This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cho, A. Articles by Langille, B. L. Search for Related Content PubMed PubMed Citation Articles by Cho, A. Articles by Langille, B. L.

(Circulation Research. 1995;76:168-175.)
© 1995 American Heart Association, Inc.

Articles

Apoptosis (Programmed Cell Death) in Arteries of the Neonatal Lamb

Aesim Cho, David W. Courtman, B. Lowell Langille

From the Banting and Best Diabetes Centre, The Toronto Hospital Research Institute, and the Department of Pathology, University of Toronto, Ontario, Canada.

Correspondence to B. Lowell Langille, PhD, The Toronto Hospital Research Institute, 200 Elizabeth St, Toronto, ON M5G 2C4, Canada.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion Appendix 1 References

 
Abstract We have examined whether cell death contributes to postnatal remodeling of arteries in lambs. First, abdominal aortic smooth muscle cell proliferation rates fell from 2.87±0.08% per day at 3 days of age to 1.75±0.15% per day at 21 days. These proliferation rates would yield a 50% increase in DNA content in the absence of cell death. No increase in DNA content was observed (P<.05 for predicted versus measured accumulation); therefore, significant cell death was inferred. The same analysis did not indicate high cell-death rates in the carotid, renal, or iliac arteries; however, cell death was detected in situ by end-labeling partially degraded DNA with terminal deoxynucleotidyl transferase or by nuclear labeling with propidium iodide, a fluorescent dye that permeates only nonviable cells. Nuclei were labeled in all arteries, although labeling was most frequent in the abdominal aorta, a vessel that regresses substantially after birth. Cell death was apoptotic because DNA extracted from arteries and end-labeled with [³²P]dCTP produced a series of low molecular weight bands (DNA ladder) on an agarose gel, a hallmark of apoptosis. The ladder was strong for neonatal abdominal aorta but weak for other arteries. Only weak laddering was observed for fetal abdominal aortas in late gestation, confirming that high apoptosis rates in this vessel were initiated after birth. Intense DNA ladders and frequent in situ labeling indicated high rates of apoptosis in the postnatal intra-abdominal umbilical artery, another vessel that regresses after birth. We conclude that apoptosis contributes to postpartum arterial remodeling . This contribution is greatest in arteries that regress after birth.

Key Words: cell suicide • physiological cell death • arterial remodeling

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion Appendix 1 References

 
Apoptosis (programmed cell death, physiological cell death) is cell suicide, apparently controlled by a balance of cell-survival and cell-death signals,¹ that contributes to many fundamental biological processes, including morphogenesis,^{2 3 4} remodeling of mature tissues,^{5 6 7} and negative selection of T lymphocytes.⁸ Apoptosis differs morphologically from necrosis: cells exhibit cytoplasmic and nuclear condensation, and then they fragment into apoptotic bodies without membrane disruption. Apoptotic bodies are phagocytosed by neighboring cells or macrophages without inducing an inflammatory response.^{1 9} Nuclear condensation apparently results from release of endonucleases that degrade DNA into large fragments of 50 to 300 kb^{10 11} and then into smaller fragments of 180 to 200 bp because of internucleosomal cleavage.¹

Apoptosis has been studied extensively in embryogenesis. It is responsible for digit formation in the developing limb bud,¹² the loss of 50% of developing central neurons,³ and the early organization of renal tissues in the metanephros.² Despite this emphasis, recent evidence indicates that apoptosis in developmental tissue remodeling has been underestimated, mostly because of the difficulty in detecting the dying cells. Detection is difficult because apoptosis is identifiable, either morphologically or biochemically, for only 1 to 3 hours.^{3 13}

The short duration of detectability has masked relatively high rates of apoptosis during embryonic morphogenesis³ ; therefore, its role in much slower rates of tissue remodeling during later development has been unexplored. We became interested in the role of cell death in the remodeling of arteries that occurs as they adapt to perinatal changes in cardiovascular function. Our interest was stimulated by data indicating that DNA accumulation in the abdominal aorta of neonatal lambs was much below that predicted from mitosis rates, if all cells were assumed to survive.¹⁴ The abdominal aorta undergoes particularly extensive remodeling in association with a 95% decrease in blood flow when the placenta is lost at birth.^{15 16} However, virtually all other large arteries undergo very large changes in blood flow in the perinatal period, and available data indicate that substantial remodeling also occurs in these vessels.¹⁵

The present study tested for direct evidence that programmed cell death occurs in large arteries perinatally. Several independent techniques confirm high cell-death rates, and our data implicate apoptosis as a mode of cell death. Apoptosis rates were particularly high in the abdominal aorta after, but not before, birth. This finding implicates apoptosis in the pronounced postnatal remodeling of this artery that occurs after birth.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion Appendix 1 References

 
Arterial Smooth Muscle Cell Replication
Lambs aged 3, 12, and 21 days (five lambs at each age) were given intramuscular injections of [³H]thymidine (500 µCi/kg) at 17, 9, and 1 hour before they were killed by intravenous injection of euthanasia solution, 200 mg/mL N-[2-m-methophenyl-2-ethylbutyl-(1)]-2 hydroxybutyramide, 50 mg/mL 4,4'-methylene bis(cyclohexyltrimethylamonium iodide), and 5 mg/mL tetracaine hydrochloride (T-61, Hoechst Canada, Inc). The heart and central arteries were exposed by performing a bilateral thoracotomy, and the ascending aorta was cannulated. The arterial system was fixed by perfusion with 1% paraformaldehyde and 1% glutaraldehyde in phosphate buffer for 30 minutes at a pressure of 70 mm Hg for 3-day-old lambs and 100 mm Hg for 12- and 21-day-old lambs. Arterial tissues (aorta and the iliac, superior mesenteric, carotid, and renal arteries) were removed and immersed in fixative for 18 to 24 hours. Unbranched segments were embedded in paraffin, and then 5-mm sections were deparaffinized in xylene, air-dried overnight, dipped in Kodak NTB-2 emulsion (Eastman Kodak), and stored at 4°C for 2 weeks in total darkness. The slides were developed in Kodak D-19 developer and stained with Mayer's hematoxylin (Sigma Chemical Co). Labeled (replicating) and unlabeled cells were counted to determine the percentage of smooth muscle cells that were replicating.

Arterial Tissue DNA Content
The same arteries listed above, but from different lambs, were harvested between well-defined anatomic landmarks at 3 and 21 days after birth. Ten lambs were studied at each age. The iliac artery was excised between the aorta and the origin of the external iliac artery, the renal arteries were removed from the aorta to a bifurcation site just proximal to the hilus of the kidney, the abdominal aorta was excised between the left renal artery and the iliac bifurcation, and the left carotid artery was removed from its origin to its termination. All vessels were cleared of adventitia. The vessel segments were homogenized, and DNA content of each artery was measured by a fluorometric assay.¹⁷ This method exploits the enhancement by tissue DNA of the fluorescence of bisbenzimidazole (Hoechst 33258, Sigma), which is then compared with DNA standards.

DNA contents at 3 days of age and cell replication data were used to predict net accumulation of DNA that would occur between 3 and 21 days of age in the absence of cell death. We assumed that DNA content is proportional to cell number, since the great majority of vascular cells are diploid in normotensive animals.¹⁸ Then DNA content [D(t)] at any time (t) after an initial measurement (D_i) is given by the following differential equation:

(1)

where the time dependence of k(t) allows for age-dependent cell replication rates. The general solution for this equation is as follows:

(2)

Noting that the mean value for k(t) over 0 to t, <k>, is given by

(3)

Equation 2 becomes

(4)

Equation 4 is identical to the solution for purely exponential growth, except that a constant, k, is replaced by the mean value for k(t). For our particular case, DNA content at 21 days of age is as follows:

(5)

where <k> must be evaluated. According to Equation 1,

(6)

For finite t, we define

(7)

Since we assume

(8)

where N(t) is cell number, it follows that (t, t=1) equals the fractional cell replication rate per day (percent cells replicating per day÷100) that was measured experimentally. (t, t=1) is a good approximation of k(t), provided that replication rates are sufficiently low. The fractional replication rates in the present study were consistently <0.05, for which the error inherent in the above approximation is on the order of 2.5% (see "Appendix"). Thus, measured replication rates, (t, t=1), were averaged to estimate <k>, and Equation 5 was used to estimate DNA content at 21 days. The methods for averaging (t, t=1) and the statistical techniques used to make comparisons with measured DNA content at 21 days of age are given in "Statistical Analysis."

In Situ DNA End-Labeling of Apoptotic Cells
The method for end-labeling apoptotic cells was adapted from Gavrieli et al.¹⁹ It is based on the preferential binding of terminal deoxynucleotidyl transferase to 3'-OH ends of DNA. Six-micrometer sections of arteries from six lambs were deparaffinized by heating at 60°C and transferred to xylene and then hydrated in a descending alcohol series (100%, 95%, 75%, and 0%). The sections were incubated with 20 µg/mL proteinase K (Boehringer Mannheim) for 15 minutes at room temperature, and endogenous peroxidase was inactivated by treating with 2% H₂O₂. The tissue sections then were incubated with biotin- ylated dUTP (0.3 pmol/µL) and 0.3 U/µL terminal deoxynucleotidyl transferase (Boehringer Mannheim) in TDT buffer (mmol/L: sodium cacodylate 140, CoCl₂ 1, and Tris-HCl 30 at pH 7.2) for 1 hour at 37°C. After end-labeling, the sections were treated with 2% bovine serum albumin for 10 minutes, followed by a 30-minute incubation at 37°C with extra-avidin peroxidase (Sigma). Aminoethylcarbazole was used as a chromogen to detect the biotin-labeled nuclei.

DNA Isolation and Detection of Oligonucleosomes
The method, based on the same principle as in situ end-labeling, was modified from Rosl.²⁰ Initially, DNA was extracted from arterial tissues of nine newborn lambs and purified by using a standard extraction method. Briefly, tissues were cut into very small pieces and incubated overnight in DNA lysis buffer (20 mmol/L Tris-HCl [pH 7.4], 1% sodium dodecyl sulfate, 5 mmol/L EDTA, and 100 µg/mL proteinase K) at 50°C. The DNA was extracted with salt-saturated phenol and chloroform and precipitated with 100% ethanol.

Five micrograms of arterial tissue DNA or 1 µg of 100-bp DNA marker (GIBCO BRL Life Technologies Inc) was incubated with 10 µCi of [³²P]dCTP (ICN Biomedical Canada Ltd) and 10 U of Klenow polymerase (Pharmacia Biotech Inc) for 15 minutes at 30°C. The end-labeling reaction was terminated by addition of 10 mmol/L EDTA. Unincorporated nucleotides were removed by using a Magic DNA clean-up system (Promega Corp) following the manufacturer's instructions.

Radiolabeled DNA was electrophoresed on a 1.8% agarose gel and blotted onto Hybond nylon membrane, and the membrane was used to expose Kodak X-OMAR X-ray film for 1 to 2 hours.

Since we hypothesized that apoptosis may be a feature of postnatal remodeling of the abdominal aorta, we also prepared autoradiographs of gels after running DNA from abdominal aortas of six late-gestation fetuses (134 days of gestation; full term, 145 days). Results were compared with autoradiographs of gels prepared with abdominal aortic DNA from six 3-day-old lambs.

In Vivo Labeling of Nonviable Cells
Three lambs, at 3 days of age, were anesthetized by intramuscular injection of 0.08 mL/kg xylazine (20 mg/mL) plus 0.72 mL/kg ketamine (100 mg/mL) and then given intravenous injections of 5 µmol/kg propidium iodide (Calbiochem-Novabiochem Corp). After 15 minutes, the lambs were killed by intravenous injections of T-61. The ascending aorta was catheterized as described above, and the arterial system was flushed with 50 mL of Tyrode's and perfusion-fixed with 3% phosphate-buffered paraformaldehyde for 30 minutes. Arteries were harvested, mounted onto glass slides with glycerol/phosphate-buffered saline, and viewed under a laser-scanning confocal microscope (Bio-Rad MRC 600). Cells that did not exclude the propidium iodide, as indicated by intense nuclear fluorescence, were considered nonviable.

Statistical Analysis
ANOVA was used to test for significant differences in cell replication rates among different arteries. No differences were indicated, so no multiple comparison tests were required. Unpaired t tests were used to test for significant changes in cell replication rates between 3 and 21 days after birth. Unpaired t tests were also used to test for significant differences in cell death rates of 3-day-old abdominal versus thoracic aortas, as determined from counts of cells end-labeled by using terminal deoxynucleotidyl transferase.

A Monte Carlo simulation,²¹ based on Equation 5, was used to test for significant differences between observed DNA contents at 21 days of age and DNA contents predicted on the assumption of no cell death. First, measured means and standard deviations for DNA contents at 3 days and for cell replication rates at 3, 12, and 21 days of age were assumed to represent means and standard deviations of normally distributed populations. MINITAB software (Minitab Inc) was then used to randomly select data sets from these populations that were equal in number to experimental sample sizes. Mean values for D(3) were substituted directly into Equation 5. In addition, each set of values for (3, t=1), (12, t=1), and (21, t=1) was fitted to a quadratic, and <k> was calculated, according to Equation 3, by integrating the quadratic. This value for <k> also was substituted into Equation 5, and D(21) was estimated according to the equation. The sampling process and calculations were repeated 2000 times to construct a population of values for D(21). The measured value for mean DNA content at 21 days of age was inserted into this population, and the P value for comparison of predicted versus measured DNA contents was set to

(9)

where Ñ is the number of entries in the population, including the measured DNA content, that are equal to or greater than the measured DNA content.

	Results

Top Abstract Introduction Materials and Methods Results Discussion Appendix 1 References

 
Arterial Smooth Muscle Cell Proliferation Rates and DNA Accumulations
Smooth muscle cell replication rates are presented in Fig 1. The decreases in cell replication rates between 3 and 21 days of age were significant for all arteries studied (unpaired t tests, P<.05).

View larger version (16K): [in this window] [in a new window]   Figure 1. Graph showing medial smooth muscle cell replication rates in abdominal aortas and renal, iliac, carotid, and superior mesenteric arteries of 3-, 12-, and 21-day-old lambs. Values are expressed as percent per day.

Replication rates in the abdominal aorta decreased steadily from 2.87±0.08% to 1.75±0.15% between 3 and 21 days after birth. On the assumption that no cell death occurs during the 18-day period, a net increase in DNA of 50% was predicted; however, we observed no DNA accumulation despite the relatively high cell proliferation rate (Table 1). The difference between predicted and observed DNA accumulations was statistically significant (P<.05). In contrast, no significant difference between predicted and observed DNA accumulations was observed in the distributing arteries that we assessed: the renal, carotid, and iliac arteries.

View this table: [in this window] [in a new window]   Table 1. Measured Versus Predicted DNA at 21 Days of Age

It was noteworthy that there were only minor differences between arteries in cell replication rates despite large differences in DNA accumulation (0% to 80%). Replication rates averaged over the experimental time period (3 to 21 days of age) were 2.28% per day for the abdominal aorta, 2.67% per day for the renal artery, 2.29% per day for the carotid artery, and 2.09% per day for the iliac artery. We also measured replication rates in the superior mesenteric artery. Cells in this vessel divided at a rate that averaged 3.64% per day over the 21-day period. These time-average replication rates could not be compared statistically because they were derived from a nonhomogeneous population due to age dependence. However, the time-averaged values for each artery were all very close to replication rates for the same vessel at the midpoint of the experimental period (12 days of age). At this age, there were no significant differences in replication rates for different arteries (one-way ANOVA, P>.05).

In Situ End-Labeling of Nonviable Cells
In all arteries examined, we detected cells positively stained by in situ end-labeling of fragmented DNA by using terminal deoxynucleotidyl transferase and biotinylated dUTP. These cells were rare, presumably because of the brief duration of apoptosis, and they occurred more often in the abdominal aorta than in the thoracic aorta or in the iliac or renal arteries. Most frequently, isolated cells were detected (Fig 2A), although islands of 2 to 10 stained cells were sometimes seen (Fig 2B). We noted no consistent pattern of distribution of labeled cells; however, their incidence was sufficiently rare to preclude easy recognition of such a pattern.

View larger version (0K): [in this window] [in a new window]   Figure 2. Photomicrographs showing nuclei with fragmented DNA that was detected by in situ DNA end-labeling using terminal deoxynucleotidyl transferase and biotinylated dUTP. A and B, Abdominal aorta from 3-day-old lamb. Apoptotic nucleus is indicated by arrowhead. C, Intra-abdominal umbilical artery from 3-day-old lamb. Many nuclei in this regressing vessel are labeled. Bars=10 µm.

We quantified the difference in cell-labeling rates for the postnatal thoracic and abdominal aortas. Thus, 0.058±0.036% of abdominal aortic smooth muscle cells versus 0.0030±0.0018% of thoracic aortic cells were labeled (P<.005). It is noteworthy that even though labeling of abdominal aortic cells is 20-fold higher than thoracic aortic labeling rates, the labeling indices for this vessel are still only a small fraction (1/30th to 1/40th) of daily cell replication rates, as measured by uptake of [³H]thymidine. Low labeling probably reflects the short duration of apoptosis and the likelihood that cells contain fragmented DNA in sufficient quantities for detection of labeling for a portion of this time.

To assess arterial sites where cell death may be more common, we end-labeled segments of the intra-abdominal umbilical artery from two lambs at 3 days of age. This artery undergoes a rapid regression soon after birth. Large numbers of cells in the inner media of the umbilical artery were intensely stained when end-labeled with biotinylated dUTP (Fig 2C).

Positive controls indicated that we were correctly detecting fragmented DNA and cell death. DNAse treatment of normal arteries to promote end-labeling resulted in a high number of positive cells (data not shown). Also, in small intestine, epithelial cells mature as they migrate from the crypt and eventually undergo apoptotic cell death at the villus tip.^{1 19} When sections of small intestine from postnatal lamb were end-labeled, cells at the villus tip showed intense staining, indicating the presence of DNA fragmentation in dying cells (data not shown).

Detection of DNA Fragmented Into Oligonucleosomes
When DNA extracted from the abdominal aortas of 3-day-old lambs was end-labeled with [³²P]dCTP and electrophoresed on an agarose gel, DNA fragments in multiples of 200 bp were detected (Fig 3A). These data indicate cleavage of DNA at internucleosomal sites. DNA extracted from the umbilical arteries of 3-day-old lambs also demonstrated this laddering phenomenon on an agarose gel, whereas DNA extracted from thoracic aorta or distributing arteries showed much fainter laddering. Furthermore, much weaker signals from oligonucleosome ladders were detected from fetal abdominal aortas when compared with the neonatal vessels (Fig 3B).

View larger version (71K): [in this window] [in a new window]   Figure 3. A, Autoradiographs of DNA gel showing fragmented DNA, end-labeled with [32P]dCTP, from arteries of 3-day-old lambs. UA indicates umbilical artery; TA, thoracic aorta; and AA, abdominal aorta. Fragmentation of DNA into oligonucleosomes, as evidenced by a ladder pattern in autoradiographs of DNA gels, was seen in AA and UA but not in TA. B, Autoradiographs of DNA gel showing fragmented DNA, end-labeled with [32P]dCTP, from AAs of two sheep fetuses at 134 days of gestation and two lambs at 3 days after birth. A ladder pattern, indicative of apoptosis, was observed for neonatal (3-day-old) AA. A much fainter oligonucleosome ladder was observed when using fetal AA at 134 days of gestation. This observation is evidence that apoptosis is upregulated during postnatal remodeling of the AA.

In Vivo Detection of Nonviable Cells
Nonviable cells were detected by intravenous infusion of the membrane-impermeable DNA binding dye, propidium iodide. The distribution of positive cells was highly reminiscent of the distribution observed after end-labeling with terminal deoxynucleotidyl transferase. Nonviable smooth muscle cells were rare, but they were detected in all vessels examined. Most positively stained cells were isolated (Fig 4A), but localized clusters of labeled cells were detected occasionally (Fig 4B). Some of these cells showed normal ellipsoidal nuclear morphology that resembled the shape of the nuclei of viable cells that were stained after membrane permeabilization (not shown). In other nonviable cells, staining did not reveal normal nuclear morphology; instead, DNA was condensed into a globular pattern (Fig 4A) suggestive of the chromatin condensation, a characteristic of apoptosis.^{1 9} Nonviable endothelial cells also were detected by this technique. Also, many cells, both smooth muscle and endothelium, were labeled in postnatal intra-abdominal umbilical arteries.

View larger version (55K): [in this window] [in a new window]   Figure 4. In vivo detection of nonviable cells in postnatal arteries. Nonviable cells were detected by fluorescent staining using the DNA-binding fluorescent dye propidium iodide, which cannot permeate the membranes of viable cells. Abdominal aortas from 3-day-old lambs reveal individual apoptotic cells (A) or cells in clusters (B). Fluorescence in panel A shows a pattern reminiscent of chromatin condensation at the perimeter of the nucleus that is produced by apoptosis; however, this was not seen in all cases. Bars=10 µm.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion Appendix 1 References

 
There has been much interest in the control of arterial tissue growth and remodeling during development, in part because of its relation to pathological processes including hypertension and atherosclerosis.^{22 23} Much of this interest has been directed toward growth factors that control vascular cell mitosis rates.²² Our data indicate that the growth of cell populations in arteries also depends on rates of cell death. Cell death in the neonatal abdominal aorta was implicated by DNA accumulation rates that were inconsistent with cell mitosis rates. Previously, Bendeck and Langille¹⁴ reported a similar inconsistency; however, those experiments relied on postmortem determinations of arterial cell replication rates from in vitro thymidine uptake. In the present study, cell replication rates were based on in vivo thymidine incorporation. Also, several additional lines of evidence supported the inference of cell death and implicated apoptosis, including in situ and biochemical evidence of DNA fragmentation and failure to exclude the fluorescent dye propidium iodide.

The concept of apoptosis arose from observations that many instances of cell death that serve a physiological or developmental function display a specific sequence of morphological changes that leads to cell fragmentation and phagocytosis by neighboring cells or macrophages. Apoptosis differs from necrosis, which occurs when cells are lethally injured by trauma or toxins. Necrosis is characterized by swelling of organelles and the cytoplasm, membrane rupture, and disintegration of the cell. Apoptosis produces compaction and segregation of chromatin at the periphery of the nucleus and condensation of the cytoplasm, followed by nuclear fragmentation and the budding off of intact cell fragments (apoptotic bodies) that are phagocytosed by neighboring cells or macrophages.

Recently, however, there have been reports of physiological cell death that is not apoptotic²⁴ and of variant forms of apoptosis. Variant forms of apoptosis include, for example, those in which DNA is cleaved into large fragments (50 to 300 kb) but not into relatively short oligonucleosomes.¹¹ Consequently, we examined postnatal arteries for evidence of cell death and to determine whether such death displayed characteristics of apoptosis.

We attempted to detect apoptosis by transmission electron microscopy (TEM), but the results proved inconclusive. Some cells displayed morphological features that could be interpreted as apoptotic, but we did not detect unequivocal chromatin condensation or apoptotic bodies. It is possible that vascular cells follow a normal biochemical pathway to apoptosis without an obvious display of many features of apoptotic morphology. However, we favor an alternative explanation, ie, that the sampling limitations of TEM render detection of apoptotic morphology improbable because of the brief duration of this morphology and the relative rarity of apoptosis. In many apoptotic cell populations, a large fraction of the cells die rapidly; thus, even very short-lived morphological changes can be detected. By contrast, death rates of the cells in the slowly remodeling arteries we studied were generally below mitosis rates of 2% per day (DNA content increased in most vessels). Since apoptotic morphology persists for only 1 to 3 hours,^{3 13} <1 cell in 1000 would be detected by TEM. One reason for favoring this explanation was that fluorescence labeling of DNA in nonviable cells frequently showed condensation of DNA into discrete packages reminiscent of apoptotic chromatin condensation seen by TEM in other cells. These features were detectable with the laser-scanning confocal microscope, because large volumes of tissue could be scanned quickly with this device.

We confirmed apoptotic cell death, because gel electrophoresis of DNA from neonatal arteries yielded a ladder pattern with bands at multiples of 180 to 200 bp. This pattern is a widely accepted marker of apoptosis that has been observed in many systems.¹ Apoptosis can occur in rare circumstances without producing these small molecular weight DNA fragments, apparently because the morphological features that define apoptosis coincide with DNA cleavage into larger fragments of 50 to 300 kb¹¹ ; nonetheless, the endonuclease that generates the smaller fragments is not associated with nonapoptotic processes. Detection of high numbers of dying cells, by in situ end-labeling and by in vivo fluorescent labeling with propidium iodide, in those vessels that yielded intense DNA ladders (postnatal abdominal aorta and intra-abdominal umbilical arteries) indicates that apoptosis contributes to the pattern of cell death observed in perinatal arteries, although other modes of cell death may also occur.

The high rate of programmed cell death in the abdominal aorta provides an illustrative example of the role of this process in the remodeling of arteries that occurs in response to altered function. The abdominal aorta goes into virtual growth arrest over the perinatal period examined in the present study, and its diameter decreases substantially.¹⁵ These phenomena are related to a very dramatic decrease in blood flow, by >90%, at parturition that is associated with loss of the placenta. The placenta, which receives almost half of the combined output of both ventricles in the fetus, is supplied by umbilical arteries that arise at the termination of the aorta. It is possible that apoptosis in this vessel, and others, is influenced by hemodynamic signals that regulate other aspects of arterial growth.^{14 15} Low apoptosis rates in fetal abdominal aorta and the intra-abdominal umbilical artery support our inference that postnatal apoptosis is related to vessel remodeling after birth. All other postnatal arteries we examined yielded evidence of apoptosis but at a much lower frequency than in these arteries.

Although the discrepancy between cell replication rates and DNA accumulation first directed our attention toward cell death, the cell replication rates also were striking for their consistency among different arteries. These data suggest that systemic factors may control vascular cell replication rates in large arteries. The striking differences in DNA accumulation in different arteries, despite an invariant mitosis rate, indicate that cell death rates may determine differential growth of cell populations in different arteries.

We observed endothelial cell and smooth muscle cell apoptosis, both of which may contribute to remodeling of arteries. Endothelial cell apoptosis has been reported previously in vessels of the regressing corpus luteum,⁵ and we have some unpublished evidence that it occurs in endothelium of arteries that remodel in response to chronic reductions in blood flow. In general, endothelial cell populations respond to changes in vessel size by restoring a normal density of the cells on the vessel surface.²⁵ Thus, cells are deleted from vessels that are decreasing in size, like those supplying the regressing corpus luteum, or that are carrying reduced blood flow for other reasons. Apoptosis may be a common means of remodeling the endothelium to restore the initial status of the monolayer after vessel diameter reduction.

The functional significance of apoptosis of vascular smooth muscle cells probably is more complex and also more critical in determining arterial structure. Localized cell death will modulate regional smooth muscle populations, and since these cells make virtually all matrix components in the media, all constituents of the media may be affected. Thus, coordinated controls of mitosis and apoptosis are probably critical in reshaping the arterial system during development. In postnatal arteries, this reshaping involves changes in the arterial lengths, radii, and wall thicknesses. Earlier remodeling includes complete loss of many vessels and dramatic reorganization of others, with relative positions of branch sites changing and even bypassing their neighbors. In mammals, the latter events occur, for example, as the aortic origin of the primitive left subclavian artery bypasses the junction of the aorta and ductus arteriosus.²⁶ Other examples in vertebrate development are commonplace (eg, see Willemse and Markus-Silvis²⁷ ). The importance of apoptosis in such remodeling is unknown.

Evidence from other systems confirms a morphogenetic role for apoptosis. For example, cell death defines digit formation in embryonic limb buds,¹² neuronal and oligodendricyte cell death directs the formation of appropriate cell-cell connections within the central nervous system,³ and epithelial cell apoptosis is important in involution of the prostate after castration.⁶ The importance of apoptosis is observed across many species. Apoptosis plays an important part in the development of the body plan of Caenorhabditis elegans, since approximately one seventh of all cells that form during development of this nematode die before maturation.^{28 29} Furthermore, immunolocalization in fetuses of an important inhibitory regulator of apoptosis, bcl-2, suggests an important role in many aspects of morphogenesis.³⁰

The importance of apoptosis in the mature vasculature is unknown; however, there has been much recent interest in the reexpression of developmental growth controls in adult vascular pathologies. There is some evidence that cell death may influence these disorders. Intimal thickening that follows arterial endothelial trauma ultimately ceases, for unknown reasons, even though luminal smooth muscle cells continue to divide at a high rate.³¹ This observation implies that cell death is compensating for ongoing proliferation and preventing ultimate occlusion of the artery. Other vascular trauma can induce transient overproliferation during repair that ultimately resolves.^{32 33} The deletion of the excess cells is poorly understood, but it may be important in the progression of sequelae to vascular injury.

Apoptosis in both normal and pathological vessel remodeling can be investigated by mapping the phenomena as arteries restructure; however, this is not straightforward. The brief duration of detectable apoptosis and a relatively low daily apoptosis rate in neonatal arteries preclude statistical mapping using most available methods of detection. TEM is too time consuming and provides too small a sampling volume for these studies, and it is reliant on three-dimensional reconstructions from tissue sections. End-labeling fragmented DNA in situ is slightly more efficient, but it is also time consuming and reliant on three-dimensional reconstruction. Uptake of fluorescent labels may be the most promising approach, since relatively large volumes of tissue in whole-mount preparations can be examined rapidly by fluorescence laser scanning confocal microscopy.

Although our data suggest that local regulators of cell death may be very important in arterial development, the nature of these regulators in vascular tissues is unknown. Transforming growth factor-ß enhances³⁴ and platelet-derived growth factor and insulin-like growth factor inhibit cell death in some tissues^{3 34} ; however, their effects on apoptosis in arteries are unknown. Likewise, a host of regulators of apoptosis have been implicated in other systems, but their effect in the vasculature has not been studied. Members of the bcl-2 family are particularly important,³⁵ but p53^{36 37} and steroid hormones³⁸ also deserve examination in vascular tissue apoptosis. Recently, interleukin-converting enzyme, which displays significant sequence identity to ced-3, a cell death gene in C elegans, has been implicated in control of cell death in mammalian cells.^{39 40} Finally, very recent evidence indicates that cultured vascular smooth muscle cells can be driven into apoptosis by growth inhibition if downregulation of c-myc is prevented.⁴¹

In summary, the present study demonstrates that apoptosis occurs in neonatal arteries and that cell death rates approach mitosis rates in some arteries. Vessel-specific differences in DNA (cell) accumulation probably are attributable largely to differences in cell death rates, since mitosis rates were consistent from vessel to vessel. Our findings indicate that programmed cell death is a potentially important mechanism for developmental remodeling of blood vessels.

	Acknowledgments

 
This study was supported by the Medical Research Council of Canada (grant PG-11115). Dr Langille is a Career Investigator of the Heart and Stroke Foundation of Ontario and A. Cho is a trainee of the Heart and Stroke Foundation of Canada. Mr Peter Lewycky, biostatistician at the Toronto Hospital Research Institute, provided statistical consultation and analysis on a fee-for-service basis.

	Appendix 1

Top Abstract Introduction Materials and Methods Results Discussion Appendix 1 References

 
We demonstrate that the replication rates we measured over a 24-hour period [(t, t=1)] are sufficiently low to provide a reliable estimate of k(t) in Equation 1. Also, we provide a brief description of how to predict DNA accumulations, assuming no cell death, when replication rates are not low. Substituting Equation 2 into Equation 7 yields the following:

(10)

which reduces to

(11)

Expanding the exponential term as a power series gives

(12)

Thus,

(13)

For our data, (t, t=1), the fraction of cells replicating in a given day, is consistently <0.05. Since all terms on the right hand side (RHS) of Equation 13 are positive, we also have k(t)dt<0.05. It follows that the second-order term is <0.00125, ie, <2.5% of the first-order term. Consequently, this term and higher order terms are neglected and

(14)

The RHS of Equation 14 is the mean of k(t) over 1 day. Thus, (t, t=1) is a good measure of the mean value of k(t) for the day if and only if the replication rate is low. If (t, t=1) is not low, then (t, t=1) overestimates k because of the higher order terms in Equation 13.

In the event that replication rates are too high to neglect higher order terms, it is still feasible to predict DNA accumulation (or cell population growth), assuming no cell death, based on time-varying cell replication rates. By definition,

(15)

Substituting the RHS of Equation 2 into Equation 15 yields

(16)

Coefficients can be derived for a polynomial fit to k(t), which can then be substituted into Equation 2 to solve for D(t). We will demonstrate this for the case of quadratic fits. Assume for the moment that coefficients exist; ie,

(17)

for some a, b, and c. Substituting Equation 17 into Equation 16 yields

(18)

Therefore,

(19)

Consequently, if raw data for (t, t=1) are converted to the logarithmic function on the left hand side (LHS) of Equation 19 and fitted to the quadratic

(20)

then, by equating coefficients in Equations 19 and 20, we have

(21)

The LHS of Equation 20, and hence the coefficients , ß, , a, b, and c, always exist, since (t, t=1) is always positive. An explicit expression for Equation 17, in terms of the experimentally derived values , ß, and , can now be substituted into Equation 2 to calculate D(t).

Received May 6, 1994; accepted October 6, 1994.

	References

Top Abstract Introduction Materials and Methods Results Discussion Appendix 1 References

 

1. Kerr JFR, Harmon BV. Definition and incidence of apoptosis: an historical perspective. In: Tomei LD, Cope FO, eds. Apoptosis: The Molecular Basis of Cell Death. Plainview, NY: Cold Spring Harbor Laboratory Press; 1991:5-29.
2. Koseki C, Herzlinger D, Al-Awqati Q. Apoptosis in metanephric development. J Cell Biol. 1992;119:1327-1333. [Abstract/Free Full Text]
3. Raff MC, Barres BA, Burne JF, Coles HS, Ishizaki Y, Jacobson MD. Programmed cell death and the control of cell survival: lessons from the nervous system. Science. 1993;262:695-700. [Abstract/Free Full Text]
4. Koseki C. Cell death programmed in uninduced metanephric mesenchymal cells. Pediatr Nephrol. 1993;7:609-611. [Medline] [Order article via Infotrieve]
5. Azmi TI, O'Shea JD. Mechanism of deletion of endothelial cells during regression of the corpus luteum. Lab Invest. 1984;51:206-217. [Medline] [Order article via Infotrieve]
6. Buttyan R. Genetic response of prostate cells to androgen deprivation: insights into the cellular mechanism of apoptosis. In: Tomei LD, Cope FO, eds. Apoptosis: The Molecular Basis of Cell Death. Plainview, NY: Cold Spring Harbor Laboratory Press; 1991:157-173.
7. Polunovsky VA, Chen B, Henke C, Snover D, Wendt C, Ingbar DH, Bitterman PB. Role of mesenchymal cell death in lung remodeling after injury. J Clin Invest. 1993;92:388-397.
8. Owens GP, Hahn WE, Cohen JJ. Identification of mRNAs associated with programmed cell death in immature thymocytes. Mol Cell Biol. 1991;11:4177-4188. [Abstract/Free Full Text]
9. Wyllie AH, Kerr JFR, Currie AR. Cell death: the significance of apoptosis. Int Rev Cytol. 1980;68:251-306. [Medline] [Order article via Infotrieve]
10. Brown DG, Sun X-M, Cohen GM. Dexamethasone-induced apoptosis involves cleavage of DNA to large fragments prior to internucleosomal fragmentation. J Biol Chem. 1993;268:3037-3039. [Abstract/Free Full Text]
11. Oberhammer F, Wilson JW, Dive C, Morris ID, Hickman JA, Wakeling AE, Walker PR, Sikorska M. Apoptotic death in epithelial cells: cleavage of DNA to 300 and/or 50 kb fragments prior to or in the absence of internucleosomal fragmentation. EMBO J. 1993;12:3679-3684. [Medline] [Order article via Infotrieve]
12. Hammar SP, Mottet NK. Tetrazolium salt and electron-microscopic studies of cellular degeneration and necrosis in the interdigital areas of the developing chick limb. J Cell Sci. 1971;8:229-251. [Abstract/Free Full Text]
13. Bursch W, Paffe S, Putz B, Barthel G, Schulte-Hermann R. Determination of the length of the histological stages of apoptosis in normal liver and in altered hepatic foci of rats. Carcinogenesis. 1992;11:847-853. [Abstract/Free Full Text]
14. Bendeck MP, Langille BL. Rapid accumulation of elastin and collagen in the aortas of sheep in the immediate perinatal period. Circ Res. 1991;69:1165-1169. [Abstract/Free Full Text]
15. Langille BL, Brownlee RD, Adamson SL. Perinatal aortic growth in lambs: relation to blood flow changes at birth. Am J Physiol. 1990;259:H1247-H1253. [Abstract/Free Full Text]
16. Bendeck MP, Langille BL. Changes in blood flow distribution in the perinatal period in fetal sheep and lambs. Can J Physiol Pharmacol. 1992;70:1576-1582. [Medline] [Order article via Infotrieve]
17. Labarca C, Paigen K. A simple, rapid and sensitive DNA assay procedure. Anal Biochem. 1980;102:344-352. [Medline] [Order article via Infotrieve]
18. Owens GK, Rabinovitch PS, Schwartz SM. Smooth muscle cell hypertrophy versus hyperplasia in hypertension. Proc Natl Acad Sci U S A. 1981;78:7759-7763. [Abstract/Free Full Text]
19. Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol. 1992;119:493-501. [Abstract/Free Full Text]
20. Rosl F. A simple and rapid method for detection of apoptosis in human cells. Nucleic Acids Res. 1992;20:5243. [Free Full Text]
21. Manly BFJ. Randomization and Monte Carlo Methods in Biology. London, England: Chapman and Hall; 1991:1-281.
22. Schwartz SM, Heimark RL, Majesky MW. Developmental mechanisms underlying pathology of arteries. Physiol Rev. 1990;70:1177-1209. [Abstract/Free Full Text]
23. Keeley FW, Johnson DJ. The effect of developing hypertension on the synthesis and accumulation of elastin in the aorta of the rat. Biochem Cell Biol. 1986;64:38-43. [Medline] [Order article via Infotrieve]
24. Schwartz LM, Smith SW, Jones MEE, Osborne BA. Do all programmed cell deaths occur via apoptosis? Proc Natl Acad Sci U S A. 1993;90:980-984. [Abstract/Free Full Text]
25. Langille BL, Bendeck MP, Keeley FW. Adaptations of carotid arteries of young and mature rabbits to reduced carotid blood flow. Am J Physiol. 1989;256:H931-H939. [Abstract/Free Full Text]
26. Heuser CH. The branchial vessels and their derivatives in the pig. Contrib Embryol. 1923;15:121-139.
27. Willemse JJ, Markus-Silvis L. The shifting of the aortic origin of the brachial arteries in the metamorphosing eel Anguilla anguilla (L.), with remarks on the shifting mechanisms in arterial junctions in general. Acta Anat (Basel). 1985;121:216-222. [Medline] [Order article via Infotrieve]
28. Sulston JE, Horvitz HR. Post-embryonic cell lineages of the nematode Caenorhabditis elegans. Dev Biol. 1977;82:110-156.
29. Suston JE, Schierenberg E, White JG, Thomson N. The embryonic cell lineage of the nematode Caenorhabditis elegans. Dev Biol. 1983;100:64-119. [Medline] [Order article via Infotrieve]
30. LeBrun DP, Warnke RA, Cleary ML. Expression of bcl-2 in fetal tissues suggests a role in morphogenesis. Am J Pathol. 1993;142:743-753. [Abstract]
31. Clowes AW, Reidy MA, Clowes MM. Kinetics of cellular proliferation after injury, I: smooth muscle cell growth in the absence of endothelium. Lab Invest. 1983;49:327-333. [Medline] [Order article via Infotrieve]
32. Reidy MA, Schwartz SM. Endothelial regeneration, III: time course of intimal changes after small defined injury to rat aortic endothelium. Lab Invest. 1981;44:301-308. [Medline] [Order article via Infotrieve]
33. Jamal A, Bendeck M, Langille BL. Structural changes and recovery of function after arterial injury. Arterioscler Thromb. 1992;12:307-317. [Abstract/Free Full Text]
34. Rotello RJ, Lieberman RC, Purchio AF, Gerschenson LE. Coordinated regulation of apoptosis and cell proliferation by transforming growth factor ß1 in cultured uterine epithelial cells. Proc Nat Acad Sci U S A. 1991;88:3412-3415. [Abstract/Free Full Text]
35. Jacobson MD, Burne JF, King MP, Miyashita T, Reed JC, Raff MC. Bcl-2 blocks apoptosis in cells lacking mitochondrial DNA. Nature. 1993;361:365-369. [Medline] [Order article via Infotrieve]
36. Lowe SW, Ruley HE. Stabilization of the p53 tumor suppressor is induced by adenovirus 5 E1A and accompanies apoptosis. Genes Dev. 1993;7:535-545. [Abstract/Free Full Text]
37. Lowe SW, Schmitt EM, Smith SW, Osborne BA, Jacks T. p53 is required for radiation-induced apoptosis in mouse thymocytes. Nature. 1993;362:847-849. [Medline] [Order article via Infotrieve]
38. Grassilli E, Carcereri de Prati A, Monti D, Troiano L, Menegazzi M, Barbieri D, Franceschi C, Suzuki H. Studies of the relationship between cell proliferation and cell death, II: early gene expression during concanavalin A-induced proliferation or dexamethasone-induced apoptosis of rat thymocytes. Biochem Biophys Res Commun. 1992;188:1261-1266. [Medline] [Order article via Infotrieve]
39. Miura M, Zhu H, Rotello R, Hartweig EA, Yuan J. Induction of apoptosis in fibroblasts by IL-1ß-converting enzyme, a mammalian homolog of the C. elegans cell death gene ced-3. Cell. 1993;75:653-660.
40. Yuan J, Shaham S, Ledoux S, Ellis HM, Horvitz HR. The C. elegans cell death gene ced-3 encodes a protein similar to mammalian interleukin-1b-converting enzyme. Cell. 1993;75:641-652. [Medline] [Order article via Infotrieve]
41. Bennett MR, Evan GI, Newby AC. Deregulated expression of the c-myc oncogene abolishes inhibition of proliferation of rat vascular smooth muscle cells by serum reduction, interferon-, heparin, and cyclic nucleotide analogues and induces apoptosis. Circ Res. 1994;74:525-536.[Abstract/Free Full Text]

This article has been cited by other articles:

				M. C.H. Clarke, T. D. Littlewood, N. Figg, J. J. Maguire, A. P. Davenport, M. Goddard, and M. R. Bennett Chronic Apoptosis of Vascular Smooth Muscle Cells Accelerates Atherosclerosis and Promotes Calcification and Medial Degeneration Circ. Res., June 20, 2008; 102(12): 1529 - 1538. [Abstract] [Full Text] [PDF]

				K. Nishimura, W. Li, Y. Hoshino, T. Kadohama, H. Asada, S. Ohgi, and B. E. Sumpio Role of AKT in cyclic strain-induced endothelial cell proliferation and survival Am J Physiol Cell Physiol, March 1, 2006; 290(3): C812 - C821. [Abstract] [Full Text] [PDF]

				M. E. Safar and H. S. Boudier Vascular Development, Pulse Pressure, and the Mechanisms of Hypertension Hypertension, July 1, 2005; 46(1): 205 - 209. [Abstract] [Full Text] [PDF]

				B. L. Langille and D. Dajnowiec Cross-Linking Vasomotor Tone and Vascular Remodeling: A Novel Function for Tissue Transglutaminase? Circ. Res., January 7, 2005; 96(1): 9 - 11. [Full Text] [PDF]

				M. Kiyono and M. Shibuya Bone Morphogenetic Protein 4 Mediates Apoptosis of Capillary Endothelial Cells during Rat Pupillary Membrane Regression Mol. Cell. Biol., July 1, 2003; 23(13): 4627 - 4636. [Abstract] [Full Text] [PDF]

				S. Taurin, V. Seyrantepe, S. N. Orlov, T.-L. Tremblay, P. Thibault, M. R. Bennett, P. Hamet, and A. V. Pshezhetsky Proteome Analysis and Functional Expression Identify Mortalin as an Antiapoptotic Gene Induced by Elevation of [Na+]i/[K+]i Ratio in Cultured Vascular Smooth Muscle Cells Circ. Res., November 15, 2002; 91(10): 915 - 922. [Abstract] [Full Text] [PDF]

				E. Durand, Z. Mallat, F. Addad, F. Vilde, M. Desnos, C. Guerot, A. Tedgui, and A. Lafont Time courses of apoptosis and cell proliferation and their relationship to arterial remodeling and restenosis after angioplasty in an atherosclerotic rabbit model J. Am. Coll. Cardiol., May 15, 2002; 39(10): 1680 - 1685. [Abstract] [Full Text] [PDF]

				Z. S. Jackson, A. I. Gotlieb, and B. L. Langille Wall Tissue Remodeling Regulates Longitudinal Tension in Arteries Circ. Res., May 3, 2002; 90(8): 918 - 925. [Abstract] [Full Text] [PDF]

				A. H. M. Hassan, I. M. Lang, M. Ignatescu, R. Ullrich, D. Bonderman, P. Wexberg, F. Weidinger, and H. D. Glogar Increased intimal apoptosis in coronary atherosclerotic vessel segments lacking compensatory enlargement J. Am. Coll. Cardiol., November 1, 2001; 38(5): 1333 - 1339. [Abstract] [Full Text] [PDF]

				H. D. Intengan and E. L. Schiffrin Vascular Remodeling in Hypertension: Roles of Apoptosis, Inflammation, and Fibrosis Hypertension, September 1, 2001; 38(3): 581 - 587. [Abstract] [Full Text] [PDF]