabilization was done using 0.1% Triton X-100 for 2 min. FITC-coupled rat anti-mouse CD68 (1:10, Serotec), Pacific blue-coupled rat anti-mouse CD11b (1:10, Biolegend), and phycoerythrin-coupled mouse anti-NK1.1 (PK136, 1:10, Abcam) were used. Vessels were embedded in Tissue-Tek (Sakura) and placed on a microscope slide. Fluorescent images were made using a confocal microscope (model TCS-SP2, Leica). Excitation wavelengths were 405, 488, and 560 nm, and section thickness was 0.37 μm.
A Student's t-test or one-way ANOVA was used to assess the differences among the defined groups with SPSS version 12.1. The values obtained from the treatment groups were compared with control values using Dunnett's t-test. All data are reported as means ± SE.
Blood pressure and vessel remodeling.
The effect of nitric oxide blockade on blood pressure was tested after 3 days and 4 wk of treatment in both BALB/c and C57BL/6 mice. The mean arterial pressure of untreated BALB/c mice (117 ± 3 mmHg) was significantly higher than C57BL/6 mice (101 ± 4 mmHg). After 3 days of l-NAME treatment, blood pressure was significantly elevated in both strains. However, after 4 wk, mean arterial blood pressure was not significantly increased by nitric oxide inhibition in BALB/c mice (+9 mmHg; not significant). In C57BL/6 mice, a profound increase in blood pressure was observed, to 154 ± 5 mmHg (+53 mmHg; P < 0.001). This increase in blood pressure was significantly higher in C57BL/6 compared with BALB/c mice (P < 0.001). The average data on blood pressure are shown in Fig. 1A.
To quantify vascular remodeling, the lumen diameter and wall thickness of the mesenteric arteries were determined at 80 mmHg. These data were used to calculate the wall-to-lumen ratio, a parameter that allows for the comparison of vessels from different sizes. In BALB/c mice, the wall-to-lumen ratio was not altered after 4 wk of l-NAME treatment: 0.104 ± 0.003 vs. 0.113 ± 0.003 (untreated vs. l-NAME). In C57BL/6 mice, a highly significant increase in the wall-to-lumen ratio was observed with l-NAME treatment, from 0.104 ± 0.002 to 0.126 ± 0.003 (P < 0.001). For both underlying parameters, wall thickness and lumen diameter, a nonsignificant change was observed. Thus the lumen diameter showed a tendency to decrease and the wall cross-sectional area tended to increase with l-NAME treatment (Table 2). Between the two strains, the increase in the wall-to-lumen ratio by l-NAME treatment was significantly different (P < 0.001). Average data are shown in Fig. 1B. Heart rate was increased in C57BL/6 mice treated with l-NAME but not in BALB/c mice. The relative heart weight was unchanged in both strains (Table 2).
Additional experiments using a lower dose (0.5 g/l) of l-NAME were done to rule out a potential interference of toxicity from an overdosage of l-NAME. After a 4-wk period of l-NAME treatment, still only C57BL/6 mice became hypertensive: 119 ± 6 mmHg (+16 mmHg; n = 5; P < 0.05), albeit to a lesser extent as with the high-dose group. BALB/c mice did not respond to the lower dose of l-NAME treatment with a sustained increase in blood pressure: 123 ± 10 mmHg (+6 mmHg; n = 5).
To gain insight in the phenotype of the inflammatory response, the systemic effects of l-NAME treatment on the expression of inflammatory cytokines were studied in spleen tissue. The gene expression of proinflammatory cytokines IFN-γ, IL-1, TNF-α, and IL-4 and the anti-inflammatory cytokine IL-10 were determined after 3 days and 4 wk of l-NAME treatment. In both strains of mice, a complex inflammatory response was found after 3 days of l-NAME treatment. Hypertensive BALB/c mice demonstrated a fivefold upregulation of IFN-γ gene expression after 3 days of l-NAME treatment and a decrease in IL-1 gene expression. The expression of TNF-α and IL-4 was unchanged, whereas IL-10 expression significantly decreased. Data are shown in Fig. 2, left. Despite a continuous treatment with l-NAME, after 4 wk of l-NAME treatment, no differences in the expression of any of the cytokines were noted compared with untreated BALB/c mice.
In C57BL/6 mice, a significant sixfold increase in the expression of IFN-γ was found after 3 days of l-NAME treatment. The expression of IL-1 decreased, whereas the expression of TNF-α and IL-4 was not significantly changed. The expression of IL-10 was significantly decreased. Data are shown in Fig. 2, right. No difference in the expression of any of the cytokines was found after 4 wk of treatment with l-NAME compared with untreated mice.
To study the local inflammatory response to l-NAME treatment, mesenteric arteries were analyzed for the expression of CD68, a macrophage marker; CD11b, a leukocyte integrin expressed by macrophages, NK cells, and neutrophils; factor XIII (FXIII); tissue-type transglutaminase (TG2); and transforming growth factor-β (TGF-β). Both FXIII and TG2 are transglutaminases and were previously identified as crucial enzymes in small artery remodeling (1, 3, 21). In BALB/c mice after 3 days of l-NAME treatment, a significant increase in the expression of CD68 was observed. The expression of CD11b, however, was not significantly changed compared with that in untreated mice. FXIII expression increased significantly (8-fold) after 3 days l-NAME treatment, whereas TG2 gene expression did not differ compared with that in untreated mice. The expression of TGF-β tended to decrease in 3-day hypertensive BALB/c mice. The average data are shown in Fig. 3, left. After 4 wk of l-NAME treatment, there was no difference in gene expression in l-NAME-treated mice compared with untreated mice.
Similar to BALB/c mice, C57BL/6 mice subjected to 3 days of l-NAME treatment showed a significant increase in the expression of CD68. However, opposite to BALB/c, a marked increase in the expression of CD11b was found. The expression of FXIII also significantly increased, whereas TG2 and TGF-β gene expression did not significantly change after 3 days of l-NAME treatment. The average data are shown in Fig. 3, right. After 4 wk of l-NAME treatment, none of the tested genes was differently expressed from untreated C57BL/6 mice.
The phenotype of the inflammatory response did not demonstrate a clear TH1/TH2 dichotomy in BALB/c versus C57BL/6 mice. To further study the relevance of the TH1/TH2 balance in hypertension, we aimed to promote a TH1 response in BALB/c mice using TCR-V peptides. The combined treatment with l-NAME and TCR-V peptides caused a small, but significant, increase in blood pressure compared with untreated controls: 132 ± 3 mmHg vs. 117 ± 3 mmHg (P < 0.05). Data are shown in Fig. 4A. Unexpectedly, the additional treatment with TCR-V peptides did cause a substantial increase in the wall-to-lumen ratio in BALB/c mice to 0.136 ± 0.007 (P < 0.001; untreated vs. l-NAME + TCR). Data are shown in Fig. 4B. The wall cross-sectional area of the vessels from animals treated with l-NAME and TCR peptides was not significantly different from untreated controls (Table 2).
NK gene complex.
The increase in expression of CD11b in C57BL/6 mice was the only noted difference in the inflammatory response between the two strains. We therefore further explored the source of CD11b expression in the vessels of C57BL/6 mice treated with l-NAME for 3 days, using confocal microscopy. We found that staining for CD11b colocalized with the staining for NK1.1, suggesting that most of these cells are NK cells. In arteries from C57BL/6 mice, we also observed CD68-positive, CD11b-negative cells in the adventitia of the arteries, indicating the presence of macrophages. We did not stain vessels from BALB/c mice, since these did not show an increase in the expression of CD11b and do not express NK1.1. Representative images are shown in Fig. 5. Building upon these findings, we continued our work using BALB.B6-Cmv1r mice. This strain expresses the NK gene complex of the C57BL/6 strain in a BALB/c background. We observed that these mice display a similar sensitivity to develop hypertension and remodeling as C57BL/6 mice. Thus blood pressure increased from 111 ± 11 to 155 ± 7 mmHg for untreated versus l-NAME-treated mice. The wall-to-lumen ratio increased from 0.107 ± 0.003 to 0.125 ± 0.007 for untreated vs. l-NAME-treated mice. Data for increases in blood pressure (as compared with untreated controls) are shown in Fig. 4A and for vascular remodeling in Fig. 4B.
Endothelial function and vascular design.
To study endothelial function and, in particular, the impact of the l-NAME treatment, we tested the response to methacholine in carotid artery segments of the mice. Here we chose carotid arteries because these vessels fully depend on nitric oxide to mediate the response to methacholine (7). We found that the relaxation to methacholine was reduced in both strains after 3 days of l-NAME treatment (Fig. 6, A and B). Of note, we observed a remarkably smaller response in untreated BALB/C mice compared with C57BL/6 mice (Fig. 6, A vs. B). Congenic BALB.B6-Cmv1r mice showed a similar response as BALB/C mice.
An interesting aspect of the mesenteric circulation of C57BL/6 and BALB/c mice was the large difference in the number of collateral arteries. A twofold difference in the percentage of collateral arteries was observed. The relatively low number of collateral arteries in BALB/c mice was also found in the congenic BALB.B6-Cmv1r strain. Data are shown in Fig. 6C. Typical images are shown in Fig. 6D.
The main new finding of the present study is that C57BL/6 mice are highly susceptible to develop l-NAME-induced hypertension and vascular remodeling compared with the BALB/c strain. The C57BL/6 and BALB/C strain differed in their inflammatory response and endothelium-dependent dilation and also markedly differed in vascular design. The difference in the inflammatory response was an increased expression of CD11b in vessels from C57BL/6 only, which reflected the recruitment of NK cells. Building up to these findings, further experiments were done with the congenic BALB.B6-Cmv1r mouse strain, which expresses the NK gene complex of the C57BL/6 in a BALB/c background. These mice showed a similar endothelial response and vascular design as BALB/C mice. However, they showed a similar sensitivity to develop hypertension and vascular remodeling as the C57BL/6 strain. These results, therefore, suggest that the NK gene complex is an important determinant in the genetically determined sensitivity to develop hypertension and remodeling after inhibition of nitric oxide.
Earlier suggestions for a difference in the sensitivity to develop hypertension in different mouse strains comes from the work of Yu et al. (31). Although these authors focused on cardiac remodeling, they also used l-NAME to induce hypertension in BALB/c, C57BL/6, and SCID mice and found that systolic blood pressure increased to 112 mmHg (+35 mmHg) in BALB/c and to 131 mmHg (+45 mmHg) in C57BL/6 mice. We found more pronounced differences in blood pressure, which may relate to the methods of blood pressure measurement (tail cuff vs. catheter) and the age of the mice (2 vs. 3 to 4 mo). Similar to assumptions of these authors with respect to cardiac remodeling, we first hypothesized that the difference between C57BL/6 and BALB/c mice in terms of hypertension and vascular remodeling could be related to a genetically determined bias toward a TH1 versus TH2 immune response. However, the expression of inflammatory cytokines in spleen tissue was rather similar between C57BL/6 and BALB/c mice. The BALB/c strain perhaps showed an increase in the TH2 cytokine IL-4, but this did not reach statistical significance. Thus, despite a clear difference in the immune background of these two strains (18, 31), their overall response to l-NAME treatment was quite similar. To further study the relevance of the TH1/TH2 balance in the context of hypertension and vascular remodeling, we aimed to promote a TH1 response in BALB/c mice with TCR peptides. These TCR-Vβ8.1 and -β5.2 peptides resemble parts of the TCR variable domains and antagonize autoantibodies that spontaneously arise with aging and after infection (17). An injection of these peptides increases the TH1/TH2 balance during immunosuppression and aging in mice (17). In our hands, the peptides only slightly increased blood pressure when combined with l-NAME but induced a disproportional inward vascular remodeling. The reason for this strong remodeling response is unclear but could relate to the stimulatory effect of T cells from TCR-V peptide-treated mice on fibroblast collagen and matrix metalloproteinase-13 synthesis (17).
For the local vascular response to l-NAME treatment, we focused on leukocyte markers and factors that could be causally related to remodeling. The temporal pattern of the local vascular responses paralleled the systemic response, with differences only after 3 days of l-NAME treatment. This temporal pattern of the inflammatory response also resembles that of flow-induced remodeling. Thus we recently found that inflammation is present only during the first stages of remodeling after a surgically modified blood flow (2). In the present study, we determined the expression of CD68, CD11b, the profibrotic cytokine TGF-β, tissue-type transglutaminase, and FXIII. With the exception of CD11b, this revealed no obvious differences between both strains. We studied transglutaminases here since previous work from our group showed that inward remodeling associated with hypertension (21) and reduced blood flow depends on transglutaminases (1, 3). These enzymes stabilize extracellular matrix proteins through glutamine-lysine cross-links, which greatly enhance mechanical and chemical stability (16). There was, however, no difference in the expression of tissue-type transglutaminase with l-NAME treatment in both strains. Besides tissue-type transglutaminase, the plasma transglutaminase FXIII is also expressed in small mesenteric arteries of mice. FXIII is well known as the fibrin-stabilizing factor of blood coagulation. Macrophages are a source of FXIII expression (13). In the present study, macrophages are attracted during l-NAME treatment, as indicated by the increase in expression of CD68. Concomitantly, we found a significant increase in FXIII expression after 3 days of l-NAME treatment in both strains of mice. However, since we found no differential expression between both strains, we cannot relate the mRNA expression of FXIII to the difference in blood pressure and remodeling between C57BL/6 and BALB/c mice. It should, however, be noted here that we analyzed the inflammatory markers only at the mRNA level. For instance, the activity of FXIII is nitric oxide dependent (5), and therefore differences at the protein level may exist that have not been appreciated.
The NK gene complex.
One of the key observations in the current work is the difference in the expression of CD11b between C57BL/6 and BALB/c mice. The CD11b (or αM) subunit is part of the CD11b/CD18 integrin, which is also known as the macrophage-1 antigen. This integrin is expressed on the surface of macrophages, monocytes, granulocytes, and NK cells. The CD11b/CD18 integrin binds to ligands such as ICAM-1 and plays a pivotal role in phagocytosis, cell-mediated killing, chemotaxis, and cellular activation (25). Since we found that most CD11b-positive cells also stain positively for NK1.1, the majority of these cells represent NK cells. NK cells are known to play a role in atherosclerosis (28), arteriogenesis (26), and outward remodeling of uterine vessels during pregnancy (11). We therefore further explored the role of the NK gene complex that encodes many of the NK cell receptors (30) by using the congenic BALB.B6-Cmv1r strain. With this approach, we eliminated other differences in the genetic background between C57BL/6 and BALB/c mice, which may obscure a comparison between these strains. In particular, we found a difference in vascular design (number of collaterals) and endothelial function, as indicated by the response to methacholine. In addition, other differences may obscure a comparison between C57BL/6 and BALB/c mice, such as a possible difference in the maturity of the immune system of these relatively young mice.
The NK gene complex encodes for many NK cell receptors (30) but also other genes of which the LOX-1 gene, or oxidized LDL receptor, has been related to hypertension (14) and endothelial function (29). Our results indicate that with the NK gene complex of the C57BL/6 strain, BALB.B6-Cmv1r mice acquired a similar sensitivity to develop hypertension and vascular remodeling as C57BL/6 mice. Interestingly, the BALB.B6-Cmv1r strain did not show the typical vascular architecture of the C57BL/6 strain, since the number of collaterals was similar to the BALB/c strain. Thus, while vascular design may be considered as a result of continuous vascular remodeling during development, this remarkable difference between C57BL/6 and BALB/c mice appears not to relate to the NK gene complex.
We used the l-arginine analog l-NAME to induce hypertension in mice. Nitric oxide inhibition with chronic l-NAME treatment induces vasoconstriction (8), increased peripheral resistance (20), and the subsequently, elevated blood pressure. We used a relatively high dose of l-NAME, which could result in tissue damage, particularly in the kidney. To rule out the potential interference from l-NAME overdosage, we also tested a lower dose of l-NAME (0.5 g/l). This showed that in C57BL/6 mice, l-NAME still induced a small increase in blood pressure, whereas BALB/c mice again did not respond to l-NAME treatment.
A potential drawback of the current study is that nitric oxide inhibition could not only reduce endothelial nitric oxide production but also affect inducible and neuronal nitric oxide synthases since l-NAME is a nonspecific inhibitor of nitric oxide synthases. In addition, it could affect the immune system directly and indirectly through the modulation of proinflammatory factors such as angiotensin II. These concerns make difficult a direct extrapolation of the results to human essential hypertension.
While telemetry is considered the gold standard for blood pressure measurements in laboratory animals, we relied on invasive blood pressure recordings under anesthesia. These measurements were made before the euthanizing of the animals and subsequent vessel isolation. Therefore, we obtained information regarding blood pressure at the start, after 3 days of treatment, and at the end of the experimental period. The blood pressure load is therefore not known for the entire study period.
The absence of persistent hypertension in BALB/c mice with l-NAME treatment was paralleled by a lack of vascular remodeling, and hypertension in C57BL/6 and BALB.B6-Cmv1r mice was paralleled by clear vascular remodeling. We only determined remodeling at the level of mesenteric resistance arteries. Whether remodeling extends to other vascular beds is currently unknown, since we did not determine the total peripheral resistance. Furthermore, the relationship between blood pressure and remodeling is a chicken-and-egg phenomenon, in which the current data do not allow us to discriminate whether blood pressure remains elevated because of the remodeling or whether remodeling is present because of the persistent increase in blood pressure. With respect to vascular remodeling, we concentrated on changes in the wall-to-lumen ratio. This is because mesenteric arteries differ in size and therefore a comparison of lumen diameter and wall mass is obscured by a large anatomical variation. The wall-to-lumen ratio eliminates this source of variation and also reflects a meaningful parameter since it is part of the equation, based on the Laplace relationship, that determines the wall stress. Thus wall stress equals pressure times (radius/wall thickness).
Summary and conclusions.
The present study provides new insight into the relationship between the immune background and the sensitivity to develop hypertension induced by nitric oxide inhibition. In particular, C57BL/6 mice are more susceptible to hypertension and vascular remodeling than BALB/c mice. The low susceptibility of BALB/c mice can be overcome by the introduction of the NK gene complex of the C57BL/6 strain into the BALB/c background, such that hypertension and small artery remodeling develop in a similar manner as in C57BL/6 mice. While these findings identify the NK gene complex as a crucial determinant in the susceptibility for hypertension, we do not know which particular gene or combination of genes fulfills this role. Since the C57BL/6 strain is also known to be susceptible to atherosclerosis, it would be of interest to address the role of the NK gene complex in atherosclerotic lesion development in future experiments. This could identify a common ground in the susceptibility for hypertension and atherosclerosis. The role of the NK gene complex in human hypertension and atherosclerosis remains to be established. However, with the consideration that hypertensive disorders and atherosclerosis are strongly correlated, the possible involvement of the NK gene complex in human cardiovascular disease deserves further attention. Although further work is clearly needed, these results of the present study may help to develop anti-inflammatory strategies in cardiovascular disease.
This work was supported by The Netherlands Heart Foundation Grant 2001T038 (to E. N. T. P. Bakker and H. L. Matlung) and by the Organisatie voor Toegepast Natuurwetenschappelijk Onderzoek-Leiden University Medical Center-Vrije Universiteit Medical Center tripartite angiogenesis program, the EU European Vascular Genomics Network Grant LSHM-CT-2003-503254, and Dutch Programme Tissue Engineering Grant VGT6747 (to L. Seghers and P. H. A. Quax).
- Copyright © 2010 the American Physiological Society