Influence of elastin on rat small artery mechanical properties

Authors


Corresponding author S. M. Arribas: Departamento de Fisiología, Facultad de Medicina, Universidad Autónoma de Madrid, C/Arzobispo Morcillo 2, 28029-Madrid, Spain. Email: silvia.arribas@uam.es

Abstract

We have previously developed a method for estimating elastin content and organization in resistance arteries, where it is a minor component. The aim of the present study was to validate the method against a quantitative assay and to determine the relative importance of elastin content and organization for intrinsic elasticity of small arteries. Mesenteric third order branches (from 10-day-old, 1- and 6-month-old rats) and middle cerebral arteries (from 6-month-old rats) were pressurized. β-Values were calculated from stress–strain relationships and used as indicators of intrinsic stiffness. The same pressure-fixed arteries were used to estimate elastin content and organization in the internal elastic lamina with confocal microscopy. Collagen and elastin contents were determined by Picrosirius Red staining and radioimmunoassay for desmosine, respectively. Confocal and desmosine assays gave similar results: no difference in elastin content of mesenteric vessels from 1- and 6-month-old rats, and a significant reduction in cerebral compared to mesenteric arteries. For all parameters (elastin and collagen content, fenestrae area and internal elastic lamina thickness) the best correlation was found between β-values and fenestrae size. These data suggest that in small arteries: (1) confocal microscopy can be used as a method for the simultaneous study of changes in elastin content and organization; and (2) elastin organization might be a key determinant of intrinsic elastic properties.

The passive mechanical properties of arteries are mainly conferred by elastic fibres and collagen, the smooth muscle cells (SMC) being the active elements (Dobrin, 1978). Elastic fibres are extracellular matrix (ECM) assemblies of elastin, microfibrils and associated proteins (Kielty et al. 2002), which endow vessels with elasticity, whereas collagen, a much stiffer protein, has a role in limiting vessel distension.

Large arteries have a high degree of elasticity because elastin is a major wall component (Parks et al. 1993). This property is essential for cardiovascular function. By buffering pressure variations during the cardiac cycle, this permits a relatively constant blood flow and correct organ irrigation (Safar & London, 1994). By contrast, in resistance arteries elastin is a minor component and collagen accounts for the largest proportion of the ECM (Milnor, 1989). Elastin is restricted to a fenestrated internal elastic lamina (IEL) and to networks of elastic fibres in parts of the adventitia and media (Briones et al. 2003). The fact that in resistance arteries elastin is very scarce has lead to the conclusion that this protein has no major role in the mechanical properties of these vessels. This assumption has been favoured by the fact that elastin is difficult to quantify with the assays used for conduit arteries, based on harsh chemical treatment that require a large amount of tissue (Boumaza et al. 2001).

Elastin is autofluorescent in the range 488–515 nm and it can be detected in arteries after other fluorescent components have been eliminated (Wong & Langille, 1996; Briones et al. 2003). We have used this property to develop a method for estimatiing the elastin content in small vessels with confocal microscopy (Briones et al. 2003). This method is based on the linear relationship between elastin concentration and fluorescence intensity (Blomfield & Farrar, 1969). Elastin can be also estimated by measuring desmosine content, a residue formed a during cross-linking process, exclusive to elastin (Eyre, 1984). This approach can be used to quantify elastin in relation to total protein content in small samples (Starcher & Conrad, 1995; Starcher, 2001). The first objective of the present study was to validate the confocal method against this quantitative assay.

In our previous study, we also demonstrated that elastin seems to be an important determinant of passive mechanical properties in rat mesenteric resistance arteries. Thus, digestion of elastin with pancreatic elastase greatly increased intrinsic wall stiffness (Briones et al. 2003). Interestingly, in our experimental conditions, elastin was not eliminated from the vessel wall, but the fenestrated organization of the IEL was substantially disrupted. Based on these data, our second objective was to test the hypothesis that elastin organization is a central element contributing to the elastic properties of small arteries. For this study we used two types of vessels, rat middle cerebral and mesenteric resistance arteries, which have similar size but exhibit different mechanical properties.

Methods

Male Wistar Kyoto rats (10 days, 1 and 6 months old) were anaesthetized (50 mg kg−1 sodium pentobarbital i.p) and killed by bleeding. The experiments were approved by the animal care and use committees from Universidad Autónoma de Madrid and Universidad Autónoma de Bercelona, where the experiments were carried out. Rat mesenteric arteries (third order branches) were studied at all ages and middle cerebral arteries in 6-month-old rats, owing to methodological difficulties (see below).

Pressure myography

Mechanical properties were studied with a pressure myograph, as described by Briones et al. (2003). Briefly, the vessel was secured on two glass microcannulae, set to 70 mmHg and equilibrated (60 min, 37°C) in oxygenated calcium-free Krebs–Henseleit solution (pH = 7.3–7.4; 0Ca-KHS, mm: 119 NaCl, 4.7 KCl, 24 NaHCO3, 1.18 KH2PO4, 1.2 MgSO4 and 10 EGTA). The vessels were set to an internal pressure of 10 mmHg and a pressure–diameter curve was then obtained by increasing intraluminal pressure in 20 mmHg steps (from 20 to 140 mmHg), with internal and external diameters (Di0Ca and De0Ca, respectively) being measured. Finally, the artery was set to an internal pressure of 70 mmHg, pressure-fixed (60 min, 4% paraformaldehyde, 37°C) and stored for confocal microscopy.

We used β-value, a parameter directly proportional to incremental elastic modulus (Einc), as an indicator of intrinsic wall stiffness independent of vessel geometry. β-Values were calculated as follows. Wall stress (σ) and strain (ɛ) were calculated from Di0Ca and De0Ca measurements, as described by Baumbach & Heistad (1989). From the stress–strain relationship we calculated the incremental elastic modulus (Einc=δσ/δɛ; Dobrin, 1978) by fitting the stress–strain data from each animal to an exponential curve using the equation σ=σorigeβε, where σorig is the stress at the original diameter (diameter at 10 mmHg internal pressure). Taking derivatives on the above equation, Einc=βσ. For a given σ-value, Einc is directly proportional to β.

Confocal microscopy

The characteristics of IEL were studied with confocal microscopy as described by Briones et al. (2003). Briefly, intact pressure-fixed segments were visualized with a Leica TCS SP2 confocal system. Serial optical sections from different regions of the IEL (z step = 0.3 μm) were captured with a ×63 oil immersion objective (zoom 2 or 4) using the 488 nm line. Due to the small size of fenestrae in cerebral arteries and 10-day-old rat mesenteric arteries, the analysis of elastin organization required images to be captured at a higher magnification (×63 oil immersion objective, zoom 4). It was not possible to study cerebral arteries from young rats. Special care was taken in order to obtain all the images under identical conditions of laser intensity, brightness and contrast. Images of the lumen were obtained with a ×20 objective.

Metamorph image analysis was used for quantification. From each stack of serial images IEL thickness was measured and individual projections of the IEL were reconstructed. From these projections fluorescence intensity (average fluorescence intensity per pixel) was measured in several regions and averaged. An estimate of the amount of elastin in the IEL was made, taking into account fluorescence intensity and the volume occupied by elastin in a 1 mm length of vessel. Area of fenestrae was measured in binary images obtained from the IEL projections (Briones et al. 2003).

Desmosine content assay

Elastin content was determined by radioimmunoassay for desmosine as described by Starcher & Conrad (1995) and Starcher (2001). Briefly, fixed vessels were hydrolysed in 6 n HCl at 100°C for 24 h, evaporated to dryness, redissolved in water, vortexed, microfuged and assayed for desmosine and for total protein using a ninhydrin assay.

Collagen determination using Picrosirius Red

Collagen was determined in rings with Picrosirius Red as described by Izzard et al. (2003). Colour images were captured with a fluorescence microscope using a digital camera. The background and red-stained areas were measured in the adventitia and media with Metamorph software. Collagen contents were estimated as the difference between the background and the red-stained areas.

Statistical analysis

Results are expressed as means ±s.e.m., and n denotes the number of animals used in each experiment. For specific two means comparisons Student's unpaired t test was used. A value of P < 0.05 was considered significant.

Results

Estimations of total elastin content determined by confocal microscopy showed no difference between mesenteric arteries from 1- and 6-month-old rats and a significantly smaller elastin content in mesenteric vessels from 10-day-old rats. Middle cerebral arteries had a smaller elastin content compared to mesenteric vessels (Table 1).

Table 1.  Extracellular matrix content, internal elastic lamina organization and intrinsic stiffness in rat small mesenteric and cerebral arteries
 Mesenteric arteriesCerebral arteries

6-month-old rats
10-day-old rats1-month-old rats6-month-old rats
  1. IEL, internal elastic lamina. Total fluorescence was calculated for 1 mm length vessel, n= 5–11 rats. *P < 0.05 cerebral arteries compared to mesenteric arteries at the age of 6 months; †P < 0.05 when compared to mesenteric arteries from the previous age.

Total fluorescence (a.u.)476.3 ± 82 1277.9 ± 94†1316.4 ± 167690.0 ± 147*
Desmosine content (pm mg−1)1139.5 ± 100  1459 ± 101  908 ± 128*
IEL thickness (μm) 4.8 ± 0.2    3.5 ± 0.1+   2.5 ± 0.1†  1.9 ± 0.1*
Fenestrae area (μm2)2.99 ± 0.2   12.5 ± 1.6† 36.6 ± 5†   3.4 ± 0.3*
Medial collagen content (a.u.)  24.6 ± 5.7  25.6 ± 6.9
Adventitial collagen content (a.u.)102.9 ± 8.0   82.7 ± 10.3
Intrinsic stiffness (β-values)12.5 ± 2.1   7.9 ± 0.6†  3.98 ± 0.1†  8.9 ± 1.2*

Estimation of elastin by desmosine confirmed confocal microscopy results: a similar content in mesenteric arteries from 1- and 6-month-old rats and a significantly smaller content in cerebral vessels compared to mesenteric arteries (Table 1). The amount of tissue was not sufficient to determine desmosine in mesenteric arteries from 10-day-old rats.

All arteries showed a fenestrated IEL (Fig. 1). During growth of the rats there was a parallel and significant increase of size of the fenestrae and a reduction of IEL thickness in mesenteric arteries (Table 1). In cerebral vessels, IEL was significantly thinner and the area of individual fenestrae was much smaller when compared to age-matched mesenteric vessels (Fig. 1 and Table 1).

Figure 1.

Confocal projections of the internal elastic lamina of mesenteric and cerebral arteries from rats of different ages
Vessels were pressure-fixed at 70 mmHg and mounted intact on a slide. Projections were obtained from serial optical sections captured with a fluorescence confocal microscope (×63 oil immersion objective, zoom 4).

Collagen content was measured in the medial and adventitial layers from the mesenteric arteries of 1- and 6-month-old rats. There was no difference in collagen content between young and adult rats in the media or in the adventitia (Table 1).

β-Values were used as an indicator of intrinsic stiffness independent of vessel geometry; the higher the β-values, the larger the wall stiffness. In mesenteric arteries, β-values decreased with the age of the rat. In cerebral arteries, β-values were significantly larger when compared to mesenteric arteries from age-matched rats (Table 1). Taking all data together, the best correlation was found between β-values and the area occupied by fenestrae (r2= 0.82); however, the correlation between IEL thickness and β-values was poor (r2= 0.44).

Discussion

The role of elastin in the mechanics of large arteries has been thoroughly studied and its importance for cardiovascular function is beyond doubt. However, the function of elastin in small arteries has been, in general, neglected owing to the assumption that, being in low quantity, elastin is not a major contributor to the mechanics of resistance arteries. This has been also due, in part, to methodological difficulties associated with the determination of elastin content in small vessels. We have previously developed a method for estimating the content of elastin in resistance-size arteries by measuring, with confocal microscopy, the fluorescence intensity and volume occupied by elastin in 1 mm long segments of intact arteries (Briones et al. 2003). The present study aimed to validate this semiquantitative method against a biochemical assay that permits evaluation of the amount of elastin and total protein content of very small samples (Starcher & Conrad, 1995; Starcher, 2001). This method is based on measurements of desmosine, an amino acid cross-link found only in elastin (Eyre, 1984), and a very sensitive ninhydrin-based assay to quantify the total protein content of samples less than 500 μg. This method has been used previously to quantify the elastin content in mouse vessels (Faury et al. 2003). The present results demonstrate that the estimations of elastin content in cerebral and mesenteric resistance arteries by confocal microscopy matched the biochemical determinations of desmosine, indicating that both methods are valid for comparison of elastin content between experimental groups. Both assays have advantages and weaknesses and so they are complementary. The biochemical method has the advantage of being a quantitative assay, but has the limitation of requiring a minimum amount of tissue. However, the confocal assay: (1) allows for simultaneous assessment of the amount and organization of elastin within the same artery; and (2) it can be used in single arterial segments. Laser beams fluctuate over time and might alter intensity measurements. Therefore, the confocal determination of elastin content can only be used for comparison between experimental groups provided that all measurements are taken under identical conditions (levels of laser intensity, brightness and contrast). In conclusion, the present data demonstrate that the confocal-based method for estimation of elastin content that we developed for resistance arteries (Briones et al. 2003) is a valid semiquantitative assay, particularly useful for very small samples, that also allows for simultaneous study of elastin organization.

The second objective of the present study was to determine the relative contribution of elastin content and organization in the IEL to the intrinsic elastic properties of resistance arteries. We determined the degree of stiffness of the vascular wall by measuring the parameter β, which is independent of vessel geometry. Mesenteric resistance artery stiffness decreased with the age of the rat, as shown by the gradual reduction in β-values. This difference could be attributed to changes in the content and/or in the organization of ECM proteins. Therefore, we determined changes with age in the content of elastin and collagen, the main determinants of the passive vascular mechanical properties (Dobrin, 1978). Under normal circumstances, elastin is synthesized in quantity only in embryonic and rapidly developing tissues, and in blood vessels the rates of synthesis increase to a maximum in the perinatal period (Bendeck & Langille, 1992; Rosenbloom et al. 1993; Davis, 1995). This was also confirmed by our data showing an increase in elastin content from the 10th day to the 1st month with no further increase by the 6th month of life. We also evaluated whether changes in collagen content could participate in the reduction in stiffness of mesenteric arteries with age. As previously described, collagen was located mainly in the adventitia and, in smaller amounts, in the medial layer (Dobrin, 1978). Our data show that collagen content was not modified between 1 and 6 months of age in the adventitia or media. These data suggest that the reduction of intrinsic stiffness observed in mesenteric arteries between the 1st and the 6th month of life is not directly related to a modification in the content of the fibrous proteins collagen or elastin.

Vessels are not static entities and are able to change their structure, i.e. to remodel, throughout the life of the individual. Vascular remodelling may be considered as an adaptive process in response to long-lasting changes in the haemodynamic environment to maintain a relatively constant tensile and/or shear stress (Langille, 1993; De Mey et al. 2005). This remodelling process is accompanied not only by geometrical changes, but also by reorganization of vascular wall components. In rat mesenteric resistance arteries, the IEL remodelled with age, with a gradual reduction of IEL thickness and a parallel increase in the area occupied by fenestrae. Similar changes have been reported in rabbit carotid arteries during postnatal development (Wong & Langille, 1996). We suggest that the reorganization of IEL observed in mesenteric resistance arteries might be related to the decrease in stiffness with age. β-Values correlated better with the area occupied by fenestrae than with IEL thickness, suggesting that enlargement of fenestrae might be a physiological mechanism implicated in the reduction of mesenteric artery stiffness during vessel growth.

The quantitative and qualitative changes in elastin observed between cerebral and mesenteric vessels parallel vascular mechanical differences in these two vascular beds. Middle cerebral arteries showed less elasticity and, accordingly, the size of fenestrae and elastin content were smaller, when compared to mesenteric arteries. We suggest that these differences in the elastic properties of vessels of relatively similar size (∼250 μm internal diameter) might be related to the different physiological roles and response to haemodynamic stimuli of cerebral and mesenteric arteries. Cerebral vessels control cerebral blood flow by their ability to constrict in response to increases in intramural pressure (Faraci et al. 1989), SMCs playing a prominent role in the regulation of internal diameter. However, in mesenteric vessels, which exhibit a comparatively lower degree of myogenic constriction (Sun et al. 1992), passive elastic properties might play a key role in regulating vessel dimensions. In addition, middle cerebral arteries are tethered to the surrounding tissue and have relatively low freedom of movement, while mesenteric vessels require a larger degree of elasticity because they irrigate the intestine, which is subjected to continuous movements. In conclusion, the present data suggest that, in resistance arteries, the organization of the elastin in the IEL might be a critical factor for the intrinsic elastic properties of the vascular wall.

Appendix

Acknowledgements

This work was supported by European Union (QLG-CT-1999-00084), MCyT (BFI 2001-0638) and CAM (GR/SAL/0093/2004).

Ancillary