Skip to main content
NCBI home page
Journal of Histochemistry and Cytochemistry logoLink to Journal of Histochemistry and Cytochemistry
. 2008 Apr;56(4):359–370. doi: 10.1369/jhc.7A7324.2007

Time Courses of Growth and Remodeling of Porcine Aortic Media During Hypertension: A Quantitative Immunohistochemical Examination

Jin-Jia Hu 1, Andy Ambrus 1, Theresa W Fossum 1, Matthew W Miller 1, Jay D Humphrey 1, Emily Wilson 1
PMCID: PMC2326104  PMID: 18071063

Abstract

Arteries undergo marked structural and functional changes in human and experimental hypertension that generally involve smooth muscle cell (SMC) hypertrophy/hyperplasia as well as abnormal extracellular matrix turnover. In this study we examined time courses of changes in SMC activity and matrix protein content in a novel mini-pig aortic coarctation model. Cell proliferation was evaluated by immunostaining of Ki-67, apoptosis was assessed by TUNEL, and phenotypic changes were monitored by immunostaining three SMC contractile markers (caldesmon, calponin, and smoothelin). Changes in medial collagen and elastin were examined by picrosirius red and Verhoeff–van Gieson staining, respectively. LabVIEW-based image analysis routines were developed to objectively and efficiently quantify the (immuno)histochemical results. We found that significant cell proliferation and matrix production occurred in the early stages of this coarctation model and then declined gradually; the SMCs also tended to exhibit a less contractile phenotype following these cellular and extracellular changes. Specifically, different aspects of the phenotypic changes associated with hypertension occurred at different rates: cell proliferation and collagen production occurred early and peaked by 2 weeks, whereas changes in contractile protein expression continued to decrease over the entire 8-week study period. Temporal changes found in this study emphasize the importance of simultaneously tracing time courses of SMC growth and differentiation as well as matrix protein production and content. SMCs are multifunctional, and caution must be used to not overdefine phenotype. This manuscript contains online supplemental material at http://www.jhc.org. Please visit this article online to view these materials. (J Histochem Cytochem 56:359–370, 2008)

Keywords: vascular smooth muscle cell, phenotypic modulation, aortic coarctation, hypertension, quantitative immunohistochemistry

Arterial cells (endothelial, smooth muscle, and fibroblasts) respond to sustained changes in mechanical loading via altered gene expression (Gibbons and Dzau 1994; Davies 1995; Langille 1996; Lehoux and Tedgui 1998). Recall, therefore, that the relationship between changes in blood pressure and mean circumferential stress (σ) in the arterial wall is given by σ = Pr/h, where P is the pressure, r the pressurized inner radius, and h the pressurized wall thickness (Humphrey 2002). It is thought that biological systems adapt to maintain the native stresses of that tissue. In the case of increased blood pressure, the arterial wall adapts to normalize the circumferential stresses that are altered by changes in pressure. Whether blood pressure increases gradually (human hypertension) or abruptly (in some animal models of hypertension), wall thickness tends to increase over time in arteries until it returns circumferential stress toward baseline (Wolinsky 1970; Matsumoto and Hayashi 1994; Hu et al. 2007). This process generally involves an increased deposition of extracellular matrix (ECM) proteins as well as hyperplasia/hypertrophy of smooth muscle cells (SMCs). To better understand the process of vascular adaptation/maladaptation in hypertension, there is a need to quantify the time courses of changes in both gene expression by SMCs and wall composition during the development of hypertension.

We used a porcine aortic coarctation model of hypertension (described in Materials and Methods) and quantified changes of key participants (fibrillar collagen, elastin, and SMCs) in aortic medial growth and remodeling at 2, 4, 6, and 8 weeks of development of hypertension. Medial SMC proliferation was examined by immunostaining for Ki-67, whereas apoptosis was evaluated by a TUNEL (terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling) assay. Total cell number was confirmed in standard hematoxylin and eosin (H&E) sections. Phenotypic modulation of smooth muscle was monitored by immunostaining for three standard SMC differentiation markers (Owens 1995): caldesmon, calponin, and smoothelin. Caldesmon (CLD) is a 120 kDa actin- and calmodulin-binding protein that has been detected in SMCs and various non-muscle cells (Sobue and Sellers 1991). Calponin (CLP) is a 32-kDa thin myofibrillar filament that acts as an actin-binding protein and is implicated in the auxiliary regulatory role of SMC contraction. CLP expression is restricted to SMCs and has been shown to be a marker of the contractile phenotype of developing smooth muscle (Gimona et al. 1990). Smoothelin (SML) is a 110-kDa cytoskeletal protein exhibiting a filamentous organization and is uniquely expressed in contractile phenotype of SMC (van der Loop et al. 1997). SML shows up very late during differentiation compared with CLD and CLP (Owens 1995); it does not belong to any particular class of structural proteins but appears to be a valuable marker for assessing SMC differentiation. When considered together, expression of these proteins is considered a more sensitive indicator of SMC differentiation than is the common use of α-actin alone. Collagen and elastin were stained with picrosirius red (PSR) and Verhoeff–van Gieson (VvG) and examined under polarized and normal light, respectively. New collagen synthesis was determined by staining with an antibody that detected the uncleaved propeptide of procollagen (I). To quantify both statistical differences and apparent time courses, we developed new image analysis programs using LabVIEW and its Vision module. Three types of quantification were performed: particle counting, staining percentage, and staining intensity. LabVIEW allows image processing (e.g., color thresholding and segmentation) and analysis (e.g., particle analysis and cumulative signal strength measurement) to be integrated within a single program, thus rendering overall analysis more effective and time efficient. Complex changes in cell and matrix biology were found to occur at different rates and extents during the early development of hypertension, thus emphasizing the importance of quantifying time courses.

Materials and Methods

Animal Model

Our aortic coarctation model for studying hypertension in the mini-pig is described elsewhere (Fossum et al. 2003). Briefly, a balloon-expandable occluder is prefilled with a 50% dextrose solution, placed over a Gore-Tex soft tissue patch sheet, and secured with sutures around the aorta proximal to the diaphragm. The occluder is connected via stiff tubing to a vascular access port that is placed SC in the neck, which allows the occluder to be gradually inflated or deflated in a conscious animal. An indwelling pressure transducer is placed within the internal thoracic artery or right carotid artery and connected to an implanted telemetry unit. Arterial pressure and heart rate were recorded for 30 sec every 2 hr throughout each study period. Mean arterial pressure is defined as diastolic pressure plus one third the difference between systolic and diastolic pressure, and a daily average mean arterial pressure was used to track the condition of each animal.

The aorta was coarcted ∼1 week after surgery by adding small amounts of dextrose to the occluder over a 7- to 10-day period until the mean arterial pressure reached or exceeded 150 mmHg. Data were collected from 30 mature (7- to 16-month old) male mini-pigs: eight normotensive (NT) controls and 22 hypertensive (HT) animals. Specifically, aortas were harvested from true controls (n=4) without an occluder, from NT animals at 2 (n=3) and 4 (n=1) weeks following a sham surgery where an occluder was implanted but not inflated, and from hypertensive animals at 2 (n=6), 4 (n=6), 6 (n=4), and 8 (n=6) weeks after the animal reached its target mean arterial pressure (>150 mmHg). All animal care and use practices were approved by the University Laboratory Animal Care Committee at Texas A&M University.

Histology and Immunohistochemistry

Segments of aorta 3+ cm proximal to the occluder were dissected free of perivascular tissue and either fixed intact by immersion in 4% fresh paraformaldehyde for 1 hr or cut into sectors to release the residual stress prior to fixation (cf. Humphrey 2002). Tissues for positive control (e.g., lymph node, skin, and intestine from NT pigs) were handled and processed similarly. Lymph node was used as a positive control for TUNEL and Ki-67 because of high apoptosis and proliferation of immune cells at this site, skin was used as a positive control for immunostaining of procollagen (I) because it is a known site of collagen synthesis, and intestine was used as positive control for some of the smooth muscle differentiation markers because of its known expression.

Paraffin-embedded tissues were sectioned at 5 μm. Prior to staining, sections were deparaffinized and rehydrated. Sections were stained with H&E for general morphology, VvG for elastin (Bancroft and Gamble 2002), and PSR for fibrillar collagen (Junqueira et al. 1979) or immunostained for a panel of antibodies. Table 1 lists the antibodies examined as well as the optimal procedure for each antibody. Optimal procedures for antigen retrieval were determined by conducting both heat-induced epitope retrieval, which included trials of three retrieval solutions (citrate buffer, pH 6, EDTA, pH 8, and EDTA with urea, pH 8) and two heating procedures (microwave or pressure cooker) and enzyme digestion, which included incubation in trypsin, pepsin, or pronase at room temperature or 37C for different times. All incubations were performed in a humidified chamber at room temperature after antigen retrieval. Endogenous peroxidase was blocked with 3% hydrogen peroxide for 20 min. Slides were then incubated in blocking serum for 30 min to prevent nonspecific binding of antibodies. Slides were incubated with the primary antibody for 1 hr, washed, and incubated in a biotin-free, polymerized detection system (ImmPRESS reagent; Vector Laboratories, Burlingame, CA) for 30 min. Slides were then incubated with DAB substrate (Vector Laboratories) in the dark (for the time period indicated for specific antibodies in Table 1). Finally, slides were counterstained with hematoxylin, dehydrated, and coverslipped with Permount (Fisher Scientific; Pittsburgh, PA). To ensure standardization and reproducibility, sections were cut at the same thickness and stained in the same batch.

Table 1.

Primary antibodies used for immunohistochemistry, their vendors, and pretreatments required to unmask the epitopes as well as optimal dilutions and DAB incubation time

Antibody (clone) Source Pretreatment Dilution DAB incubation (min)
SMC differentiation markers
 SM α-actin (1A4) Dako 1 mM EDTA, pH 8, MW 30%, 5 min*2 1:500 3
 Calponin (CALP) Dako 1 mM EDTA, pH 8, MW 30%, 5 min*2 1:500 5
 Caldesmon (N5/22) Chemicon 1 mM EDTA, pH 8, MW 30%, 5 min*2 1:2500 3
 Smoothelin (R4A) Abcam 1 mM EDTA, pH 8, MW 30%, 5 min*2 1:50 5
Cell proliferation
 Ki-67 (MIB-1) Dako 1 mM EDTA, pH 8, MW 30%, 5 min*2 1:200 4
Collagen synthesis
 Carboxyl-terminal propeptide of type I collagen (SP1.D8) DSHB Trypsin, 20 min, room temperature 1:300 5
Open in a new tab

Dako, Carpinteria, CA; Chemicon, Temecula, CA; Abcam, Cambridge, MA; DSHB, Developmental Studies Hybridoma Bank, University of Iowa.

Image Acquisition and Quantification

The central medium was imaged using an Olympus BX51 light microscope customized for circularly polarized illumination (Hu et al. 2007) coupled with an Olympus DP70 digital camera (Olympus; Tokyo, Japan). Settings for microscopic illumination, exposure times, and color balance of the digital camera were maintained for each set of slides to enable subsequent comparisons. A 25% density filter avoided saturation problems when PSR-stained sections were imaged with circularly polarized light. For quantitative analysis, five representative 24-bit color images of 1360 × 1024 pixel resolution were acquired per specimen with a ×20 objective; those shown in the following figures were acquired at ×40 for better visualization. Image files were saved in tagged-image file format (TIFF).

A LabVIEW routine containing a sequence of functions including color thresholds, fill hole, particle filter, reject border, danielsson, watershed transform, and particle analysis was developed and used to count SMC nuclei. Specifically, three-dimensional (3D) red, green, blue (RGB) thresholding (Russ 2002) was used to segment all nuclei (R: 0–255, G: 0–185, B: 0–255) and DAB-stained nuclei (R: 75–255, G: 0–255, B: 0–135). Particle analysis also generated information such as the area, hydraulic radius, orientation, and elongation of a nucleus, which can be used to estimate the 3D orientation of an individual cell (Walmsley and Canham 1979). Figure 1A shows the front panel of a representative LabVIEW VI with a demonstration (see also Supplemental Materials SM1 for the actual LabView VI).

Figure 1.

Figure 1

Open in a new tab

(A–D) Front panels or user interfaces for LabVIEW routines developed and used in this study (see Supplemental Materials for the actual VIs). (A) Three-dimensional (3D) RGB thresholding is used to segment proliferating cells (DAB stained) and total cells, thus providing a Ki-67 index. (B) 3D HSL thresholding is used to segment black elastin in Verhoeff–van Gieson sections and to calculate the relative elastin content. (C) Specific DAB-stained regions are identified as foreground by 3D HSL thresholding, whereas nonspecific background is segmented further by an additional 3D HSL thresholding that removes DAB- and hematoxylin-stained regions. Staining intensity is estimated by measuring differences between the normalized energy from foreground and background regions (Matkowskyj et al. 2000). (D) Picrosirius red-stained sections are analyzed under normal light (brightfield) and circularly polarized light (darkfield). The red region, segmented by 3D HSL thresholding in the brightfield image (region of interest, ROI) was used as a mask to exclude birefringence from muscle on the darkfield images. Relative collagen content and intensity of birefringence were evaluated using the processed image.

Elastin, stained black with VvG, was segmented by another LabVIEW routine based on 3D HSL (hue, saturation, lightness) thresholding (H: 160–240, S: 120–255, L: 0–50) (Goto et al. 1992; Kohlberger et al. 1999). Thresholds in saturation and luminance efficiently excluded lightly stained black nuclei. Relative elastin content was evaluated by calculating the segmented area as a percentage of the total area (Figure 1B) (see Supplemental Materials SM2 for the LabView VI). Another 3D HSL thresholding was used to segment specific DAB-stained regions (H: 0–30, S: 12–255, L: 0–200) from nonspecific background stained regions (H: 150–30, S: 12–255, L: 0–200, followed by an inversion). This segmentation generates a region of interest, which if used as a mask permits selective processing and analysis of the original image. Figure 1C shows the front panel of the VI with an example (see Supplemental Materials SM3 for the actual VI).

Here we present staining intensity in terms of a normalized “energy” E that is defined as follows (Matkowskyj et al. 2000). In an N1 × N2 pixel image,

graphic file with name M1.gif 1

where Ii(n1,n2) is the energy contribution from color i = R, G, or B at position (n1,n2) of an image (n1=1-N1, n2=1-N2 pixels). For brightfield images (e.g., standard transmitted light), Ii(n1,n2) = 255-bi(n1,n2), noting that a brightness bi(n1,n2) = 0 corresponds to black, and a brightness of 255 corresponds to white; for darkfield images (e.g., birefringence or fluorescence), Ii(n1,n2) = bi(n1,n2).

After segmentation of foreground and background regions, the brightness of R, G, and B at each pixel in each region was extracted to calculate the corresponding normalized energy using Equation (1). A normalized energy contributed by a specific stain is

graphic file with name M2.gif 2

where subscripts s, f, and b stand for specific, foreground, and background staining, respectively. The stained percentage, defined as the area ratio of the foreground to the total image, Af/Atotal, was also determined. Relative SMC content was determined by the stained percentage of SM α-actin.

Another LabVIEW routine, based on Whittaker's method (Whittaker et al. 1994), was developed to measure relative collagen content and its normalized energy (Figure 1D) (see Supplemental Materials SM4 for the actual LabView VI). Briefly, red regions, segmented by 3D HSL thresholding (H: 130–0, S: 80–255, L: 0–255), in the brightfield image were used as a mask to exclude birefringence from muscle on darkfield images. Relative collagen content was defined as the birefringent region divided by the total region. Similarly, the intensity of birefringence was represented by the normalized energy. If we let REC be relative elastin content, RCC be relative collagen content, and RSC be relative SMC content, then relative ground substance content = 1-REC-RCC-RSC, assuming that elastin, collagen, SMCs, and ground substance (i.e., the amorphous material consisting primarily of proteoglycans) that is not visible in standard histological sections (Fawcett 1986) are the four primary structural components of the aortic media. Note that inter- or intralaboratory variations in (immuno)histological staining may affect these thresholds; those that reach best separation should be determined empirically.

Statistical Analysis

Mean staining parameters were calculated for each animal based on five representative images. Differences between groups (e.g., NT and 2, 4, 6, or 8 weeks of HT) were assessed via one-way ANOVA in conjunction with Holm–Sidak posthoc testing, with an overall significance level p=0.05. Results are reported as mean ± SE.

Results

The 24-hr average mean arterial pressures in the HT animals (151 ± 3 mmHg at 2, 174 ± 4 mmHg at 4, 172 ± 8 mmHg at 6, and 181 ± 8 mmHg at 8 weeks) were significantly different from those in the NT animals (131 ± 2 mmHg) at each end-point. Notwithstanding the increase in pressure over time during hypertension, the overall averaged mean arterial pressures were 131 ± 2 mmHg in NT and 169 ± 4 mmHg in HT, which were also significantly different (p<0.001). The average pulse pressures in HT at 2 (49 ± 2 mmHg), 4 (58 ± 5 mmHg), and 8 (55 ± 5 mmHg) weeks differed significantly from those in NT animals (36 ± 3 mmHg); if grouped separately, the averaged pulse pressures were 36 ± 3 mmHg in NT and 54 ± 2 mmHg in HT, which were also significantly different (p<0.001).

The Ki-67 antigen is present during all active phases of the cell cycle (G1, S, G2, and M) but absent in resting cells (G0); it is thus an excellent marker for cell proliferation. Ki-67 stained nuclei appeared to be distributed evenly throughout the central media in all samples (cf. Figures 2A and 2B). We found an increased Ki-67 index in the central media at 2, 4, and 6 weeks of HT although only 2 weeks reached statistical difference when compared with NT controls; similarly, the number of total nuclei per sectional area reached statistical significance only at 2 weeks of HT, thus suggesting that heightened activity was particularly strong early on (Figure 2C). Future studies should thus examine time points earlier than 2 weeks during the development of hypertension.

Figure 2.

Figure 2

Open in a new tab

(A,B) Representative Ki-67 immunohistochemical staining of porcine aorta from a control and a hypertensive animal, respectively, are shown. (C) Ki-67 index obtained from image analysis with statistical analysis as a function of the time of hypertension is shown. **Denotes significant difference from control at p<0.005. Bar = 50 μm.

TUNEL assay, which demonstrates in situ DNA fragmentation, a hallmark for apoptosis, showed strongly positive reactions on both intimal cells and fibroblasts in the adventitia in the early stage of HT (data not shown), but no marked staining in the media over the 8-week study period.

Soon after release of procollagen (I) into the extracellular space, the amino propeptides and then the carboxyl propeptides are cleaved (Davidson et al. 1975). Antibodies to the amino- or carboxyl-terminus of procollagen (I), which only bind to intracellular procollagen (I), can thus be used to examine new collagen (I) production and its localization (Botney et al. 1993). Staining intensity of procollagen (I) was increased relative to control in all HT specimens (Figures 3A and 3B) but was highest at 2 weeks (Figure 3C).

Figure 3.

Figure 3

Open in a new tab

(A,B) Representative procollagen (I) immunohistochemical staining of porcine aorta from a control and a hypertensive animal are shown, respectively. Inset in B demonstrates a positive control in which fibroblasts in the porcine skin are stained. (C) Staining intensity of procollagen (I) obtained from image analysis with statistical analysis is shown. Error bars in C are SEM. *Denotes significant difference from control at p<0.05. **Denotes difference at p<0.005. Bar = 50 μm.

Viewed under circularly polarized light, PSR-stained sections demonstrated the lamellar structure of medial collagen. Computer-assisted morphological analysis showed that relative collagen content increased in HT animals at 2 weeks, stayed high at 4 weeks, but declined toward baseline at 6 weeks. On the other hand, normalized energy for fibrillar collagen, which represents collagen density per sectional area, peaked at 2 weeks of HT (p<0.05) and then returned to the normal value at 4 weeks and remained so thereafter (Figure 4A).

Figure 4.

Figure 4

Open in a new tab

Relative contents (mean ± SEM) obtained from image analysis with statistical comparison. (A) Collagen. (B) Elastin. (C) Smooth muscle cell (SMC). (D) Bulk undetermined matrix. *Denotes significant difference from control at p<0.05. **Denotes difference at p<0.005.

Analysis also showed that the relative elastin content in HT decreased significantly (p<0.05) at 2 and 4 weeks of HT but remained at a similar level thereafter with no statistical difference from NT (Figure 4B). As the ratio of collagen and elastin is often used as an index of arterial stiffening, note that this ratio increased at 2 and 4 weeks of HT (p<0.005) but returned to baseline after 6 weeks: 0.75 ± 0.09 for NT and 1.27 ± 0.12, 1.25 ± 0.11, 0.80 ± 0.20, and 0.83 ± 0.08 at 2, 4, 6, and 8 weeks of HT (cf. Olivetti et al. 1982; Levy et al. 1994). There was no significant difference among NT and each HT for relative SMC content or the bulk ECM represented by components other than elastin or collagen (Figures 4C and 4D).

Despite no statistical difference in staining intensity of SM α-actin among groups, the medial SMCs appeared to modulate their phenotype from contractile to more synthetic in HT in terms of expression of caldesmon (CLD), calponin (CLP), and smoothelin (SML) (Figure 5). Posthoc multiple comparisons to controls showed that staining intensity of CLD reached statistical significance at 4 weeks of HT (p<0.005) and remained at 6 (p<0.05) and 8 (p<0.005) weeks of HT (Figure 6A). Staining intensities of CLP did not differ statistically between NT and HT, but temporal changes appeared similar to that of CLD (Figure 6B). Temporal changes in the staining intensity of SML were slightly different from that of CLD and CLP; it decreased as hypertension advanced but was not significantly different from the control until 8 weeks of HT (Figure 6C). Note that there was no significant difference between NT and HT in terms of stained percentage for the three SMC differentiation markers.

Figure 5.

Figure 5

Open in a new tab

Micrographs of immunohistochemical staining of porcine aorta with anti-CLD (A,D,G,J,M), anti-CLP (B,E,H,K,N), and anti-SML (C,F,I,L,O). (A–C) From the control animals. (D–F) From 2 weeks of HT. (G–I) From 4 weeks of HT. (J–L) From 6 weeks of HT. (M–O) From 8 weeks of HT. Bar = 50 μm.

Figure 6.

Figure 6

Open in a new tab

Time course of staining intensity (mean ± SEM) of SMC contractile markers obtained from image analysis. (A) Caldesmon. (B) Calponin. (C) Smoothelin. *Denotes significant difference from control at p<0.05. **Denotes difference at p<0.005.

Discussion

Diverse growth responses by SMCs and deposition patterns of ECM proteins have been reported separately for the many different vessels and durations of hypertension that have been considered via different animal models (e.g., Gabbiani et al. 1979; Olivetti et al. 1982; Owens and Reidy 1985; Baumbach and Hajdu 1993; Pauletto et al. 1994). As a result, it has been difficult to mathematically model time courses of changes in cellular and matrix activity, which is essential for understanding how the arterial wall responds to hypertension (Rachev et al. 1998; Gleason and Humphrey 2004). One goal of this work, therefore, was to quantify together the time courses of cellular and matrix changes in the aorta of a single model of hypertension over an extended (8 week) period. The porcine aorta is preferred over rodent aorta because the thicker media allows better quantification of (immuno-)histological findings, particularly for cell proliferation or apoptosis. That is, whereas mouse and rat thoracic aorta consist of only 5–7 and 6–8 musculoelastic layers, respectively, the porcine thoracic aorta consists of 60–70 layers that delimit the many medial smooth muscle cell layers (Wolinsky and Glagov 1969). The porcine vasculature is also thought to be closer to that of humans than the rodent vasculature (Johnson et al. 1999).

We found increased cell proliferation at 2, 4, and 6 weeks, but not at 8 weeks, of HT compared with NT controls, reaching statistical significance at 2 weeks of HT. Accordingly, cell number per sectional (unit) area also peaked at 2 weeks of HT. Altered cell numbers in cross-sectional images may have been due to hyperplasia or changes in orientation (e.g., cells becoming more longitudinal and less circumferential), but the former is more likely because the percentage of round cells, determined by quantifying elongation via the particle analysis, did not change at 2 weeks of HT (data not shown). Starting at 4 weeks of HT, there was no increase in cell number per sectional area, whereas cells kept proliferating without strong indication of cell death. This suggested an increase in absolute volume occupied by SMCs. Indeed, gross morphological analyses revealed an increase in wall thickness at 2 and 4 weeks of HT, which may have been due to both cell proliferation and matrix production. Our overall findings are thus consistent with results from Xu et al. (2001) who found that proliferation of medial SMCs was induced rapidly (within 3 days) after acute coarctation of the rat aorta and continued for 2 weeks, whereas no increased apoptosis was found until 4 weeks. We did not observe obvious staining of TUNEL in the media throughout the 8-week study. This suggests that both proliferation and apoptosis were delayed in our porcine model in which the coarctation was induced gradually over a 7- to 10-day period, rather than acutely as in most rodent models.

Changes in absolute and relative quantities of collagen and elastin typically reflect changes in passive mechanical properties (Wolinsky and Glagov 1964), provided there are no major changes in orientation or crosslinking of these fibers. An abnormal turnover of elastin and collagen can thus lead to a maladaptive response, which in turn can contribute to the progression of the disease. Relative collagen content was increased at 2 weeks of HT, stayed high at 4 weeks, and declined to baseline level at 6 weeks, although the normalized energy (Equation 1) was significantly higher only at 2 weeks of HT. Although these findings correlated with the expression of procollagen, which peaked at or before 2 weeks of HT and remained thereafter, they suggested that collagen degradation may have started by 4 weeks of HT. Although we did not measure imbalances in matrix metalloproteinase–tissue inhibitor of metalloproteinases (MMP-TIMP) activity, a shorter half-life of arterial collagen has been reported in hypertension (e.g., decrease from ∼70 days in NT to ∼17 days in HT rat aorta) (Nissen et al. 1978). Hence, the decrease of relative collagen content at 6 and 8 weeks of HT may have been due to enhanced collagen degradation, similar to time courses of MMP activity in arterial injury models (Strauss et al. 1996). Using a similar rabbit aortic coarctation model (Xu et al. 2000), it was found that collagen (I) increased at 3 days, remained high at 1, 2, and 4 weeks, and decreased toward normal by 8 weeks, again consistent with our results.

The relative elastin content in the HT animals decreased at 2 weeks and remained at a similar level thereafter. There are contradictory reports regarding changes in arterial elastin due to hypertension, which may be due to the use of different animal models, methods of analysis (chemical vs morphometric approaches), or definitions of content (e.g., relative content can change while absolute content remains unchanged). Wolinsky (1970) chemically assessed collagen and elastin content in the aorta of DOCA-salt hypertensive rats and reported that neither protein was significantly different between NT and HT. In contrast, a morphological study of the aorta of spontaneously hypertensive rats showed an increased collagen content in the media but showed similar absolute elastin content compared with controls (Levy et al. 1994). Our findings are not consistent with these results.

Phenotypic modulation of SMCs, manifested as increased rates of proliferation, migration, and synthesis of ECM proteins and decreased expression of SMC-specific/-selective marker genes, has been reported in atherosclerosis and hypertension (Owens et al. 2004). In addition to SM α-actin, we examined expression of three primary SMC-specific markers by immunohistochemistry. Staining intensities of CLD, CLP, and SML were all decreased at 4 weeks of HT, with similar temporal changes in CLD and CLP but not SML. This may be due to the distinct promoters involved in their gene transcription (Yoshida et al. 2004). Different staining patterns of SMC differentiation markers also demonstrated that the aortic medial SMCs may represent a spectrum of SMC phenotypes.

Most studies of phenotypic modulation of vascular SMCs have focused on neointimal development associated with atherosclerosis or restenosis. Very few have examined the time course of phenotypic changes in the media of large arteries. Regan et al. (2000) reported decreased immunostaining for SM α-actin and SM myosin heavy chain 7 days after injuring the carotid artery with a wire, with slightly increased staining for both markers by 14 days (see their Figure 2). If one focuses on these markers in the media, however, staining appears to decrease further than that of 7 days after injury. Note that they only qualitatively describe their immunostaining. It is of interest that at 2 weeks of HT when vigorous cell proliferation and collagen (I) production were observed, there was no significant difference between HT and NT in terms of SMC differentiation marker staining intensities. This suggested that the medial SMCs expressing contractile proteins could also proliferate and actively synthesize ECM proteins. Indeed, this was found by double staining at 2 weeks of HT (data not shown), which showed coexistent staining of Ki-67 and SMC differentiation markers in the same SMCs. Owens et al. (1986) similarly found that loss of SM α-actin was not a prerequisite for initiation of cell proliferation in cultured rat aortic SMCs. Furthermore, the medial SMCs continued to modulate their phenotype as determined by decreased expression of the SMC differentiation markers after 2 weeks of hypertension, whereas cell proliferation and collagen production did not correlate with this phenotypic change, which contrasts with the accepted phenotypic changes of SMCs in neointimal development (Owens 1995; Thyberg 1998).

Drawbacks of visual scoring of immunostaining include inter- and intra-observer bias and difficulty of statistical analysis due to a lack of consistent grading. In this study, LabVIEW-based image analyses provided a consistent description of staining patterns [e.g., SMC differentiation markers and procollagen (I)] and allowed quantitative comparison between HT and NT. It was challenging, if not impossible, in pilot studies to infer results from immunostaining without computer-assisted image analysis.

Segmentation of regions of interest (e.g., DAB-stained brown) from background has a significant impact on quantification of staining parameters. In general, two types of color segmentation are generally used based on dependencies on background color: (a) use of information on the foreground color only (e.g., blue channel, combined HSL, and combined RGB) (Russ 2002) and (b) use of color contrast between foreground and background (e.g., G/B) (Montironi et al. 1996), normalized colors (Ohta et al. 1980; Brey et al. 2003), and spectral ratio (Ornberg et al. 1999). We adopted 3D color thresholds (either combined HSL or RGB) to segment regions of interest because they require no information on the background color (and thus is a better choice for multicolor immunohistochemistry or special staining) while providing the best results among the first type (Brey et al. 2003). A set of predefined HSL (or RGB) thresholds was applied for all sections, which allowed more objective selection compared with using the Magic Wand in conjunction with Similar commands (Photoshop; Adobe Systems, San Jose, CA) (Lehr et al. 1997; Matkowskyj et al. 2000). The latter may yield different staining areas when subtly different colors are selected. This is, in fact, common to all interactive thresholding procedures. Quantitative immunohistochemistry often determines the amount of antigen by calculating the mean optical density (MOD) of selected regions using Photoshop or NIH image. With appropriate dilutions of antibodies, the MOD of immunostaining region relates linearly to the stained protein level (Rahier et al. 1989); such correlations have been shown in many other studies (Venter et al. 1987; Dias et al. 2000). The MOD depends, however, on the formula converting the color to the gray-scale image [e.g., gray scale = 0.3×R + 0.59×G + 0.11×B or gray scale = B - 0.3×(R+G)]. That is, MOD is excellent for evaluating the amount of a monochromatic chromogen but not as good for a spectrum of browns. On the other hand, cumulative signal strength (Matkowskyj et al. 2000), which considers equal contributions of three colors (red, green, and blue), is a better choice when accounting for a broad range of brown DAB staining as in this study and validated experimentally elsewhere (Matkowskyj et al. 2003).

Whereas existing algorithms require a series of user inputs for each image, the object-oriented programming of LabVIEW integrates all image processing and analysis as an assembly line, making the process less labor intensive (cf. Lehr et al. 1997; Matkowskyj et al. 2000; Tolivia et al. 2006). The Vision module of LabVIEW, which contains a variety of functions for both image processing and analysis, also provides flexibility for different applications including studies of cellular and extracellular changes in cancers or cardiovascular diseases. Once the routine is developed, stand-alone executable files with user-friendly interfaces can be distributed to others who have the LabVIEW run-time engine (downloadable at no cost from www.ni.com).

In summary, there is a pressing need to predict growth and remodeling of arteries in response to altered mechanical environments and to elucidate the underlying mechanisms. Temporal changes found in this study emphasize the importance of simultaneously tracing time courses of SMC growth and differentiation as well as matrix protein production and content. Moreover, as confirmed by our data, SMCs are multifunctional and caution must be used to not overdefine phenotype. Clearly, additional data will be needed to develop a predictive mathematical model of pressure-induced growth and remodeling, and objective immunohistochemical analyses will continue to play an important role. In particular, attention should be given to the development of kinetic models, which in turn may aid in the design of new treatment strategies to prevent maladaptations in hypertension and to help provide directions for tissue engineering of vascular grafts.

Supplementary Material

[Supplemental Data for this article]
jhc_7A7324.2007_index.html (1.4KB, html)

Acknowledgments

This research was supported by Grant HL-64372 from the National Institutes of Health via the Bioengineering Research Partnerships Program.

We thank Rosemary Vollmar and Jan Patterson for expert technical assistance. The procollagen type I (SP1.D8) developed by Dr. Heinz Furthmayr was obtained from the Developmental Studies Hybridoma Bank at the University of Iowa.

References

  1. Bancroft JD, Gamble M, eds. (2002) Theory and Practice of Histological Techniques, 5th ed. New York, Churchill Livingstone
  2. Baumbach GL, Hajdu MA (1993) Mechanics and composition of cerebral arterioles in renal and spontaneously hypertensive rats. Hypertension 21:816–826 [DOI] [PubMed] [Google Scholar]
  3. Botney MD, Liptay MJ, Kaiser LR, Cooper JD, Parks WC, Mecham RP (1993) Active collagen synthesis by pulmonary arteries in human primary pulmonary hypertension. Am J Pathol 143:121–129 [PMC free article] [PubMed] [Google Scholar]
  4. Brey EM, Lalani Z, Johnston C, Wong M, McIntire LV, Duke PJ, Patrick CW (2003) Automated selection of DAB-labeled tissue for immunohistochemical quantification. J Histochem Cytochem 51:575–584 [DOI] [PubMed] [Google Scholar]
  5. Davidson JM, McEneany LSG, Bornstein P (1975) Intermediates in limited proteolytic conversion of procollagen to collagen. Biochemistry 14:5188–5194 [DOI] [PubMed] [Google Scholar]
  6. Davies PF (1995) Flow-mediated endothelial mechanotransduction. Physiol Rev 75:519–560 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Dias P, Chen B, Dilday B, Palmer H, Hosoi H, Singh S, Wu C, et al. (2000) Strong immunostaining for myogenin in rhabdomyosarcoma is significantly associated with tumors of the alveolar subclass. Am J Pathol 156:399–408 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Fawcett DW (1986) Bloom and Fawcett's A Textbook of Histology. Philadelphia, W.B. Saunders
  9. Fossum TW, Baltzer WI, Miller MW, Aguirre M, Whitlock D, Solter P, Makarski LA, et al. (2003) A novel aortic coarctation model for studying hypertension in the pig. J Invest Surg 16:35–44 [PubMed] [Google Scholar]
  10. Gabbiani G, Elemer G, Guelpa C, Vallotton MB, Badonnel MC, Huttner I (1979) Morphologic and functional changes of the aortic intima during experimental hypertension. Am J Pathol 96:399–422 [PMC free article] [PubMed] [Google Scholar]
  11. Gibbons GH, Dzau VJ (1994) The emerging concept of vascular remodeling. N Engl J Med 330:1431–1438 [DOI] [PubMed] [Google Scholar]
  12. Gimona M, Herzog M, Vandekerckhove J, Small JV (1990) Smooth muscle specific expression of calponin. FEBS Lett 274:159–162 [DOI] [PubMed] [Google Scholar]
  13. Gleason RL, Humphrey JD (2004) A mixture model of arterial growth and remodeling in hypertension: altered muscle tone and tissue turnover. J Vasc Res 41:352–363 [DOI] [PubMed] [Google Scholar]
  14. Goto M, Nagatomo Y, Hasui K, Yamanaka H, Murashima S, Sato E (1992) Chromaticity analysis of immunostained tumor specimens. Pathol Res Pract 188:433–437 [DOI] [PubMed] [Google Scholar]
  15. Hu JJ, Fossum TW, Miller MW, Xu H, Liu JC, Humphrey JD (2007) Biomechanics of the porcine basilar artery in hypertension. Ann Biomed Eng 35:19–29 [DOI] [PubMed] [Google Scholar]
  16. Humphrey JD (2002) Cardiovascular Solid Mechanics: Cells, Tissues, and Organs. New York, Springer
  17. Johnson GJ, Griggs TR, Badimon L (1999) The utility of animal models in the preclinical study of interventions to prevent human coronary artery restenosis: analysis and recommendations. Thromb Haemost 81:835–843 [PubMed] [Google Scholar]
  18. Junqueira LCU, Bignolas G, Brentani RR (1979) Picrosirius staining plus polarization microscopy, a specific method for collagen detection in tissue sections. Histochem J 11:447–455 [DOI] [PubMed] [Google Scholar]
  19. Kohlberger PD, Breitenecker F, Kaider A, Losch A, Gitsch G, Breitenecker G, Kieback DG (1999) Modified true-color computer-assisted image analysis versus subjective scoring of estrogen receptor expression in breast cancer: a comparison. Anticancer Res 19:2189–2193 [PubMed] [Google Scholar]
  20. Langille BL (1996) Arterial remodeling: relation to hemodynamics. Can J Physiol Pharmacol 74:834–841 [PubMed] [Google Scholar]
  21. Lehoux S, Tedgui A (1998) Signal transduction of mechanical stresses in the vascular wall. Hypertension 32:338–345 [DOI] [PubMed] [Google Scholar]
  22. Lehr HA, Mankoff DA, Corwin D, Santeusanio G, Gown AM (1997) Application of Photoshop-based image analysis to quantification of hormone receptor expression in breast cancer. J Histochem Cytochem 45:1559–1565 [DOI] [PubMed] [Google Scholar]
  23. Levy BI, Duriez M, Phillipe M, Poitevin P, Michel JB (1994) Effect of chronic dihydropyridine (Isradipine) on the large arterial walls of spontaneously hypertensive rats. Circulation 90:3024–3033 [DOI] [PubMed] [Google Scholar]
  24. Matkowskyj KA, Cox R, Jensen RT, Benya RV (2003) Quantitative immunohistochemistry by measuring cumulative signal strength accurately measures receptor number. J Histochem Cytochem 51:205–214 [DOI] [PubMed] [Google Scholar]
  25. Matkowskyj KA, Schonfeld D, Benya RV (2000) Quantitative immunohistochemistry by measuring cumulative signal strength using commercially available software Photoshop and Matlab. J Histochem Cytochem 48:303–311 [DOI] [PubMed] [Google Scholar]
  26. Matsumoto T, Hayashi K (1994) Mechanical and dimensional adaptation of rat aorta to hypertension. J Biomech Eng 116:278–283 [DOI] [PubMed] [Google Scholar]
  27. Montironi R, Diamanti L, Thompson D, Bartels HG, Bartels PH (1996) Analysis of the capillary architecture in the precursors of prostate cancer: recent findings and new concepts. Eur Urol 30:191–200 [DOI] [PubMed] [Google Scholar]
  28. Nissen R, Cardinale GJ, Udenfriend S (1978) Increased turnover of arterial collagen in hypertensive rats. Proc Natl Acad Sci USA 75:451–453 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Ohta Y, Kanade T, Sakai T (1980) Color information for region-segmentation. Comput Graph Image Process 13:222–241 [Google Scholar]
  30. Olivetti G, Melissari M, Marchetti G, Anversa P (1982) Quantitative structural changes of the rat thoracic aorta in early spontaneous hypertension. Tissue composition, and hypertrophy and hyperplasia of smooth muscle cells. Circ Res 51:19–26 [DOI] [PubMed] [Google Scholar]
  31. Ornberg RL, Woerner BM, Edwards DA (1999) Analysis of stained objects in histological sections by spectral imaging and differential absorption. J Histochem Cytochem 47:1307–1313 [DOI] [PubMed] [Google Scholar]
  32. Owens GK (1995) Regulation of differentiation of vascular smooth muscle cells. Physiol Rev 75:487–517 [DOI] [PubMed] [Google Scholar]
  33. Owens GK, Kumar MS, Wamhoff BR (2004) Molecular regulation of vascular smooth muscle cell differentiation in development and disease. Physiol Rev 84:767–801 [DOI] [PubMed] [Google Scholar]
  34. Owens GK, Loeb A, Gordon D, Thompson MM (1986) Expression of smooth muscle-specific α-isoactin in cultured vascular smooth muscle cells. Relationship between growth and cytodifferentiation. J Cell Biol 102:343–352 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Owens GK, Reidy MA (1985) Hyperplastic growth response of vascular smooth muscle cells following induction of acute hypertension in rats by aortic coarctation. Circ Res 57:695–705 [DOI] [PubMed] [Google Scholar]
  36. Pauletto P, Chiavegato A, Giuriato L, Scatena M, Faggin E, Grisenti A, Sarzani R, et al. (1994) Hyperplastic growth of aortic smooth muscle cells in renovascular hypertensive rabbits is characterized by the expansion of an immature cell phenotype. Circ Res 74:774–788 [DOI] [PubMed] [Google Scholar]
  37. Rachev A, Stergiopulos N, Meister JJ (1998) A model for geometric and mechanical adaptation of arteries to sustained hypertension. J Biomech Eng 120:9–17 [DOI] [PubMed] [Google Scholar]
  38. Rahier J, Stevens M, de Menten Y, Henquin JC (1989) Determination of antigen concentration in tissue sections by immunodensitometry. Lab Invest 61:357–363 [PubMed] [Google Scholar]
  39. Regan CP, Adam PJ, Madsen CS, Owens GK (2000) Molecular mechanisms of decreased smooth muscle differentiation marker expression after vascular injury. J Clin Invest 106:1139–1147 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Russ JC (2002) The Image Processing Handbook. 4th ed. Boca Raton, FL, CRC Press
  41. Sobue K, Sellers JR (1991) Caldesmon, a novel regulatory protein in smooth muscle and nonmuscle actomyosin systems. J Biol Chem 266:12115–12118 [PubMed] [Google Scholar]
  42. Strauss BH, Robinson R, Batchelor WB, Chisholm RJ, Ravi G, Natarajan MK, Logan RA, et al. (1996) In vivo collagen turnover following experimental balloon angioplasty injury and the role of matrix metalloproteinases. Circ Res 79:541–550 [DOI] [PubMed] [Google Scholar]
  43. Thyberg J (1998) Phenotypic modulation of smooth muscle cells during formation of neointimal thickenings following vascular injury. Histol Histopathol 13:871–891 [DOI] [PubMed] [Google Scholar]
  44. Tolivia J, Navarro A, del Valle E, Perez C, Ordonez C, Martinez E (2006) Application of Photoshop and Scion Image analysis to quantification of signals in histochemistry, immunocytochemistry and hybridocytochemistry. Anal Quant Cytol Histol 28:43–53 [PubMed] [Google Scholar]
  45. van der Loop FTL, Gabbiani G, Kohnen G, Ramaekers FCS, van Eys JJM (1997) Differentiation of smooth muscle cells in human blood vessels as defined by smoothelin, a novel marker for the contractile phenotype. Arterioscler Thromb Vasc Biol 17:665–671 [DOI] [PubMed] [Google Scholar]
  46. Venter DJ, Kumar S, Tuzi NL, Gullick WJ (1987) Overexpression of the c-erbB-2 oncoprotein in human breast carcinomas: immunohistological assessment correlates with gene amplification. Lancet 2:69–72 [DOI] [PubMed] [Google Scholar]
  47. Walmsley JG, Canham PB (1979) Orientation of nuclei as indicators of smooth muscle cell alignment in the cerebral artery. Blood Vessels 16:43–51 [DOI] [PubMed] [Google Scholar]
  48. Whittaker P, Kloner RA, Boughner DR, Pickering JG (1994) Quantitative assessment of myocardial collagen with picrosirius red staining and circularly polarized light. Basic Res Cardiol 89:397–410 [DOI] [PubMed] [Google Scholar]
  49. Wolinsky H (1970) Response of rat aortic media to hypertension. Morphological and chemical studies. Circ Res 26:507–522 [DOI] [PubMed] [Google Scholar]
  50. Wolinsky H, Glagov S (1964) Structural basis for static mechanical properties of aortic media. Circ Res 14:400–413 [DOI] [PubMed] [Google Scholar]
  51. Wolinsky H, Glagov S (1969) Comparison of abdominal and thoracic aortic medial structure in mammals: deviation of man from usual pattern. Circ Res 25:677–686 [DOI] [PubMed] [Google Scholar]
  52. Xu CP, Lee S, Singh TM, Sho E, Li XQ, Sho M, Masuda H, et al. (2001) Molecular mechanisms of aortic wall remodeling in response to hypertension. J Vasc Surg 33:570–578 [DOI] [PubMed] [Google Scholar]
  53. Xu CP, Zarins CK, Bassiouny HS, Briggs WH, Reardon C, Glagov S (2000) Differential transmural distribution of gene expression for collagen types I and III proximal to aortic coarctation in the rabbit. J Vasc Res 37:170–182 [DOI] [PubMed] [Google Scholar]
  54. Yoshida T, Kawai-Kowase K, Owens GK (2004) Forced expression of myocardin is not sufficient for induction of smooth muscle differentiation in multipotential embryonic cells. Arterioscler Thromb Vasc Biol 24:1596–1601 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

[Supplemental Data for this article]
jhc_7A7324.2007_index.html (1.4KB, html)
jhc_7A7324.2007_1.pdf (165.3KB, pdf)
jhc_7A7324.2007_2.pdf (141.1KB, pdf)
jhc_7A7324.2007_3.pdf (206KB, pdf)
jhc_7A7324.2007_4.pdf (186.9KB, pdf)

Articles from Journal of Histochemistry and Cytochemistry are provided here courtesy of The Histochemical Society

RESOURCES