- Email: [email protected]
Collagen biosynthesis in cell culture: Comparison of corneal keratocytes and skin fibroblasts
Disponible en ligne sur www.sciencedirect.com
Pathologie Biologie 56 (2008) 66–69 http://france.elsevier.com/direct/PATBIO/
Original article
Collagen biosynthesis in cell culture: Comparison of corneal keratocytes and skin fibroblasts Effect of rhamnose-rich oligo- and polysaccharides Biosynthe`se du collage`ne en culture cellulaire : comparaison entre ke´ratocytes corne´ens et fibroblastes cutane´s Effet des oligo- et polysaccharides riches en rhamnose V. Ravelojaona, A.-M. Robert, L. Robert *, G. Renard Laboratoire de recherches ophtalmologiques, hôpital Hôtel-Dieu, université Paris-V, 1, place Parvis-Notre-Dame, 75181 Paris cedex 04, France Received 23 August 2007; accepted 29 October 2007 Available online 4 January 2008
Abstract Corneal keratocytes are often confounded with fibroblasts, although their matrix-synthetic phenotype is quite different as shown by the nature and relative amount of the different collagens and proteoglycans–glycosaminoglycans synthesized. In these experiments, we compared the concentration of collagens excreted in the culture medium by human corneal keratocytes and skin fibroblasts at three consecutives passages. Although the keratocytes excreted less collagen at earlier passages, they approached and reached fibroblasts at later passages. This can be taken as an indication of the progressive loss of a specific keratocyte phenotype with increasing passages (in vitro aging). The effect of rhamnose-rich oligo- and polysaccharides on collagen secretion also confirmed the subtle differences between these two cell–types, as well as the difference between saturation density of both cell types at confluence and the proportion of dead cells floating on top of the culture medium. These differences were also attenuated with passage number without disappearing completely. The significance of these findings will be discussed in the light of previous results in our and other laboratories on matrix-secreting phenotypes and aging. # 2007 Elsevier Masson SAS. All rights reserved. Résumé On confond souvent les kératocytes cornéens avec les fibroblastes cutanés, bien que leurs phénotypes, reflétant la synthèse de la matrice, soient différents : il suffit de regarder la nature et les quantités relatives des différents types de collagènes et protéoglycanes–glycosaminoglycanes synthétisés. Au cours des ces expériences, nous avons comparé les concentrations de collagènes excrétés dans le milieu de culture par les kératocytes cornéens et les fibroblastes de la peau, à trois passages consécutifs. Bien que les kératocytes excrètent moins de collagène aux passages précoces, ils se rapprochent, voire rejoignent les quantités de collagène synthétisées par les fibroblastes, aux passages plus tardifs. On peut considérer cela comme une indication de la perte progressive de spécificité du phénotype des kératocytes, au cours de la sénescence in vitro. L’effet des oligo- et polysaccharides riches en rhamnose, sur la sécrétion de collagène, confirme également les différences subtiles entre ces deux types cellulaires, à propos de la densité de saturation et de la proportion de cellules mortes, flottant à la surface du milieu de culture. Ces différences sont également atténuées au cours de passages sans disparaître complètement. La signification de ces découvertes sera discutée, en prenant compte les
Abbreviations: RROP-s, Rhamnose-rich oligo- and polysaccharides; ECM, Extracellular matrix; NHDF, Normal human dermal fibroblast; GAG, Glycosaminoglycans; PG, Proteoglycans * Corresponding author. E-mail address: [email protected] (L. Robert). 0369-8114/$ – see front matter # 2007 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.patbio.2007.10.003
V. Ravelojaona et al. / Pathologie Biologie 56 (2008) 66–69
67
résultats sur les phénotypes liés à la sécrétion des macromolécules de la matrice extracellulaire ainsi que sur le vieillissement, précédemment obtenus dans notre, et d’autres, laboratoires. # 2007 Elsevier Masson SAS. All rights reserved. Keywords: Collagen; Fibroblasts; Keratocytes; Cornea; Skin; Aging; Connective tissues; Rhamnose-rich oligo- and polysaccharides Mots clés : Collagène ; Fibroblastes ; Kératocytes ; Cornée ; Peau ; Vieillissement ; Tissus conjonctifs ; Oligo- et polysaccharides riches en rhamnose
1. Introduction The structure and function of connective tissues, such as the skin or the cornea are largely dependent on the nature and relative amount of extracellular matrix (ECM) – macromolecules synthesized. This ‘‘program’’ of matrix biosynthesis is changing with cell-phenotype and also with age. The determination of these parameters, the nature and relative amount of ECM – macromolecules synthesized can be taken as a characteristic of cellular phenotype. Such determinations, carried out at successive passages can give some information on the age-dependent changes of cell-phenotype. We shall describe here such experiments carried out on human corneal keratocytes and skin fibroblasts at three successive passages based on the quantitative measure of total collagen excretion in the culture medium. We also tested the action of rhamnose-rich oligo- and polysaccharides (RROP-s) previously shown to modulate matrix biosynthesis [1] and collagen synthesis by these two cell types. Both RROP-s tested (RROP-1 and RROP-3) modulated differently the collagen synthetic activity of keratocytes and fibroblasts. These results also confirm the subtle differences between these two cell-phenotypes as far as their matrix synthetic activity is concerned. 2. Materials and methods 2.1. Cell culture Human skin fibroblasts were obtained from Cambrex (Emerainville, France) (NHDF-Adult cryopreserved, product code CC-2511, lot number 4F1293) derived from the skin of a 39-year-old woman, according to the furnishers’ informations. Corneal keratocytes were obtained from corneal rings excised at transplantation from corneas kept at the French Eye Bank, as previously described [2]. Fibroblasts were cultivated in standard conditions: Dulbecco’s modified Eagle culture medium (DMEM-glutamax, InvitrogenTM) supplemented with 10% (v/v) foetal calf serum (FCS, Gibco1), antibiotics: penicillin (100 U/ml; Gibco1) and streptomycin (100 mg/ml; Gibco1), and an antifongic (0.25 mg/ml amphotericin B, Gibco1). This was designated as the complete culture medium. Keratocytes were cultivated in a D-MEM/F-12, Dulbecco’s modified Eagle culture medium, supplemented with 10% (v/v) foetal calf serum, antibiotics and antifongic (Penicillin 100 U/ ml (Gibco1), Streptomycin 100 mg/ml and Fongisone).
These two kinds of cell were incubated in a temperaturecontrolled, humidified incubator with 5% CO2 at 37 8C. Keratocytes were kept at the fifth passage in liquid nitrogen for one year. Fibroblast cultures were started freshly at the delivery of cells by Cambrex and kept also in liquid nitrogen. Fibroblasts and keratocytes were used at the 7th, 11th and 12th passages. In all experiments, cells were seeded at 50,000 cells per well, in a Nunc-plate of 12 wells. For every experiment, four to six parallel cultures were set up in order to reach significance. All cells were grown in the complete culture medium, in 75 cm2 surface ventilated culture flasks (Nunc) and subcultured by trypsinisation (0.05% trypsin, Gibco1). Culture medium was changed every two to three days. Both serial cultures of cells were used for the determination of the following parameters: time to reach confluency in the standard culture flasks of 75 cm2; the total number of cells at saturation density determined after trypsinisation on the Coulter counter, at increasing passages [3]; total amount of collagen synthesized and deposited by the cells after 72 h, using the colorimetric procedure with PicroSirius red staining [4,5]. This method has the advantage to estimate not only freshly synthesized collagen but also the total amount of collagen accumulated by the cells during the 72 h culture period. Shortly, cells were seeded in six well plates at 5.104 cells/well, after 72 h of culture, cells were washed twice with PBS then fixed for one hour with 1 ml of the Bouin’s solution at room temperature. Following fixation, cells were washed twice with distilled water and then stained with Sirius-red solution (0.5 g Sirius red F3B Gurr BDH, 500 ml saturated aqueous solution of picric acid and a little solid picric acid to ensure saturation) for one hour under shaking at room temperature. Samples were then washed consecutively with distilled water and 0.01 M HCl to remove unbound dye. Then, bound dye was solubilised by incubation in 500 ml of 0.1 M NaOH for one hour under shaking. Samples were transferred in a spectrophotometer cuvette and absorbency read at 550 nm. 2.2. Rhamnose-rich oligo- and polysaccharides All a-l-rhamnose-rich oligo- and polysaccharide preparations (RROP-s) were obtained from Solabia - BioEurope (Pantin, France). RROP-1 (commercial name Rhamnosoft1) is obtained from Klebsiella pneumoniae strains. The other polysaccharide of 5 kDa, RROP-3 (commercial name
68
V. Ravelojaona et al. / Pathologie Biologie 56 (2008) 66–69
Oligosaccharide BEC-291) was obtained from Klebsiella planticola, by hydrolysis of the 45 kDa parent-polysaccharide (RROP-2). Their structure and also their chemical and biological properties were recently described [1,6]. These two rhamnose rich-oligo- and polysacharides, prepared as described in [1,6], were added to the cultures, at the concentration of 20 mg/ml. The results of collagen synthesis were estimated on a calibration curve obtained with collagen type I, as described in [5]. As a matter of fact, this is the major collagen of both tissues: skin and cornea. 2.3. Statistics Statistical evaluation of the results was done with the Mann– Whitney distribution free U-test on results obtained with a minimal of four parallel experiments.
Fig. 1. Evolution of collagen synthesis by fibroblasts and keratocytes during the successive passages. Collagen synthesis expressed as microgram of collagen type I secreted by 106 cells in 48 h.
3. Results As shown on Table 1, the number of cells at saturation density was higher for keratocytes than for fibroblasts. The number of cells at saturation density declined slightly with passage number for both keratocytes and fibroblasts. The number of dead cells was compared for these two cell types at the twelfth passage. It was significantly more important (over 7% of total seeded cells) for keratocytes than for fibroblasts (0.56% of total seeded cells). As shown on Fig. 1, keratocytes secreted significantly lower concentrations of collagen than fibroblasts at the earliest passage (7th): about 3000 mg/106 cells for fibroblasts and only 1337 mg/106 cells for keratocytes ( p < 2.1010). This difference decreased at passage 11 (4720 mg/106 cells for fibroblasts and 4007 mg/106 cells for keratocytes ( p < 2.104)) and decreased even more at passage 12 (5715 mg/106 cells for fibroblasts and 5192 mg/106 cells for keratocytes ( p < 0.0024)). This suggests that close to the normal corneal phenotype, keratocytes secrete less collagen than fibroblasts. With increasing passages, the corneal phenotype is progressively lost and keratocytes tend to secrete as much collagen as skin fibroblasts.
Fig. 2. Modulation of collagen synthesis by RROP-s for keratocytes and fibroblasts at three increasing passages. Notice the differences in the modulation of biosynthesis in presence of RROP-1 and RROP-3. Several of the results differ nonsignificantly because of the dispersion of the results. Significant differences between values obtained with a RROP and values obtained in control cultures, are indicated by * ( p < 0.05). Results are expressed in percentage as comparison to the control cultures.
As shown on Fig. 2, the action of rhamnose-rich oligo- and polysaccharides acted differently on these two cell types at the earliest passage (no 7). Fibroblasts were hardly affected or slightly inhibited by the RROP-1 but keratocytes were quite
Table 1 Cell death and cell density at confluency for keratocytes and fibroblasts at three consecutive passages at four days after passing Number of passage
Keratocytes Percentage of cell death
Number of cells at confluency
6.46% 7.33% (***)|
1.86 10+5 (NS)| vs. fibroblasts 11th passage*vs. 12th passage*** 1.71 10+5 (***)| vs. fibroblasts 1.16 10+5 (NS) | vs fibroblasts
7 11 12
Fibroblasts Percentage of cell death
Number of cells at confluency
0.56%
1.56 10+5 vs. 11th passage* vs. 12th passage** 1.14 10+5 9.89 10+4
Cell death is expressed as a percentage of dead cells as compared to total number of cells. Significance of the results was estimated with the Mann–Whitney distribution free U-test on results obtained with a minimal of four to six parallel experiments. Values marked with a bold clover correspond to the comparison between keratocytes and fibroblasts; results obtained at the same passage. Other values correspond to the comparison between the values obtained at the earliest passages (no 7) and those obtained at later passages (no 11 or no 12). Significance of differences between keratocytes and fibroblasts and for the decrease of the saturation with increasing passages: * p < 0.05. ** p < 0.01. *** p < 0.001.NS: no significance.
V. Ravelojaona et al. / Pathologie Biologie 56 (2008) 66–69
strongly stimulated by RROP-3 and less strongly (near about +10%) by RROP-1. This rhamnose-rich oligosaccharide, RROP-3 increased much more (+35.5%) collagen biosynthesis by keratocytes than by fibroblasts (+1.8%) at passage 7. At the later passages studied (11th and 12th), these differences were attenuated. According to these results, the differences of RROP action on these two cell types are strongest at the earliest passage and decrease without completely disappearing at later passages. Only few of these differences are, however, significant because of the dispersion of individual values and keep, therefore, only an indicative value. 4. Discussion The above-presented results on saturation density of both cell types and on the proportion of dead cells floating on top of the culture fluid (Table 1) suggest subtle differences between these two cell types. More keratocytes than fibroblasts were found at saturation density, but this difference decreased with passage number. The proportion of dead cells at the twelfth passage was still higher for keratocytes than for fibroblasts. The composition of the ECM for both tissues, cornea and skin, is quite different: reflection of the differences of their matrix-synthetic phenotype. Cornea is transparent, as the result of the ordered deposition of its collagen fibers during embryonic development [7]. These fibers are composed essentially of collagen types I, V and VI; this last collagen type represents close to 20% of total corneal collagens [8]. Their rate of synthesis and turnover could be measured on calfcornea explant cultures and shown to be surprisingly rapid [9]. Collagen type III was shown to increase only in inflammatory conditions [10]. Dermal collagen of human skin is also rich in type I collagen, but also in type III, besides lower proportion of type V. Type VI collagen is hardly detectable. With age, the ratio of type III to type I collagen increases [11]. Comparable differences between cornea and skin were shown for the GAGPG-compositions of both tissues [12]. The passage-dependent phenotype shift in GAG-PG-biosynthesis was studied for keratocytes and fibroblasts both in cell and explant cultures. With increasing passages, there was a shift in the proportion of individual GAG-s synthesized but there was no decline of total synthesis for both cell types. Our present results suggest also a shift in collagen phenotype with increasing passage number for keratocytes but no decrease of synthesis up to the twelfth passage. Similar results were obtained with fibroblasts up to the twenty-fifth passage: no decline of total synthesis [13]. The relatively large difference in total collagen synthesis by keratocytes (of about 55%) below the level of fibroblasts, at the seventh passage, decreased to about 15% at passage 11 and to hardly 10% at passage 12. This progressive increase of
69
collagen synthesis by keratocytes might well be accompanied by the progressive disappearance of the qualitative differences between these two cell types, essentially by the decrease of collagen type VI. This, however, remains to be confirmed. Taken together with our data on the progressive shift of GAGPG- synthesis by increasing passages of keratocytes, the progressive loss of the corneal phenotype of keratocytes during in vitro ‘‘aging’’ seems to be confirmed. Acknowledgement Supported by the Institut DERM, Paris and by NATURA Ltda, São Paulo, Brazil. References [1] Andrès E, Molinari J, Péterszegi G, Mariko B, Ruszova E, Velebny V, et al. Pharmacological properties of rhamnose-rich polysaccharides, potential interest in age-dependent alterations of connective tissues. Pathol Biol 2006;54:420–5. [2] Isnard N, Sabatier P, Robert AM, Robert L, Renard G. MMP-type endopeptidase activity in the cornea. Its evolution during organ culture storage at the Eye Bank. Effect of hyaluronan. Pathol Biol 2005;53:424–9. [3] Péterszegi G, Molinari J, Ravelojaona V, Robert L. Effect of advanced glycation end-products on cell proliferation and cell death. Pathol Biol 2006;54:396–404. [4] Tullberg-Reinert H, Jundt G. In situ measurement of collagen synthesis by human bone cells with a Sirius red-based colorimetric microassay: effects of transforming growth factor beta 2 and ascorbic acid 2-phosphate. Histochem Cell Biol 1999;112:271–6. [5] Péterszegi G, Andrès E, Molinari J, Ravelojaona V, Robert L. Effect of cellular aging on collagen biosynthesis. I-methodological consideration and pharmacological applications. Arch Gerontol Geriatr (2007), doi: 10.1016/j.archger, 2007.08.019. [6] Ravelojaona V, Molinari J, Robert L. Protection by rhamnose-rich polysaccharides against the cytotoxicity of Maillard reaction products. Biomed Pharmacother 2006;60:359–62. [7] Robert L, Reyss-Brion M, Junqua S, Salaun J. Composition chimique de la cornée de l’embryon de poulet et de poule adulte. C R Acad Sci III 1969;269:491–3. [8] Robert L, Legeais JM, Robert AM, Renard G. Corneal Collagens. Pathol Biol 2001;49:353–63. [9] Kern P, Menasche M, Robert L. Relatives rates of biosynthesis of collagen type I, type V and type VI in calf cornea. Biochem J 1991;274:615–7. [10] Menasche M, Junqua S, Payrau P, Robert L. Étude de la régulation de la biosynthèse des macromolécules de la matrice intercellulaire de la cornée. Pathol Biol 1975;23:705–9. [11] Labat-Robert J, Kern P, Robert L. Biomarkers of connective tissue aging: biosynthesis of fibronectin, collagen type III and elastase. Ann N Y Acad Sci 1992;673:16–22. [12] Isnard N, Fodil I, Robert L, Renard G. Modulation of cell-phenotype during in vitro aging. Glycosaminoglycan biosynthesis by skin fibroblasts and corneal keratocytes. Exp Gerontol 2002;37:1377–85. [13] Ravelojaona V, Robert L, Robert AM. Effect of cellular aging on collagen biosynthesis. II-collagen synthesis and deposition by a human skin fibroblast strain over 25 passages. Arch Gerontol Geriatr (2007), doi: 10.1016/j. archger. 2007.08.017.
Pathologie Biologie 56 (2008) 66–69 http://france.elsevier.com/direct/PATBIO/
Original article
Collagen biosynthesis in cell culture: Comparison of corneal keratocytes and skin fibroblasts Effect of rhamnose-rich oligo- and polysaccharides Biosynthe`se du collage`ne en culture cellulaire : comparaison entre ke´ratocytes corne´ens et fibroblastes cutane´s Effet des oligo- et polysaccharides riches en rhamnose V. Ravelojaona, A.-M. Robert, L. Robert *, G. Renard Laboratoire de recherches ophtalmologiques, hôpital Hôtel-Dieu, université Paris-V, 1, place Parvis-Notre-Dame, 75181 Paris cedex 04, France Received 23 August 2007; accepted 29 October 2007 Available online 4 January 2008
Abstract Corneal keratocytes are often confounded with fibroblasts, although their matrix-synthetic phenotype is quite different as shown by the nature and relative amount of the different collagens and proteoglycans–glycosaminoglycans synthesized. In these experiments, we compared the concentration of collagens excreted in the culture medium by human corneal keratocytes and skin fibroblasts at three consecutives passages. Although the keratocytes excreted less collagen at earlier passages, they approached and reached fibroblasts at later passages. This can be taken as an indication of the progressive loss of a specific keratocyte phenotype with increasing passages (in vitro aging). The effect of rhamnose-rich oligo- and polysaccharides on collagen secretion also confirmed the subtle differences between these two cell–types, as well as the difference between saturation density of both cell types at confluence and the proportion of dead cells floating on top of the culture medium. These differences were also attenuated with passage number without disappearing completely. The significance of these findings will be discussed in the light of previous results in our and other laboratories on matrix-secreting phenotypes and aging. # 2007 Elsevier Masson SAS. All rights reserved. Résumé On confond souvent les kératocytes cornéens avec les fibroblastes cutanés, bien que leurs phénotypes, reflétant la synthèse de la matrice, soient différents : il suffit de regarder la nature et les quantités relatives des différents types de collagènes et protéoglycanes–glycosaminoglycanes synthétisés. Au cours des ces expériences, nous avons comparé les concentrations de collagènes excrétés dans le milieu de culture par les kératocytes cornéens et les fibroblastes de la peau, à trois passages consécutifs. Bien que les kératocytes excrètent moins de collagène aux passages précoces, ils se rapprochent, voire rejoignent les quantités de collagène synthétisées par les fibroblastes, aux passages plus tardifs. On peut considérer cela comme une indication de la perte progressive de spécificité du phénotype des kératocytes, au cours de la sénescence in vitro. L’effet des oligo- et polysaccharides riches en rhamnose, sur la sécrétion de collagène, confirme également les différences subtiles entre ces deux types cellulaires, à propos de la densité de saturation et de la proportion de cellules mortes, flottant à la surface du milieu de culture. Ces différences sont également atténuées au cours de passages sans disparaître complètement. La signification de ces découvertes sera discutée, en prenant compte les
Abbreviations: RROP-s, Rhamnose-rich oligo- and polysaccharides; ECM, Extracellular matrix; NHDF, Normal human dermal fibroblast; GAG, Glycosaminoglycans; PG, Proteoglycans * Corresponding author. E-mail address: [email protected] (L. Robert). 0369-8114/$ – see front matter # 2007 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.patbio.2007.10.003
V. Ravelojaona et al. / Pathologie Biologie 56 (2008) 66–69
67
résultats sur les phénotypes liés à la sécrétion des macromolécules de la matrice extracellulaire ainsi que sur le vieillissement, précédemment obtenus dans notre, et d’autres, laboratoires. # 2007 Elsevier Masson SAS. All rights reserved. Keywords: Collagen; Fibroblasts; Keratocytes; Cornea; Skin; Aging; Connective tissues; Rhamnose-rich oligo- and polysaccharides Mots clés : Collagène ; Fibroblastes ; Kératocytes ; Cornée ; Peau ; Vieillissement ; Tissus conjonctifs ; Oligo- et polysaccharides riches en rhamnose
1. Introduction The structure and function of connective tissues, such as the skin or the cornea are largely dependent on the nature and relative amount of extracellular matrix (ECM) – macromolecules synthesized. This ‘‘program’’ of matrix biosynthesis is changing with cell-phenotype and also with age. The determination of these parameters, the nature and relative amount of ECM – macromolecules synthesized can be taken as a characteristic of cellular phenotype. Such determinations, carried out at successive passages can give some information on the age-dependent changes of cell-phenotype. We shall describe here such experiments carried out on human corneal keratocytes and skin fibroblasts at three successive passages based on the quantitative measure of total collagen excretion in the culture medium. We also tested the action of rhamnose-rich oligo- and polysaccharides (RROP-s) previously shown to modulate matrix biosynthesis [1] and collagen synthesis by these two cell types. Both RROP-s tested (RROP-1 and RROP-3) modulated differently the collagen synthetic activity of keratocytes and fibroblasts. These results also confirm the subtle differences between these two cell-phenotypes as far as their matrix synthetic activity is concerned. 2. Materials and methods 2.1. Cell culture Human skin fibroblasts were obtained from Cambrex (Emerainville, France) (NHDF-Adult cryopreserved, product code CC-2511, lot number 4F1293) derived from the skin of a 39-year-old woman, according to the furnishers’ informations. Corneal keratocytes were obtained from corneal rings excised at transplantation from corneas kept at the French Eye Bank, as previously described [2]. Fibroblasts were cultivated in standard conditions: Dulbecco’s modified Eagle culture medium (DMEM-glutamax, InvitrogenTM) supplemented with 10% (v/v) foetal calf serum (FCS, Gibco1), antibiotics: penicillin (100 U/ml; Gibco1) and streptomycin (100 mg/ml; Gibco1), and an antifongic (0.25 mg/ml amphotericin B, Gibco1). This was designated as the complete culture medium. Keratocytes were cultivated in a D-MEM/F-12, Dulbecco’s modified Eagle culture medium, supplemented with 10% (v/v) foetal calf serum, antibiotics and antifongic (Penicillin 100 U/ ml (Gibco1), Streptomycin 100 mg/ml and Fongisone).
These two kinds of cell were incubated in a temperaturecontrolled, humidified incubator with 5% CO2 at 37 8C. Keratocytes were kept at the fifth passage in liquid nitrogen for one year. Fibroblast cultures were started freshly at the delivery of cells by Cambrex and kept also in liquid nitrogen. Fibroblasts and keratocytes were used at the 7th, 11th and 12th passages. In all experiments, cells were seeded at 50,000 cells per well, in a Nunc-plate of 12 wells. For every experiment, four to six parallel cultures were set up in order to reach significance. All cells were grown in the complete culture medium, in 75 cm2 surface ventilated culture flasks (Nunc) and subcultured by trypsinisation (0.05% trypsin, Gibco1). Culture medium was changed every two to three days. Both serial cultures of cells were used for the determination of the following parameters: time to reach confluency in the standard culture flasks of 75 cm2; the total number of cells at saturation density determined after trypsinisation on the Coulter counter, at increasing passages [3]; total amount of collagen synthesized and deposited by the cells after 72 h, using the colorimetric procedure with PicroSirius red staining [4,5]. This method has the advantage to estimate not only freshly synthesized collagen but also the total amount of collagen accumulated by the cells during the 72 h culture period. Shortly, cells were seeded in six well plates at 5.104 cells/well, after 72 h of culture, cells were washed twice with PBS then fixed for one hour with 1 ml of the Bouin’s solution at room temperature. Following fixation, cells were washed twice with distilled water and then stained with Sirius-red solution (0.5 g Sirius red F3B Gurr BDH, 500 ml saturated aqueous solution of picric acid and a little solid picric acid to ensure saturation) for one hour under shaking at room temperature. Samples were then washed consecutively with distilled water and 0.01 M HCl to remove unbound dye. Then, bound dye was solubilised by incubation in 500 ml of 0.1 M NaOH for one hour under shaking. Samples were transferred in a spectrophotometer cuvette and absorbency read at 550 nm. 2.2. Rhamnose-rich oligo- and polysaccharides All a-l-rhamnose-rich oligo- and polysaccharide preparations (RROP-s) were obtained from Solabia - BioEurope (Pantin, France). RROP-1 (commercial name Rhamnosoft1) is obtained from Klebsiella pneumoniae strains. The other polysaccharide of 5 kDa, RROP-3 (commercial name
68
V. Ravelojaona et al. / Pathologie Biologie 56 (2008) 66–69
Oligosaccharide BEC-291) was obtained from Klebsiella planticola, by hydrolysis of the 45 kDa parent-polysaccharide (RROP-2). Their structure and also their chemical and biological properties were recently described [1,6]. These two rhamnose rich-oligo- and polysacharides, prepared as described in [1,6], were added to the cultures, at the concentration of 20 mg/ml. The results of collagen synthesis were estimated on a calibration curve obtained with collagen type I, as described in [5]. As a matter of fact, this is the major collagen of both tissues: skin and cornea. 2.3. Statistics Statistical evaluation of the results was done with the Mann– Whitney distribution free U-test on results obtained with a minimal of four parallel experiments.
Fig. 1. Evolution of collagen synthesis by fibroblasts and keratocytes during the successive passages. Collagen synthesis expressed as microgram of collagen type I secreted by 106 cells in 48 h.
3. Results As shown on Table 1, the number of cells at saturation density was higher for keratocytes than for fibroblasts. The number of cells at saturation density declined slightly with passage number for both keratocytes and fibroblasts. The number of dead cells was compared for these two cell types at the twelfth passage. It was significantly more important (over 7% of total seeded cells) for keratocytes than for fibroblasts (0.56% of total seeded cells). As shown on Fig. 1, keratocytes secreted significantly lower concentrations of collagen than fibroblasts at the earliest passage (7th): about 3000 mg/106 cells for fibroblasts and only 1337 mg/106 cells for keratocytes ( p < 2.1010). This difference decreased at passage 11 (4720 mg/106 cells for fibroblasts and 4007 mg/106 cells for keratocytes ( p < 2.104)) and decreased even more at passage 12 (5715 mg/106 cells for fibroblasts and 5192 mg/106 cells for keratocytes ( p < 0.0024)). This suggests that close to the normal corneal phenotype, keratocytes secrete less collagen than fibroblasts. With increasing passages, the corneal phenotype is progressively lost and keratocytes tend to secrete as much collagen as skin fibroblasts.
Fig. 2. Modulation of collagen synthesis by RROP-s for keratocytes and fibroblasts at three increasing passages. Notice the differences in the modulation of biosynthesis in presence of RROP-1 and RROP-3. Several of the results differ nonsignificantly because of the dispersion of the results. Significant differences between values obtained with a RROP and values obtained in control cultures, are indicated by * ( p < 0.05). Results are expressed in percentage as comparison to the control cultures.
As shown on Fig. 2, the action of rhamnose-rich oligo- and polysaccharides acted differently on these two cell types at the earliest passage (no 7). Fibroblasts were hardly affected or slightly inhibited by the RROP-1 but keratocytes were quite
Table 1 Cell death and cell density at confluency for keratocytes and fibroblasts at three consecutive passages at four days after passing Number of passage
Keratocytes Percentage of cell death
Number of cells at confluency
6.46% 7.33% (***)|
1.86 10+5 (NS)| vs. fibroblasts 11th passage*vs. 12th passage*** 1.71 10+5 (***)| vs. fibroblasts 1.16 10+5 (NS) | vs fibroblasts
7 11 12
Fibroblasts Percentage of cell death
Number of cells at confluency
0.56%
1.56 10+5 vs. 11th passage* vs. 12th passage** 1.14 10+5 9.89 10+4
Cell death is expressed as a percentage of dead cells as compared to total number of cells. Significance of the results was estimated with the Mann–Whitney distribution free U-test on results obtained with a minimal of four to six parallel experiments. Values marked with a bold clover correspond to the comparison between keratocytes and fibroblasts; results obtained at the same passage. Other values correspond to the comparison between the values obtained at the earliest passages (no 7) and those obtained at later passages (no 11 or no 12). Significance of differences between keratocytes and fibroblasts and for the decrease of the saturation with increasing passages: * p < 0.05. ** p < 0.01. *** p < 0.001.NS: no significance.
V. Ravelojaona et al. / Pathologie Biologie 56 (2008) 66–69
strongly stimulated by RROP-3 and less strongly (near about +10%) by RROP-1. This rhamnose-rich oligosaccharide, RROP-3 increased much more (+35.5%) collagen biosynthesis by keratocytes than by fibroblasts (+1.8%) at passage 7. At the later passages studied (11th and 12th), these differences were attenuated. According to these results, the differences of RROP action on these two cell types are strongest at the earliest passage and decrease without completely disappearing at later passages. Only few of these differences are, however, significant because of the dispersion of individual values and keep, therefore, only an indicative value. 4. Discussion The above-presented results on saturation density of both cell types and on the proportion of dead cells floating on top of the culture fluid (Table 1) suggest subtle differences between these two cell types. More keratocytes than fibroblasts were found at saturation density, but this difference decreased with passage number. The proportion of dead cells at the twelfth passage was still higher for keratocytes than for fibroblasts. The composition of the ECM for both tissues, cornea and skin, is quite different: reflection of the differences of their matrix-synthetic phenotype. Cornea is transparent, as the result of the ordered deposition of its collagen fibers during embryonic development [7]. These fibers are composed essentially of collagen types I, V and VI; this last collagen type represents close to 20% of total corneal collagens [8]. Their rate of synthesis and turnover could be measured on calfcornea explant cultures and shown to be surprisingly rapid [9]. Collagen type III was shown to increase only in inflammatory conditions [10]. Dermal collagen of human skin is also rich in type I collagen, but also in type III, besides lower proportion of type V. Type VI collagen is hardly detectable. With age, the ratio of type III to type I collagen increases [11]. Comparable differences between cornea and skin were shown for the GAGPG-compositions of both tissues [12]. The passage-dependent phenotype shift in GAG-PG-biosynthesis was studied for keratocytes and fibroblasts both in cell and explant cultures. With increasing passages, there was a shift in the proportion of individual GAG-s synthesized but there was no decline of total synthesis for both cell types. Our present results suggest also a shift in collagen phenotype with increasing passage number for keratocytes but no decrease of synthesis up to the twelfth passage. Similar results were obtained with fibroblasts up to the twenty-fifth passage: no decline of total synthesis [13]. The relatively large difference in total collagen synthesis by keratocytes (of about 55%) below the level of fibroblasts, at the seventh passage, decreased to about 15% at passage 11 and to hardly 10% at passage 12. This progressive increase of
69
collagen synthesis by keratocytes might well be accompanied by the progressive disappearance of the qualitative differences between these two cell types, essentially by the decrease of collagen type VI. This, however, remains to be confirmed. Taken together with our data on the progressive shift of GAGPG- synthesis by increasing passages of keratocytes, the progressive loss of the corneal phenotype of keratocytes during in vitro ‘‘aging’’ seems to be confirmed. Acknowledgement Supported by the Institut DERM, Paris and by NATURA Ltda, São Paulo, Brazil. References [1] Andrès E, Molinari J, Péterszegi G, Mariko B, Ruszova E, Velebny V, et al. Pharmacological properties of rhamnose-rich polysaccharides, potential interest in age-dependent alterations of connective tissues. Pathol Biol 2006;54:420–5. [2] Isnard N, Sabatier P, Robert AM, Robert L, Renard G. MMP-type endopeptidase activity in the cornea. Its evolution during organ culture storage at the Eye Bank. Effect of hyaluronan. Pathol Biol 2005;53:424–9. [3] Péterszegi G, Molinari J, Ravelojaona V, Robert L. Effect of advanced glycation end-products on cell proliferation and cell death. Pathol Biol 2006;54:396–404. [4] Tullberg-Reinert H, Jundt G. In situ measurement of collagen synthesis by human bone cells with a Sirius red-based colorimetric microassay: effects of transforming growth factor beta 2 and ascorbic acid 2-phosphate. Histochem Cell Biol 1999;112:271–6. [5] Péterszegi G, Andrès E, Molinari J, Ravelojaona V, Robert L. Effect of cellular aging on collagen biosynthesis. I-methodological consideration and pharmacological applications. Arch Gerontol Geriatr (2007), doi: 10.1016/j.archger, 2007.08.019. [6] Ravelojaona V, Molinari J, Robert L. Protection by rhamnose-rich polysaccharides against the cytotoxicity of Maillard reaction products. Biomed Pharmacother 2006;60:359–62. [7] Robert L, Reyss-Brion M, Junqua S, Salaun J. Composition chimique de la cornée de l’embryon de poulet et de poule adulte. C R Acad Sci III 1969;269:491–3. [8] Robert L, Legeais JM, Robert AM, Renard G. Corneal Collagens. Pathol Biol 2001;49:353–63. [9] Kern P, Menasche M, Robert L. Relatives rates of biosynthesis of collagen type I, type V and type VI in calf cornea. Biochem J 1991;274:615–7. [10] Menasche M, Junqua S, Payrau P, Robert L. Étude de la régulation de la biosynthèse des macromolécules de la matrice intercellulaire de la cornée. Pathol Biol 1975;23:705–9. [11] Labat-Robert J, Kern P, Robert L. Biomarkers of connective tissue aging: biosynthesis of fibronectin, collagen type III and elastase. Ann N Y Acad Sci 1992;673:16–22. [12] Isnard N, Fodil I, Robert L, Renard G. Modulation of cell-phenotype during in vitro aging. Glycosaminoglycan biosynthesis by skin fibroblasts and corneal keratocytes. Exp Gerontol 2002;37:1377–85. [13] Ravelojaona V, Robert L, Robert AM. Effect of cellular aging on collagen biosynthesis. II-collagen synthesis and deposition by a human skin fibroblast strain over 25 passages. Arch Gerontol Geriatr (2007), doi: 10.1016/j. archger. 2007.08.017.