Phenotypic Heterogeneity of Rat Arterial Smooth Muscle Cell Clones
Implications for the Development of Experimental Intimal Thickening
Abstract It is well accepted that smooth muscle cells (SMCs) cultured from normal rat arterial media have different morphological and biological features compared with SMCs cultured from experimental intimal thickening (IT) 15 days after endothelial injury. It is not known, however, whether the phenotypic modulation producing IT cells occurs in any medial SMCs or only in a particular SMC subpopulation. To distinguish among these possibilities, the phenotypic features of SMC clones derived from normal adult media and the IT 15 days after endothelial lesion were analyzed according to morphological appearance, replicative activity in the presence and absence of fetal calf serum, and [3H]thymidine incorporation and motile activity; these features were compared with those of the respective SMC parental populations. Two categories of SMC clones predominated: spindle clones, with morphological features similar to those of the parental population from the normal media, and epithelioid clones, with morphological features similar to those of the IT parental population. Both categories were present among clones produced from normal media and IT; however, spindle was more common among normal media clones, and epithelioid, among IT clones. The behavior in vitro was distinct for each category of clones and did not depend on their origin. Our results are compatible with the possibility that the SMC population of IT in vivo derives mainly from SMCs belonging to the category exhibiting epithelioid features in vitro.
Dr Göran Hansson kindly acted as Guest Editor for this manuscript.
- Received December 1, 1995.
- Accepted January 26, 1996.
Experimental arterial IT after endothelial injury represents the most-used model for restenosis and SMC accumulation during atheromatous plaque formation.1 It is well established that IT depends on migration of SMCs from the media and on their replication1 ; more recently, SMC apoptosis has been suggested as playing a role in IT evolution.2 SMC migration and replication result from the combined action of several cytokines, growth factors, and extracellular matrix components.1 3 These factors may act on all medial SMCs, as is generally accepted, or, alternatively, on SMCs that are selectively prone to produce the IT population upon appropriate stimulation. Cultured SMCs from IT 15 days after endothelial injury exhibit distinct morphological and biological features compared with SMCs from normal media.4 This observation supports the hypothesis that SMCs in the two situations are biologically different but does not distinguish between the two possibilities mentioned above. It is more and more evident that SMCs are heterogeneous in several respects, including their capacity to react to the stimuli that play a role in IT formation.5 6 7 Thus, IT development may also depend on the phenotype of SMCs that are subject to migratory and replicative stimuli.
SMC heterogeneity has been mainly demonstrated by culturing populations with different phenotypes or expressing different proteins from the media of animals at various ages,8 9 10 11 from different portions of the arterial tree,12 13 or from different portions of the same artery.4 14 Thus, distinct SMC populations can be obtained from the luminal portion of the normal aortic media compared with the abluminal portion,15 from the IT induced by endothelial injury compared with the underlying media,4 or from IT at different times after endothelial injury.14 SMC clones produced from the normal aortic media of rats at different ages also display different cytoskeletal features.16
To analyze whether the normal arterial media contains SMCs capable of expressing in vitro features present in the whole population of IT cells, we characterized SMC clones from the normal aortic media and IT 15 days after endothelial injury in adult rats. We report here that such SMC clones can be classified into distinct categories on the basis of their phenotypic features irrespective of their origin; moreover, a proportion of SMC clones obtained from normal media has the same phenotypic features of the SMC whole population derived from experimental IT 15 days after endothelial injury. These findings are compatible with the possibility that the SMC whole population of IT in vivo derives essentially from a small number of distinct SMCs resident in the normal media.
Cell Culture and SMC Cloning
A total of 40 adult (6- to 8-week-old) Wistar rats were used. SMCs were isolated by enzymatic digestion9 14 from the thoracic aortic media and from the 15-day-old IT produced by removing the endothelium of the thoracic aorta by means of an inflated embolectomy catheter.17 18 Cells were plated at a density of approximately 3×103 cells/cm2 in DMEM (GIBCO) supplemented with 10% FCS (Biological Industries). Whole SMC populations were grown up to the fifth passage in the same medium. For cloning, SMCs were obtained from normal aortic media and IT 15 days after endothelial injury.16 Four cloning experiments per situation were performed. Cloning was performed 48 hours after plating of the primary culture, when the whole SMC population is still available; this procedure allows us to obtain isolated cells more easily than at the time of plating. SMCs were cloned by limiting dilution (0.5 or 0.75 cell/well) into FCS-coated 96-well plates in DMEM containing 20% FCS. Cloned SMCs were expanded up to the third passage under the same conditions; they were then cultured as the whole SMC populations.16 The number and features of clones were calculated from three experiments per condition. In all experiments, SMCs were studied between the sixth and 15th passages.
Cell Proliferation and [3H]Thymidine Incorporation
For evaluation of proliferation, SMCs were plated in DMEM in the presence of 10% FCS at a concentration of 18×103 cells/cm2 and were counted after 7 days by using a hemocytometer. Cells were also plated in DMEM supplemented with 2% PDS14 or in a synthetic medium (Monomed, CSL Ltd) at a density of 7×103 cells/cm2 and were counted 7 days after plating by using a fluorescence-activated cell sorter (FACScan, Beckton Dickinson). The results were calculated as the ratio of counted to seeded cells.
For [3H]thymidine incorporation, SMCs were plated at a concentration of 7×103 cells/cm2 and synchronized for 4 days in DMEM supplemented with 10−9 mol/L selenium, 10−9 mol/L insulin (both, Collaborative Research), and 5 μg/mL transferrin (SIT; GIBCO).19 Fresh medium plus 10% FCS and 0.1 μCi/mL [3H]thymidine (specific activity, 5 Ci·mmol−1·L−1; Amersham) were added for 16 hours. [3H]thymidine incorporation was evaluated by trichloracetic acid precipitation and counting in a scintillation counter (Beckman Instruments). Results were expressed as the ratio of total [3H]thymidine incorporation in FCS to total [3H]thymidine incorporation in SIT.
For evaluation of cell migration, SMCs were plated in DMEM in the presence of 10% FCS at a concentration of 11×103 cells/cm2. Confluent cultures were scratched with silicone rubber to obtain an 0.8-mm-wide in vitro wound.20 After 24 hours, migrating cells invading the empty space were counted by using a Zeiss Axiophot photomicroscope (Carl Zeiss), a charge-coupled device camera (type FMC-4005, AVT-Horn), and a Zeiss VIDAS (Carl Zeiss) interactive image-analysis system. Six randomly preselected fields (length, 2.5 mm) were analyzed per experiment. Results were calculated as the total number of migrated cells per field.
SMCs were cultured in the presence of 10% FCS and fixed at subconfluence or confluence; results were similar in both conditions. Double immunofluorescence staining was performed directly in the Petri dish or on cytocentrifuged SMCs. A mouse monoclonal IgG2a specific for α-SMA,21 and three affinity-purified rabbit polyclonal IgGs specific for desmin,18 SM myosin heavy chains,22 and human von Willebrand factor and reacting with rat von Willebrand factor (Sigma) were used.9 Cell counts were made on cytocentrifuged cells16 by using the Zeiss interactive image-analysis system.
Results are given as mean±SEM. For statistical evaluation, the results were analyzed by means of Student’s t test. Differences were considered statistically significant at values of P<.05.
Cloning efficiency, ie, the percentage of clones growing in 96-well dishes, and the percentage of clones obtained after expansion to the fifth passage were higher in clones derived from IT (41% and 24%, respectively) than in clones derived from normal media (15% and 7%, respectively) (Table 1⇓).
Clear differences were noted in the morphology of normal media and IT whole SMC populations.4 14 SMCs cultured from normal media were spindle shaped and displayed the classic “hills-and-valleys” pattern at confluence. SMCs cultured from IT appeared epithelioid and grew in a single layer. Both phenotypes persisted up to the 15th passage.
Clones derived from both populations were heterogeneous in their morphology and could be divided into four categories: (1) spindle-shaped, corresponding to the hills-and-valleys appearance; (2) epithelioid, polygonal cells growing in a monolayer; (3) thin-elongated, which consisted of densely packed elongated cells forming typical whorls; and (4) senescent, characterized by large vacuolated cells that could not reach confluence (Fig 1⇓). The proportions of these categories were different in clones derived from normal media and IT (Table 2⇓). Spindle-shaped clones were the main phenotype among those derived from normal media, whereas epithelioid clones were the most common among those derived from IT (40% and 63% of the total number of clones, respectively). Thin-elongated and senescent morphologies were more common in clones derived from normal media than in those derived from IT. Between the fifth and 15th passages several clones, independent of their morphology, underwent senescent changes and were discarded. Thus, independent of their origin, SMC clones could be classified on the basis of their morphology into four categories. Nevertheless, the predominance of a category appears to depend on the origin of the clones. To see whether biological features of the different clones are also independent of their origin, we studied six spindle-shaped, six epithelioid, and four thin-elongated clones for cell proliferation in the presence and absence of FCS, [3H]thymidine incorporation, migration, and cytoskeletal features. Half the clones of each category were randomly taken from those of normal media and half from those of IT. Cell proliferation and [3H]thymidine incorporation were similar among clones having a similar morphology irrespective of their origin. Moreover, spindle-shaped clones displayed features similar to those of the whole SMC population isolated from normal media, whereas epithelioid clones displayed features similar to those of the whole SMC population derived from IT. For this reason we concentrated our efforts on the analysis of these two clonal SMC populations.
Because SMCs generally continue to proliferate after reaching confluence, in our clones we examined both cell number 7 days after plating in the presence of 10% FCS, when cells are postconfluent,16 and DNA synthesis by means of [3H]thymidine incorporation after FCS stimulation of growth-arrested subconfluent cultures.
In the presence of 10% FCS, the number of cells counted 7 days after plating was higher in spindle-shaped than in epithelioid clones (6.4±0.5 and 4.1±0.2, respectively, P<.01). The proliferation rates of spindle-shaped and epithelioid clones did not differ from those of normal media and IT whole SMC populations, respectively (Fig 2A⇓). SMCs from epithelioid clones stopped growing when confluent (about the fifth day after plating), which has been noted for whole SMC populations isolated from IT.4 14 Thin-elongated clones showed a very high cell proliferation rate (14.7±1.2) when compared with the other categories and whole SMC populations (P<.001 in all cases).
The [3H]thymidine incorporation rate, which was measured 16 hours after FCS was added in synchronized SMC clones, was lower in spindle-shaped (2.6±0.2) than epithelioid (4.4±0.2; P<.001) clones (Fig 2B⇑). The results of spindle-shaped and epithelioid clones were similar to those of normal media and IT whole SMC populations, respectively (Fig 2B⇑). Thin-elongated clones (2.1±1.1) behaved similarly to spindle-shaped clones. Thus, although epithelioid cells (from clones or whole SMC populations) stop growing at confluence, they incorporate more [3H]thymidine than spindle-shaped cells when they are stimulated by FCS after synchronization. This shows a further biological difference between spindle and epithelioid SMC clones.
Because SMCs isolated from IT and cultured as a whole SMC population grow in the presence of PDS,4 14 we tested whether epithelioid clones behave similarly (Fig 2C⇑). Epithelioid clones showed a doubling of their cell number (2.1±0.3) in the presence of 2% PDS 6 days after plating, whereas spindle-shaped clones remained quiescent (0.7±0.1; P<.05). Thin-elongated clones behaved as the spindle-shaped clones did. As expected, normal media whole SMC populations did not replicate in the presence of PDS, whereas IT whole SMC populations showed an increase in cell number, an increase that was also observed for epithelioid clones (Fig 2C⇑).
Migration experiments showed that spindle-shaped clones display a migration activity lower than that of epithelioid clones (95±3 and 197±26, respectively; P<.05) (Fig 3⇓). Whole SMC populations from normal media behaved as spindle-shaped clones; whole SMC populations from IT showed results similar to those of epithelioid clones (Fig 3⇓). Thin-elongated clones showed a migration activity intermediate between that of spindle-shaped and epithelioid clones (142±15).
Cytoskeletal features of whole SMC populations cultured from normal media and IT were in accordance with previous reports.9 14 16 The percentage of α-SMA–positive cells derived from normal media was higher than the one derived from IT.14 No whole SMC population cells expressed desmin and very few of them (<5%) contained SM myosin.14 16 Irrespective of their morphology, all clones expressed a percentage of cells positive to α-SMA and SM myosin heavy chains much higher than those of the corresponding whole SMC populations. Moreover, unlike the whole SMC populations,9 14 16 some clones showed desmin-positive cells up to the 15th passage. Thus, in all clones, >80% of SMCs expressed α-SMA and 20% to 90% of cells expressed SM myosin heavy chains (Table 3⇓). Three of six tested spindle-shaped clones contained desmin; surprisingly, all thin-elongated clones expressed desmin in 10% to 30% of the cells. The tested epithelioid clones were negative for this protein. All SMC populations, particularly the IT whole SMC populations and the epithelioid clones, did not express von Willebrand factor, indicating that they did not contain endothelial cells.
Our results indicate that when classified on the basis of well-accepted criteria such as cell morphology, FCS-dependent and FCS-independent growth, [3H]thymidine incorporation, migratory activity, and cytoskeletal protein content, SMC clones from the aortic media and from IT 15 days after endothelial injury fall into four main phenotypic categories: spindle, epithelioid, thin-elongated, and senescent. To our knowledge, ours is the first report of SMC clones derived from the normal adult media that display all the features of the SMC whole population derived from IT 15 days after endothelial injury. It is noteworthy that SMCs belonging to a given phenotypic category express similar features irrespective of their origin; moreover, we have not seen modulation in vitro among the four phenotypic categories with the exception that the first three can evolve into the senescent one. Spindle and epithelioid clones are the most frequent: their biological features correspond to those of the whole SMC populations cultured from the normal aortic media and the IT 15 days after endothelial injury, respectively.4 14 Interestingly, the proportion of spindle SMC clones is predominant among those obtained from the normal media, whereas the proportion of epithelioid clones is predominant among those obtained from IT. These results indicate that the potential to develop clones with the spindle or epithelioid phenotype is different in the SMC populations of normal media and IT and that there is a correlation between the phenotypic features of the whole SMC populations derived from the aortic media or IT and the phenotypic features of the majority of the clones derived from the same locations.
Due to their similarities with whole SMC populations, we concentrated our efforts in the comparison of spindle and epithelioid clones. Their biological features are distinct, and, as far as we have observed, noninterchangeable; this is compatible with the possibility that they derive from distinct precursor SMCs. Epithelioid clones grow independent of FCS, similar to the whole SMC population from IT.4 However, epithelioid clones (like the whole SMC population from IT) stop growing at confluence, even in the presence of FCS; this probably contributes to their epithelioid appearance and is at variance with spindle clones and with the whole SMC population from the normal aortic media of adult rats. It is noteworthy that epithelioid clones (like the whole population of IT) have a high [3H]thymidine incorporation when growing at low cell density. However, our data do not clarify whether epithelioid SMCs have a shorter lag phase or whether a larger amount of cells enter in S phase compared with spindle-shaped SMCs. These observations show that each of these clonal or whole SMC populations possesses some properties that do not exactly correspond to the definition of normal growth in vitro; this may result from autocrine production of growth factor8 and may reflect their behavior in vivo when stimulated by serum components.
The migratory activities of spindle and epithelioid clones differ significantly, albeit quantitatively, and coincide with those of the corresponding whole populations. These findings are not surprising if one considers that SMCs producing IT after endothelial injury in vivo display an important motile activity.23
The cytoskeletal features of all clones, irrespective of their origin, are in agreement with those from other reports.16 It is noteworthy that in general clones show more differentiated properties compared with the whole SMC populations from both normal media16 and IT.14 Thus, >80% of SMCs from every clone are positive for α-SMA, and all clones contain a significant proportion of SM myosin heavy chain–positive cells. The expression of this protein is a reliable marker of terminal SMC differentiation.24 Moreover, some clones had a proportion of desmin-positive cells, particularly the spindle and thin-elongated clones. Desmin disappears in early phases of culture in whole SMC populations.21 Our observations in clonal populations are in line with the well-known cytoskeletal heterogeneity of vascular SMCs in vivo.25 Further studies are needed to understand the mechanisms of the higher capacity of in vitro differentiation of clonal SMC populations compared with their respective parental populations. In any event, our results confirm their SMC nature.
Several laboratories, including ours, have attempted to characterize SMC whole populations from the aortic media of rats at different ages.6 9 10 11 Fetal SMCs exhibit FCS-independent growth,11 contrary to newborn and adult SMCs.6 9 An SMC population derived from 12-day-old rats also exhibits an epithelioid phenotype and FCS-independent growth.10 Finally, old (≥18 months) rats produce epithelioid cells, which, however, do not stop growing at confluence.8 Our results show that the epithelioid serum-independent phenotype can be produced in vitro by cloning SMCs of the normal media early in primary culture. Since the whole SMC population from IT 15 days after endothelial injury has an epithelioid phenotype,4 14 it is plausible that the precursors for this SMC population are present in the media at all ages; it is conceivable that these precursor cells are the major source of SMCs in IT. Further studies are needed to pinpoint the factors influencing the emergence of a given category of SMCs when placed in culture or when stimulated in vivo by endothelial injury.
The cytoskeletal features of SMCs present in IT 15 days after endothelial injury are those of poorly differentiated SMCs but become similar to those of normal medial SMCs 60 days after endothelial injury.26 Correspondingly, the whole population of SMCs cultured from IT 15 days after endothelial injury has epithelioid features, whereas the whole population of SMCs cultured from IT 60 days after endothelial injury has spindle features.4 14 The SMC number per unit volume in IT 60 days after endothelial injury is smaller than that of SMCs in IT 15 days after endothelial injury.26 These findings are compatible with the existence of at least two populations of SMCs, both in normal media and IT 15 days after endothelial injury. To explain the change at 60 days after endothelial injury, one can hypothesize an important decrease of one of the two populations, eg, by means of apoptosis,2 rather than a phenotype modulation, as is now generally accepted.26 We are now working to verify this possibility.
Evidence for SMC heterogeneity in vivo and in vitro has been produced by means of different criteria and by using, in general, whole SMC populations.4 5 6 7 8 9 10 11 12 13 14 15 16 24 27 Our present results are in line with these reports and extend the notion of heterogeneity to clonal SMC populations derived from both normal media and IT. Benditt and Benditt28 have proposed that human atheromatous plaques are essentially monoclonal; the present observation that SMCs are heterogeneous10 12 27 is compatible with this hypothesis.
In conclusion, with the limitation that in vitro results do not necessarily reflect in vivo properties, our results suggest that SMCs react heterogeneously to stimulation by serum growth factors when placed in culture and are compatible with the possibility that the SMC population of IT in vivo derives mainly from a subpopulation of medial SMCs capable of producing in vitro the epithelioid phenotype. Further studies analyzing the biochemical and biological properties of clonal populations29 may furnish tools capable of distinguishing their respective precursors in vivo.
Selected Abbreviations and Acronyms
|DMEM||=||Dulbecco’s modified Eagle’s medium|
|FCS||=||fetal calf serum|
|SMA||=||smooth muscle actin|
|SMC||=||smooth muscle cell|
This work was partially supported by the Swiss National Science Foundation (grant 31-40372.94). We thank K. Grandchamp for help in statistical analysis, E. Denkinger and J.C. Rumbeli for photographic work, and M. Vitali for typing the manuscript.
Bondjers G, Glukhova M, Hansson GK, Postnov YV, Reidy MA, Schwartz SM. Hypertension and atherosclerosis: cause and effect, or two effects with one unknown cause? Circulation. 1991;84(suppl V):V-I2-V-16.
Walker LN, Bowen-Pope DF, Ross R, Reidy MA. Production of platelet-derived growth factor-like molecules by cultured arterial smooth muscle cells accompanies proliferation after arterial injury. Proc Natl Acad Sci U S A. 1986;83:7311-7315.
Majesky MW, Benditt EP, Schwartz SM. Expression and developmental control of platelet-derived growth factor A-chain and B-chain/sis genes in rat aortic smooth muscle cells. Proc Natl Acad Sci U S A. 1988;85:1524-1528.
McCaffrey TA, Nicholson AC, Szabo PE, Weksler ME, Weksler BB. Aging and arteriosclerosis: the increased proliferation of arterial smooth muscle cells isolated from old rats is associated with increased platelet-derived growth factor-like activity. J Exp Med. 1988;167:163-174.
Cook CL, Weiser MC, Schwartz PE, Jones CL, Majack RA. Developmentally timed expression of an embryonic growth phenotype in vascular smooth muscle cells. Circ Res. 1994;74:189-196.
Frid MG, Moiseeva EP, Stenmark KR. Multiple phenotypically distinct smooth muscle cell populations exist in the adult and developing bovine pulmonary arterial media in vivo. Circ Res. 1994;75:669-681.
Giachelli CM, Bae N, Almeida M, Denhardt DT, Alpers CE, Schwartz SM. Osteopontin is elevated during neointima formation in rat arteries and is a novel component of human atherosclerotic plaques. J Clin Invest. 1993;92:1686-1696.
Orlandi A, Ehrlich HP, Ropraz P, Spagnoli LG, Gabbiani G. Rat aortic smooth muscle cells isolated from different layers and at different times after endothelial denudation show distinct biological features in vitro. Arterioscler Thromb. 1994;14:982-989.
Villaschi S, Nicosia RF, Smith MR. Isolation of a morphologically and functionally distinct smooth muscle cell type from the intimal aspect of the normal rat aorta: evidence for smooth muscle cell heterogeneity. In Vitro Cell Dev Biol Anim. 1994;30A:589-595.
Bochaton-Piallat ML, Gabbiani F, Ropraz P, Gabbiani G. Age influences the replicative activity and the differentiation features of cultured rat aortic smooth muscle cell populations and clones. Arterioscler Thromb. 1993;13:1449-1455.
Gabbiani G, Gabbiani F, Heimark RL, Schwartz SM. Organization of actin cytoskeleton during early endothelial regeneration in vitro. J Cell Sci. 1984;66:39-50.
Skalli O, Ropraz P, Trzeciak A, Benzonana G, Gillessen D, Gabbiani G. A monoclonal antibody against alpha-smooth muscle actin: a new probe for smooth muscle differentiation. J Cell Biol. 1986;103:2787-2796.
Clowes AW, Clowes MM. Kinetics of cellular proliferation after arterial injury, IV: heparin inhibits rat smooth muscle mitogenesis and migration. Circ Res. 1986;58:839-845.
Gabbiani G, Schmid E, Winter S, Chaponnier C, De Chastonay C, Vandekerckhove J, Weber K, Franke WW. Vascular smooth muscle cells differ from other muscle cells: predominance of vimentin filaments and a specific alpha-type actin. Proc Natl Acad Sci U S A. 1981;78:298-302.
Shanahan CM, Weissberg PL, Metcalfe JC. Isolation of gene markers of differentiated and proliferating vascular smooth muscle cells. Circ Res. 1993;73:193-204.
Benditt EP, Benditt JM. Evidence for a monoclonal origin of human atherosclerotic plaques. Proc Natl Acad Sci U S A. 1973;70:1753-1756.