Compromised Elastin Network Load Transfer in Aging Human Aorta

Abstract

Elastic fibers are essential for passive arterial load bearing and must withstand billions of loading cycles throughout life. Although arterial stiffening is a hallmark of cardiovascular aging, how elastin mechanical function evolves with aging in humans remains poorly understood. This study investigated age-dependent changes in the mechanical properties and microstructure of purified human aortic elastin networks.

Purified elastin networks were obtained from human thoracic aortas (17 in total; 9 males, 8 females; ages 26–92 years). Uniaxial tensile testing was performed, and tangent modulus at 1.1 stretch was calculated from Cauchy stress-stretch responses. Elastin microstructure was examined using multiphoton microscopy, histology, and scanning electron microscopy (SEM), combined with quantitative image analysis.

Elastin tangent modulus decreases significantly with aging in male subjects (p < 0.05). Multiphoton imaging revealed progressive reduction in fiber straightness with age (p < 0.05 in males), with young elastin networks characterized by straight, continuous fibers that primarily reoriented under load, whereas older networks exhibited increased waviness, fragmentation, and loss of load-bearing fibers. Histological analysis further demonstrated a significant age-related reduction in lamellar length in males (p < 0.05), indicating increased network discontinuity. SEM further demonstrated pronounced ultrastructural deterioration in older male samples, with elastic fibers appearing fragmented and frayed compared to the continuous and organized fibers in younger males. These microstructural alterations were consistent with diminished mechanical engagement of elastin fibers under tensile loading in the older male samples. In contrast, female samples did not exhibit consistent age-dependent deterioration in either mechanical or microstructural metrics.

Graphical Abstract

1. Introduction

Arteries are complex organs with a hierarchical ultrastructure, in which the extracellular matrix (ECM) bears the majority of wall stress and largely determines the mechanical behavior [1]. One of the major ECM components is elastic fiber, which forms concentric layers of elastic lamellae interspersed with collagen fibers, smooth muscle cells, proteoglycans and other extracellular components. This lamellar structure provides both structural integrity and elasticity to the vessel wall [2]. Elastic fibers dominate the mechanical responses at low strains, providing compliance during the early stages of arterial deformation, whereas collagen fibers are progressively recruited with increasing deformation, leading to strain stiffening. This coordinated, sequential engagement of elastic and collagen fibers enables large elastic arteries such as aorta to accommodate the cyclic diameter changes imposed by pulsatile blood flow during each cardiac cycle [3, 4, 5].

Elastic fibers play a dominant role in passive arterial load bearing and withstand billions of loading cycles over a human lifetime. Numerous studies have demonstrated that arterial compliance decreases with age, manifested by reduced cyclic circumferential strain in the thoracic aorta and increased arterial stiffness [6, 7]. Elastin, which made up 90% of elastin fibers [8], has a slow turnover rate with an estimated half-life of 40–80 years, and is therefore particularly vulnerable to cumulative mechanical and biochemical damage over time [9, 10]. Over decades of repetitive loading, histological studies consistently report age-associated fragmentation, thinning, and loss of elastic lamellar integrity, changes that compromise arterial elasticity and promote pathological remodeling such as aortic dilation and stiffening [11, 12, 13]. As elastin progressively degrades without effective replacement, mechanical load is increasingly transferred to the stiffer collagen fibers, further accelerating vascular stiffening and functional decline [14], [15], [16]While mechanical fatigue alone can drive progressive elastin damage, age-associated biochemical modifications substantially exacerbate elastin degradation by reducing fatigue resistance, impairing structural integrity, and increasing susceptibility to enzymatic cleavage. As a long-lived ECM protein, elastin is progressively modified by both enzymatic and non-enzymatic age-related chemical reactions over time [13], which alter its molecular structure and macroscopic mechanical behavior, thereby accelerating functional deterioration during aging.

To elucidate the contribution of elastin to arterial function, several studies have directly examined the mechanical behavior of purified elastin network. Zou and Zhang (2009) characterized purified bovine elastin networks using biaxial tensile test and demonstrated that the elastin network possesses anisotropic stress-strain behavior and primarily governs the low-strain response of the arterial wall. Their study further showed regional variations in elastin mechanical properties, a finding also reported in Lillie et al. (2012), highlighting non-uniform mechanical loading of elastin within the vessel wall. More recently, Wang et al. (2021) investigated the role of elastin in interlamellar bonding within the aortic media using peeling test. By comparing purified porcine aortic elastin with intact arterial media, they showed that isolated elastin exhibits substantially lower strength and energy release than native tissue. Collectively, these studies employed purified elastin network to probe elastin’s functional contribution to arterial mechanics, while also emphasizing the critical role of structural context in interpreting mechanical behavior. Additional studies have examined purified elastin subjected to ex vivo nonenzymatic glycation treatment and reported significant alterations in elastic and viscoelastic properties [16, 17, 18], accompanied by molecular dynamical and structural modifications of elastin [20]. Together, these findings demonstrate that biochemical modifications can induce measurable mechanical consequences, underscoring the need to study elastin aging under physiologically relevant mechanical and biochemical environments.

While prior work has substantially advanced our understanding of elastin’s important role in arterial mechanical function and the effects of biochemical modification, critical gaps remain. Most existing studies rely on animal tissues or elastin subjected to ex vivo biochemical treatments, leaving the mechanics and structure of aged human elastin largely unexplored. Moreover, how progressive aging of the human elastin network alters its mechanical function and microarchitecture remains poorly understood. To address these gaps, this study investigates age-related changes in the mechanical properties and microstructure of the human elastin network using an integrated experimental framework that combines biomechanical testing, multiphoton imaging, histology, scanning electron microscopy (SEM), and quantitative imaging analysis to directly link age-associated structural alterations of elastin to corresponding changes in mechanical function.

2. Material and Methods

2.1. Sample preparation

The use of human tissue was approved by the Institutional Biosafety Committee at Boston University. Human descending thoracic aortas were procured from the National Disease Research Interchange (NDRI), transported to laboratory on dry ice, and stored at −80°C until experiment. Samples from older donors became available more quickly than those from younger donors, resulting in variable storage durations, with some samples stored for several years before testing. All samples were collected from approximately the same anatomical location of descending thoracic aorta. A summary of donor characteristics, including age, sex, race, and cause of death is provided in Table 1. Before experiments, samples were thawed at −20°C for 5 hours, followed by 3 hours at 4°C and 1 hour at room temperature. Square human descending thoracic aorta samples were purified via cyanogen bromide (CNBr) treatment to remove cells, collagen, and other ECM components [21]. Briefly, each sample was incubated in 50 mg/mL CNBr in 70% formic acid at room temperature for 19 hours, followed by heating to 60 °C for 1 hour with gentle stirring, and then boiled for 5 minutes to inactivate CNBr. The purified elastin network was rinsed thoroughly and stored in 1× phosphate-buffered saline (PBS) solution. Each sample then was cut with a dimension of approximately 12 × 6 mm with the longer axis aligned in the circumferential direction. A total of 17 samples were obtained from 17 donor subjects (9 males and 8 females). Male ages ranged from 26 to 92 years, while female ages ranged from 39 to 90 years. There was no significant difference in age distribution between male and female cohorts. The full cohort (9 male, 8 female) was used for uniaxial tensile testing, elastic fiber straightness analysis from multiphoton microscopy (MPM), and elastic lamellar length from histology. A subset of eight samples (4 male, 4 female), selected to provide comparable age distributions between sexes, was used for multiphoton imaging under in situ loading experiments.

Table 1.

Summary of donor characteristics for human descending thoracic aorta samples included in this study. Donor characteristics include age, sex, race, and cause of death. Race: C, Caucasian; B, Black. Cause of death: SIGSW, self-inflicted gunshot wound; MVA, motor vehicle accident; CPA, cardiopulmonary arrest; CVA, cerebrovascular accident; ABI, anoxic brain injury; HD, heart disease; COPD, chronic obstructive pulmonary disease; CHF, congestive heart failure.

Donor	Age (years)	Sex	Race	Cause of Death
1	26	M	C	MVA
2	32	M	C	CPA
3	39	F	C	CVA
4	42	M	C	CPA
5	44	F	C	ABI
6	47	M	C	CPA
7	49	M	B	CVA
8	57	F	C	ABI
9	59	M	C	HD
10	62	F	C	CVA
11	64	F	Unknown	CPA
12	70	M	C	CVA
13	72	F	C	CPA
14	74	F	C	COPD
15	87	M	C	CPA
16	90	F	C	CPA
17	92	M	C	CHF

2.2. Uniaxial tensile testing

To characterize the elastic and viscoelastic mechanical properties of the elastin network, uniaxial tensile tests were conducted using a uniaxial tensile tester equipped with a 5 N load cell (ADMET, Inc., Norwood, MA). Samples were mounted on the testing device submerged in 1× PBS under room temperature to maintain hydration. A small tensile preload, between 0.003–0.005 N, was applied to remove slack and establish the preloaded reference configuration. Elastin samples were then subjected to 20 cycles of uniaxial stretch-control loading to a maximum stretch of 1.1 at a loading rate of 5 mm/min. The stress-stretch relationship was derived from the final unloading cycle, where Cauchy stresses were calculated using Equation 1 with incompressibility assumption. The reference configuration (λ = 1.0) was defined as the sample configuration following preload. For the final unloading curve, the stress was subsequently offset such that the stress at λ = 1.0 was set to zero. Tangent modulus was calculated by differentiating the fitted 3rd-order polynomial curve to the stress-stretch response plus evaluating the resulting derivative at the stretch of 1.1. The polynomial fit was needed to eliminate the experimental noise and provide a stable estimate of the tangent modulus while capturing the nonlinear stress-stretch response.

$$\sigma =\frac{F\lambda }{T_0{W}_0}$$ (1)

where $\sigma$ is Cauchy stress, $F$ is the applied force measured, and $\lambda$ is stretch. The initial thickness $T_0$ and the initial width $W_0$ was measured using a digital caliper at multiple locations and then averaged.

2.3. Multiphoton Microscopy (MPM)

Multiphoton images were captured using a FVMPE-RS confocal microscope (Olympus Life Sciences, Inc.) equipped with a 25× water-immersion objective lens (NA 1.05). A femtosecond IR pulse laser was set to 805 nm to produce two-photon excitation fluorescence (2PEF) signals for elastin (495–540 nm). Samples were placed under the microscope and submerged in 1× PBS solution during imaging (Fig. 4j). A stack of images (~60 to 160) were acquired from the intimal side of each sample with an increment of 1 μm. The images were adjusted for consistent brightness and contrast and underwent noise filtration in ImageJ as a preprocessing step. Maximum intensity projection was performed to every five consecutive slices within each image stack, generating a single consolidating image to ensure comprehensive coverage and optimal representation of the microstructure. Subsequently, CT-FIRE (v3.0) was used to quantify fiber straightness for each consolidated image within the stack. Specifically, CT-FIRE was used to automatically segment individual elastin fibers within each multiphoton image prior to quantitative analysis. Fiber straightness was defined as the ratio of the distance between the two endpoints of a fiber to the contour length measured along the fiber path. The resulting straightness value ranges from 0 to 1, where a value of 1 represents a perfectly straight fiber and lower values indicate increasing fiber undulation. Histograms of fiber straightness were plotted for each sample to enable comparison of straightness distributions across samples of different ages. Median fiber straightness was also obtained for each sample.

Figure 4.

Representative multiphoton images from male donors showing the effect of stretch on elastin fiber organization within elastin network. (a-c) a 32-year-old, (d-f) a 59-year-old, and (g-i) an 87-year-old male imaged at 1.0, 1.1, and 1.2 stretch. C and L indicate the circumferential and longitudinal anatomical directions of the samples. Scale bar represents 50 μm for all images. Probability density function (PDF) of fiber straightness distribution for (j-l) the 32-year-old male samples shown in a-c, (m-o) the 59-year-old male sample in (d-f), and (p-r) the 87-year-old male sample in (g-i).

To assess fiber network structure changes under mechanical loading, elastin samples (4 male and 4 female subjects) were imaged while the samples were subjected to mechanical loading at 1, 1.1, and 1.2 stretch using a custom-made stretching device. To evaluate network structure under mechanical loading, fiber straightness was quantified for each sample at varying stretches using CT-FIRE. Histograms of fiber straightness were analyzed following the procedure described above, and the median value was determined for each sample at each stretch.

2.4. Histology

To evaluate age-related differences in elastin architecture, histological cross-sectional images of the elastin samples were also acquired. Purified elastin samples were cut into thin strips in both the circumferential and longitudinal directions, fixed in 4% paraformaldehyde for 24 hours, and stored in 1× PBS. The samples were then embedded in paraffin, sectioned into 4 μm slices, and stained with Verhoeff-Van Gieson (VVG). An Olympus VS120 slide scanner (Evident Scientific, Waltham, MA) was then used to obtain the images. Quantitative analysis of elastic lamellar integrity was subsequently performed using CT-FIRE tracker, from which individual elastic lamella were tracked in both young and aged samples. The median elastic lamellar length for each sample was then obtained.

2.5. Scanning Electron Microscopy (SEM)

Surface morphology of the elastin network was examined using a Phenom PROX Desktop SEM (Thermo Fisher Scientific) operated at 5 kV to investigate fiber-level architecture and identify age-related alterations in microstructure. Prior to SEM, elastin samples were fixed in Karnovsky’s fixative kit (2% paraformaldehyde, 2.5% glutaraldehyde and 0.1 M sodium phosphate buffer, Fisher Scientific) for 24 hs and then dehydrated in a graded ethanol series (0%, 25%, 50%, 75%, 90%, and 100%), for 10 min in each solution. The samples were then dried using a critical point dryer (Supercritical Autosamdri^®−815). A very thin layer of platinum was coated on the cross-section of the samples using a Sputter Coater (Cressington 108).

2.6. Statistical Analysis

Each donor was considered an independent biological sample for statistical analysis. For measurements obtained from multiple fibers or lamellae within a donor, donor-specific values were obtained for each measurements, which were then used for statistical comparisons. Fiber straightness parameters from multiphoton imaging analysis are reported as the median, with 95% confidence intervals to illustrate the trend across age groups, and are presented in the results as median ± standard error (SE). Differences in fiber straightness parameters across stretch levels were further evaluated using two-way analysis of variance (ANOVA). Post hoc multiple comparisons were performed to assess differences in straightness between stretch conditions within each sample. Lamellar length from histology images is shown as median ± standard deviation (SD). Additionally, thickness, tangent modulus, straightness parameters, interquartile range (IQR), and lamellar length were all analyzed using linear regression fits across age. Linear regression was performed to assess the relationship between variables. The significance of the slope differing from zero was evaluated using an unpaired, two-tailed t-test, with a threshold of p < 0.05 considered statistically significant. For tangent modulus, median straightness, and lamellar length, differences between male and female age-related regression slopes were further evaluated using ANCOVA to assess the significance of the age × sex interaction.

3. Results

The Cauchy stress-stretch curves of all samples are shown in Fig. 1a and Fig. 1b for male and female samples, respectively. In male samples, with aging, the Cauchy stress-stretch curves become less steep and progressively shift down (Fig. 1a). Accordingly, the tangent modulus decreases significantly with age (Fig. 1c; p = 0.0219), ranging from 322.6 to 54.6 kPa. For female samples, no obvious age-dependent changes were observed in the Cauchy stress-stretch curves (Fig. 1b). The corresponding tangent modulus for female samples, ranging from 57.154 to 129.893 kPa, lacks significant age-dependent trend (Fig. 1c; p = 0.7786). However, the difference between male and female regression slopes was not statistically significant (p = 0.0792). The thickness of the purified elastin network in male specimens increased significantly with age (p = 0.0103), whereas the thickness increase was not significant (Fig. 1d; p = 0.1550) in female specimens.

Figure 1.

Age- and sex-dependent mechanical properties of purified elastin network from human thoracic aorta. Quantitative analyses were performed using the full cohort of 17 samples (9 male and 8 female); each data point or curve represents an individual donor. (a) Cauchy stress-stretch response of circumferential strips of purified elastin isolated from male subjects across ages. (b) Cauchy stress-stretch response of purified elastin from female subjects across ages. (c) Elastin sample thickness as a function of age for male and female samples, showing an age-associated increase in thickness for both sexes, with a more pronounced, significant increase in males (p = 0.0103) and no-significant change in females (p = 0.7074). (d) Tangent modulus as a function of age, demonstrating a significant decrease with age in male samples (p = 0.0219) and no significant age-dependent change in females. Symbols with darker shades indicate older samples and lighter shades indicate younger samples.

Representative multiphoton images (Figs. 2a-d) for the elastin network from male samples, ages 26, 49, 59, and 87, show that the elastic fibers in younger male samples appear to be more organized, dense, continuous and straight as opposed to the disorganized, recoiled and randomly structured fibers in the older samples. In comparison, elastin fibers in female elastin network (Figs. 2e-h), ages 39, 44, 57, and 74, appear more uniform, well-organized and intact than those in older male samples, and age-related disorganization of the network appears to be less pronounced. Figs. 2i-p show the corresponding histograms of straightness parameters distribution for these multiphoton images. The histograms of younger male (26 years old) sample show sharper peaks and smaller variation between the straightness of fibers, where most fibers have very high straightness above 0.9, indicating relatively straight fibers in these samples. In contrast, the histograms of older male sample (49, 59, and 87 years old) demonstrate a progressively broader distribution in fiber straightness due to the presence of undulated fibers, whereas there are still straight fibers present, the frequency of these fibers is less. The histograms of younger female samples (39 years old) appear similar to those of younger male samples (26 years old) samples. However, compared with the older male samples, the older female samples (44, 57, and 74 years old) exhibit a relatively straighter fiber network with less obvious changes in fiber straightness distributions (Figs. 2m-p), further supporting the more organized and straight elastin fibers in female than in male samples.

Figure 2.

Representative multiphoton images showing age- and sex-dependent changes in elastin fiber organization in purified elastin network. (a-d) male subjects at ages of 26, 49, 59, and 87 years old, and (e-h) female subjects at ages of 39, 44, 57, and 74 years old. C and L indicate the circumferential and longitudinal anatomical directions of the samples. Scale bar represents 50 μm for all images. With increasing age, elastin fibers appeared progressively more disorganized and wavier, an effect more pronounced in males. (i-l) Histograms of fiber straightness corresponding to images in (a-d), showing a shift from sharp, narrow distributions at high straightness in younger samples to broader distributions in older samples. (m-p) Corresponding histograms for female images in (e-h), showing no clear age-dependent trend.

Figure 3a further shows the straightness parameter quantification results for all samples. In male samples, median straightness decreases significantly with age (p < 0.0001), ranging from 0.967 ± 0.001 to 0.838 ± 0.006, whereas female samples do not exhibit a consistent age-related trend (p = 0.7929). Furthermore, the median fiber straightness exhibited significantly age-related trends between males and females (age × sex interaction, p = 0.0082). The IQR of fiber straightness (Fig. 3b) increases significantly with age in male samples but shows only a slight, non-significant increase in female samples (p < 0.0001 and p = 0.3123 for male and female samples, respectively). Taken together, multiphoton imaging and quantification results demonstrate that older males have fibers ranging widely from straight to wavy, while younger males exhibit more uniform straightness, whereas age-related variability is less pronounced in females.

Figure 3.

Age- and sex-dependent changes in elastin fiber straightness performed using the full cohort of 17 donors (9 male and 8 female) (a) Median fiber straightness as a function of age for male and female purified elastin network. Male elastin network exhibited a decrease in median straightness with aging (p < 0.0001), whereas females showed no clear age-dependent trend (p = 0.5699). Shaded regions indicate 95% confidence bounds for each sex. (b) Interquartile range (IQR) of fiber straightness versus age, showing an age-associated increase in variability for both sexes, with a more pronounced, significant increase in males (p < 0.0001) and non-significant in females (p = 0.3123).

To assess the elastin fiber network integrity and its response to mechanical loading, multiphoton images were obtained while the elastin network was subjected to mechanical loading. Representative images from male samples, aged 32, 47, 59, and 87 years, at different stretches are presented in Figs. 4a-i with the corresponding fiber straightness histograms presented in Figs. 4j-r. In the younger male sample (32 years old), the fiber network appeared straight and continuous. Upon loading, these fibers primarily underwent reorientation, aligning more closely in the circumferential direction corresponding to the applied load (Figs. 4a-c). The corresponding histograms show a similar trend (Figs. 4j-la-c), i.e., the distributions are already relatively sharp in the relaxed state and become slightly sharper with increasing stretch. The 59-year-old male sample exhibits fibers that become both straighter and more uniformly aligned under loading (Figs. 4d-f). The corresponding histograms also demonstrate an increase in sharpness with increasing stretch (Figs. 4m-o). In contrast, the older male sample (87 years old) contains a mixture of fragmented, coiled, and relatively straighter fibers. With increasing load, some fibers exhibit slight straightening, while fragmented fibers show minimal structural change. The fragmented fibers appear non-functional and do not contribute to load bearing, whereas wavy but continuous fibers straighten under load (Figs. 4g–i). The corresponding histogram initially shows a broad distribution and exhibits minimal sharpening with increasing stretch (Figs. 4p-r). Representative images from a 44-year-old and a 72-year-old female sample are shown in Fig. 5. The relatively younger female sample (44 years old) exhibits increasing fiber straightness with increasing load (Figs. 5a–c), and the corresponding histograms become sharper as stretch increases (Figs. 5g–i). The older sample (72 years old), which displays a more disorganized network consisting of a combination of fragmented and intact fibers, shows slight straightening with increasing load (Figs. 5d–f). The corresponding histograms also become slightly sharper with increasing stretch (Figs. 5j–l).

Figure 5.

Representative multiphoton images from female donors showing the effect of stretch on elastin fiber organization and straightness within female elastin network. (a-c) a 44-year-old and (d–f) a 72-year-old female sample at 1.0, 1.1, and 1.2 stretch. C and L indicate the circumferential and longitudinal anatomical directions of the samples. Scale bar represents 50 μm for all images. Probability density function (PDF) of fiber straightness distribution for (g-i) the 44-year-old sample shown in a-c, and (j-l) the 74-year-old sample shown in d-f.

Figure 6 shows the median straightness parameter of each sample vs. stretch. Similar to what is observed in the histograms of male samples, in Figure 6b the fibers of the 32-year-old sample are already predominantly straight, with median straightness of 0.947 ± 0.001, and maintain their straightness throughout stretching, with median straightness of 0.947 ± 0.001 at stretch 1.1 and 0.954 ± 0.001 at 1.2, respectively. In contrast, the 47 and 59 years old samples show pronounced and continuous straightening with stretch. The 47-year-old sample’s straightness increases from 0.924 ± 0.003 at no stretch to 0.933 ± 0.002 at 1.1 stretch, and then to 0.943 ± 0.002 at 1.2 stretch. The 59-year-old sample’s median straightness increases from 0.892 ± 0.004 at no stretch to 0.923 ± 0.002 at stretch 1.1, and then to 0.946 ± 0.001 at stretch 1.2. The older sample (87 years old), with a median straightness of 0.861 ± 0.004 at no stretch, shows moderate straightening, with median straightness increases to 0.893 ± 0.003 at 1.1 stretch and 0.891 ± 0.004 at 1.2 stretch. As shown in Fig. 6, the median straightness of the younger samples is higher than that of the older samples, consistent with findings from MPM in the relaxed state. Figure 6b illustrates the changes in median straightness across different stretch levels for female samples. The age-related trend is less pronounced compared to that observed in male samples.

Figure 6.

Effect of stretch on elastin fiber straightness across age and sex. Median fiber straightness as a function of applied stretch for (a) male elastin samples (32, 47, 59, and 87 years old), and (b) female elastin samples (39, 44, 57, and 72 years old) at stretches λ = 1.0, 1.1, and 1.2. Error bars represent standard error (SE) (* p < 0.05, ** p < 0.01, ^† p < 0.001).

The youngest female sample’s (39 years old) median straightness increases from 0.950 ±0.002 at no stretch to 0.954 ± 0.001 and 0.962 ± 0.001 at stretch values of 1.1 and 1.2, respectively. The most notable increase with stretching is observed in the 44-year-old female sample, from stretches 1.0 to 1.1, with median straightness increasing from 0.892 ± 0.004 to 0.916 ± 0.002 and subsequently to 0.942 ± 0.001 at 1.2 stretch. The 57-year-old sample’s median straightness increases from 0.948 ± 0.001 at no stretch to 0.950 ± 0.001 and 0.965 ± 0.001 at stretches 1.1 and 1.2, respectively. In contrast, the oldest sample’s (72 years old) median straightness shows slight increase from 0.903 ± 0.003 at no stretch to 0.910 ± 0.004 at stretch 1.1 to 0.912 ± 0.004 at 1.2 stretch.

Assessment of lamellar structural integrity and fragmentation associated with aging is presented in Fig. 7, where lamellar length from histology images was quantified for both male and female samples. The younger male sample exhibits more tightly packed lamellae that are generally continuous and of greater average length (Fig. 7a). In contrast, the older male sample shows lamellae that are more fragmented, loosely connected, and structurally disorganized (Fig. 7b). Male elastin network exhibits a significant decreasing trend in median lamellar length with aging (Fig. 7c) (p = 0.0172), with values ranging from 65.4 ± 0.8 μm to 94.4 ± 3.2 μm. The female elastin networks (Figs. 7d and 7e) do not exhibit substantial differences in lamellar structure between the younger and older samples. Consistently, the median lamellar lengths of female specimens did not exhibit a clear age-related trend (Fig. 7f) (p = 0.7684). However, the age-related trends in lamellar length did not differ significantly between males and females (age × sex interaction, p = 0.1287).

Figure 7.

Age- and sex-dependent changes in elastin lamellar organization. Representative histological cross-section of (a) a younger male sample (32 years old), (b) an older male sample (87 years old), (d) a younger female sample (44 years old), and (e) an older female sample (74 years). Elastin lamellar length as a function of age for (c) male samples, showing a significant decrease in lamellar length with aging (p = 0.0172); and (f) female samples, showing no strong age-dependent trend and approximately constant lamellar length (p = 0.7684). Lamellar length was quantified using 17 donors (9 male and 8 female). Error bars represent SD.

The SEM images of elastin networks further show the striking age-related ultrastructural differences between the younger and older elastin samples. In the 32-year-old male elastin sample (Figs. 8a and 8b), the elastin fibers appear intact, continuous, and uniform in diameter. In contrast, in the 87-year-old male elastin sample (Figs. 8c and d), the elastic fibers appear degraded, fragmented, and frayed, indicating age-related deterioration of the elastin fibers. Similarly, the 39-year-old female sample shows a continuous and relative uniform elastin network (Figs. 8e and f). In the 74-year-old female sample, the fibers appear more disorganized compared to the younger female sample; however, the network remains largely continuous, with no obvious evidence of frayed or fragmented fibers (Figs. 8g and h).

Figure 8.

Representative SEM images showing the ultrastructure of elastin fiber network (n = 4). (a, b) a 32-year-old male subject, revealing continuous, well-integrated fiber networks; (c, d) an 87-year-old male sample, showing pronounced fiber fragmentation, increased fuzziness, and frayed, irregular fiber morphology; (e, f) a 39-year-old female subject, showing continuous, organized fiber network; and (g, h) a 74-year-old female subject with less pronounced signs of fragmentation or frayed like fibers. The scale bars represent 15 μm in all images.

4. Discussion

Age-related arterial stiffening is a hallmark of cardiovascular aging, yet the specific contribution of elastin, the primary determinant of arterial compliance at physiological strains, remains poorly understood in humans. Previous studies have largely inferred elastin’s role from intact tissue measurements, animal models, or ex vivo chemically modified preparations, limiting direct insight into how natural aging alters elastin structure-function relationships. By integrating mechanical testing of purified human aortic elastin network with quantitative multiscale microstructural imaging and analyses across a broad age range, the present study directly links elastin mechanics to its underlying architecture. Across both sexes, variations in mechanical behavior were consistently explained by corresponding changes in elastin microstructure, demonstrating a direct structure-function relationship at the network level. Notably, this relationship manifested differently with aging: male elastin exhibited progressive fragmentation and loss of network integrity accompanied by reduced mechanical engagement under tensile loading, whereas female elastin showed relative preservation of both microstructure and mechanical properties. Together, these findings provide the first direct experimental evidence in humans that age-related changes in arterial elastin mechanics are governed by microstructural integrity, with pronounced sex-specific trajectories.

Our results demonstrate a strong relationship between microstructural integrity and macroscopic mechanical behavior in both sexes, with aging-related alterations being more pronounced in male samples, where aging was associated with decreased fiber straightness, increased IQR, greater variability in fiber straightness (Fig. 3), as well as pronounced fragmentation observed in multiphoton imaging, histology, and SEM (Figs. 2, 7, and 8), and a corresponding decline in elastic modulus (Fig. 1c).The presence of coiled or wavy fibers, indicative of localized rupture or loss of continuity within the elastin network, supports the interpretation that mechanical weakening arises from progressive deterioration of elastin architecture. These observations align with and extend previous findings on elastin degradation in aging and disease. Prior studies have reported fragmented, “moth-eaten” elastic fibers in Fibulin-4 mutant mouse aorta [22], age-related widening of gaps within elastic lamellar layers in rat aorta [23], and greater elastic fiber fragmentation with disrupted lamellar continuity in older mouse aorta [24]. Comparable degeneration including coiled and discontinuous elastic fibers has also been documented in bovine abdominal aorta [25]. Our results provide direct structural-mechanical evidence in human tissue that aging-related fragmentation of the elastin network leads to impaired mechanical performance, especially in male samples.

The histological analyses (Fig. 7) revealed a pronounced decrease in elastic lamellar length with age in male samples, reflecting fragmentation and discontinuity of the elastin network. To characterize this trend, linear regression was used to describe the overall relationship between lamellar length and age. The linear model was intended to capture the observed trend rather than imply that elastin degradation progresses linearly throughout aging. A larger cohort spanning a broader age range will be needed to more fully characterize the progression of lamellar degradation. Complementary observations from SEM imaging (Fig. 8) further illustrate these age-related changes at the ultrastructural level - fibers from older samples exhibit extensive fragmentation and a frayed appearance reminiscent of worn rope, whereas fibers from younger samples appear largely intact, showing minimal signs of damage or disintegration. Collectively, these findings underscore the progressive deterioration of elastin architecture with aging. It is worth mentioning that potential confounding effects arising from age-related increases in sample thickness were carefully evaluated when comparing elastin networks across all ages. Sample thickness tended to increase with age (Fig.1d), likely due to progressive weakening and loosening of network integrity. However, the observed age-related weaking of the elastin network, including the reduction in tangent modulus (Figs. 1a and 1c), cannot be explained solely to the thickness changes. To further eliminate thickness-dependent effects, mechanical responses were also analyzed using tension, which is independent of sample thickness. These analyses revealed similar age-dependent decline in elastin network mechanical function (results not shown), consistent with findings obtained using stress-based measures.

To reconcile the apparent contrast between the age-related increase in stiffness reported for the intact aorta [26] and the decrease in modulus observed for isolated elastin in the present study (Fig. 1c), it is important to consider both the hierarchical nature of arterial mechanics and age-related ECM constituents remodeling. Measurements of intact vessels integrate the combined contributions of multiple matrix components and cells, whereas in this study, testing of purified elastin network probes the behavior of a single structural constituent. Age-related stiffening of the arterial wall is strongly influenced by changes in the relative abundance and organization of collagen and elastic fibers. Quantitative imaging studies have shown that collagen content increases with age, while elastic fiber content decreases, as assessed using normalized autofluorescence and second-harmonic generation imaging [27]. These measurements primarily reflect changes in matrix composition and organization rather than the material properties of the fiber network. Prior work has suggested that arterial remodeling in aging promotes earlier engagement of collagen during loading, shifting more of the mechanical load from the fragmented elastin network to collagen fibers, which are 100–1000 times stiffer than elastin [28]. Within this framework, the present findings further suggest that the mechanical contribution of elastin to arterial mechanics declines with age (Figs. 4 and 5), providing a mechanistic explanation for how arterial stiffening can occur in native tissue despite progressive weakening of the elastin network.

To establish a clearer link between the mechanical behavior and microstructural alterations of the elastin network, we examined these two aspects concurrently by stretching the samples under a multiphoton microscope (Fig. 4 and 5). The observations were particularly revealing. The combined mechanical and microstructural results provide a coherent explanation for the age-dependent decrease in tangent modulus observed in older male samples (Fig. 1c). The microstructural analysis shows that these samples contain highly disrupted and fragmented elastin networks (Figs. 4g-i and Figs. 5d-f), which will lead to fewer load-bearing fibers and reduced network continuity, as the fragmented and coiled fibers cannot bear load and thus compromise load transfer within the network. In contrast, younger, structurally intact networks primarily reorient under load, efficiently transferring load through continuous fiber pathways (Figs. 4a-c and Figs. 5a-c). This correspondence between fiber organization and mechanical behavior suggests that progressive elastin fragmentation with age compromises the structural integrity and load transfer within the network, ultimately contributing to the lower tangent modulus measured in aged male elastin network.

Our findings provide multi-scale structural evidence of age-related elastin deterioration. Across several imaging modalities, we observed progressive fragmentation and loss of structural integrity in elastin fibers with aging. Multiphoton microscopy (Figs. 2 and 3) reveals a transition from organized, dense, and relatively straight fibers in younger samples to increasingly disorganized, recoiled, and heterogeneous networks in older samples, accompanied by a significant decrease in median straightness and increased variability. Histological analysis (Fig. 7) further demonstrates a reduction in lamellar continuity and length with age, indicating fragmentation of the elastin network at the tissue level. These observations are supported by the ultrastructural evidence from scanning electron microscopy (Figs. 8), where older male samples exhibit frayed, discontinuous, and degraded fibers compared to the intact and uniform morphology observed in younger specimens. Elastin undergoes age-related degradation through both enzymatic cleavage and irreversible molecular modifications. Previous studies demonstrated that aging increases elastin susceptibility to enzymatic susceptibility. Human skin elastin from younger individuals was reported to be more resistant to enzymatic cleavage than elastin from older donors, which already exhibited microfissures and cracks [29]. Accordingly, progressive elastin degradation, such as partial fragmentation and fibrillar disintegration, associated with intrinsic aging and photoaging was reported [30]. Subsequent enzymatic digestion revealed increased cleavage in specific elastin domains with age and environmental exposure. Together, these studies show that structural deterioration of elastin fibers enhances protease-mediated fragmentation, a phenomenon consistently observed across various tissues [29, 30, 31, 32, 33]. Non-enzymatic molecular aging further contributes to the vulnerability of elastin fibers to fragmentation and loss of integrity. Even modest glycation can alter elastin conformation, ion binding, and mechanical stiffness, demonstrating that chemical modifications alone can compromise fiber stability [34]. Age-related accumulation of irreversible molecular modifications in elastin, including amino-acid racemization and advanced glycation end products (AGEs), progressively weaken elastin structure and increase susceptibility to fragmentation [35]. Collectively, these studies show that non-enzymatic molecular aging acts together with enzymatic mechanisms to drive the progressive fragmentation and thinning of elastin fibers observed in aging tissues [34, 35, 36, 37]. Taken together, these results support the notion that age-related structural deterioration of elastin occurs across multiple length scales and is consistent with increased susceptibility to both enzymatic degradation and non-enzymatic molecular damage described in prior studies.

Our study showed that female samples exhibited preserved microstructure (Figs. 2e-h, 3a,7e, and 8e-h) and maintained mechanical function (Fig. 1c). Sex-specific differences in elastin degradation have been reported in both animal and human studies. Male mice carrying severe Marfan syndrome mutations exhibit greater aortic stiffening and more pronounced ECM remodeling than females [27]. Similarly, age-related arterial thickening occurs in both sexes but is delayed by roughly a decade in women [38]. Estrogen-related signaling has been proposed as a key protective factor - deletion of the estrogen receptor was reported to abolish sex differences in vascular stiffness, blood pressure, and endothelial function [39]. Female coronary arteries also display inherently greater vasodilatory capacity, mediated in part by nitric oxide-dependent pathways, which reduces mechanical stress on the vessel wall [40]. Collectively, these observations suggest that multiple female-favorable mechanisms, including hormonal and genetic factors, may contribute to preservation of elastin microstructure and mechanical function during aging. However, the contribution of menopausal status and hormonal changes could not be evaluated in the present study and requires further investigation. Future studies are needed to understand how these mechanisms may contribute to the relative maintenance of elastin integrity observed in female specimens. It is also important to note that the relatively small cohort limits our ability to determine whether the observed differences in age-related trends between male and female samples reflect true sex-specific effects. Although the regression trends differed between sexes, the age × sex interaction was statistically significant only for fiber straightness, but not for tangent modulus or lamellar length. These findings should therefore be interpreted with caution, and larger cohorts are needed to establish whether sex-specific differences in age-related elastin remodeling are present.

5. Limitations

Several limitations of this study warrant consideration. Human elastin samples were obtained from both female and male donors, while the sample size was limited (n = 17). The division of samples by sex (9 male and 8 female donors) further limited statistical power for detecting sex-dependent effects; therefore, observed sex-related trends should be interpreted with caution and require validation in larger cohorts. Additionally, menopausal status and related hormonal factors were not available for female donors, limiting interpretation of potential sex-related differences in elastin aging.

Multiphoton imaging was confined to surface regions, with an imaging depth of approximately 60–160 μm, and therefore did not capture potential depth-dependent variations in elastin microstructure across the full aortic wall thickness, which have been reported previously [41]. Donor-specific clinical factors such as hypertension and smoking history, both of which are known to influence aortic structure and mechanical behavior [42], were not analyzed separately and may contribute to inter-sample variability. Other donor-specific factors, including race, medical history, and cause of death, were also not independently analyzed and may contribute to variability in tissue properties. For the in situ loading experiment, the subset of specimens was selected to provide representation across donor ages and sexes while maintaining comparable age distributions between male and female samples; however, the limited number of specimens available for this experiment may not fully capture the biological variability across the broader donor cohort. In addition, the age-dependent thickness analysis for female samples (Fig. 1d) should be interpreted with caution because the oldest female sample (90 years old) had a substantial influence on the regression outcome. Sensitivity analysis showed that exclusion of this individual data point altered the statistical significance of the regression, highlighting the limitation associated with the small sample size and limited representation of advanced ages in the female cohort. Elastin isolation using cyanogen bromide treatment removes associated microfibrillar components [43]; accordingly, the mechanical and microstructural properties reported here primarily reflect the intrinsic properties of the elastin core and may not fully represent the mechanics of intact elastic fiber network in vivo.

Additionally, although long-term storage at −80°C has been reported to have minimal effects on arterial mechanical properties [44], [45], [46], potential effects of variable storage duration across specimens cannot be completely excluded. Uniaxial tensile testing was limited to a maximum stretch of 1.1 because higher stretches frequently resulted in tearing of the purified elastin network, particularly in older samples. This limitation was necessary to maintain specimen integrity across all donors; consequently, the reported mechanical properties primarily characterized the mechanical response of the elastin network in this limited range of deformation. Finally, purified elastin networks were used in this study. Removal of associated microfibrils and potential effects of the purification process may alter network-level mechanical behavior. Thus, the measured properties may not fully represent the mechanical behavior of elastic fiber networks within the native arterial wall.

6. Conclusion

This study provides direct experimental evidence linking age-related microstructural changes in human aortic elastin to alterations in mechanical function. In male specimens, elastin stiffness decreases with age, accompanied by reduced fiber straightness, increased fragmentation, and shorter lamellar lengths, indicating diminished mechanical engagement under tensile loading. In contrast, female elastin exhibited relative preservation of mechanical properties and microstructure, although these observations should be interpreted with caution given the limited sample size and lack of information regarding menopausal status. By integrating mechanical testing with detailed microstructural characterization, this study advances understanding of elastin’s contribution to arterial wall biomechanics and the structural basis of vascular aging. These findings provide a foundation for predictive modeling of elastin behavior under physiological loading and highlight the need for future studies to further elucidate factors that influence elastin integrity. Overall, the results offer important insights into the interplay between microstructure and mechanics in human arterial elastin.

Statement of Significance.

This study provides direct experimental evidence linking age-related microstructural degradation of isolated human elastin networks to impaired mechanical function. Using purified elastin from naturally aged human thoracic aortas, we demonstrate that aging in males is associated with increased elastin fragmentation, fiber waviness, reduced lamellar continuity, and diminished mechanical stiffness, whereas female elastin networks exhibit relative structural and functional preservation. These findings identify deterioration of elastin network integrity as a potential mechanobiological contributor to arterial aging and stiffening. By integrating biomechanics with multiscale imaging of aged human tissues, this work advances understanding of elastin’s role in vascular biomechanics and suggests potential sex-specific differences in elastin aging.