MEDICINA - Volumen 56 - Nº 4, 1996
MEDICINA (Buenos Aires) 1996





Departamento de Microbiología, Facultad de Medicina, Universidad de Buenos Aires

Key words: aging astrocytes, dBcAMP treatment, image analysis


Since changes in cell morphology are conspicuous features of astrocyte reaction, we resorted to an histometric approach to evaluate age influence on such morphological response to activating stimuli. To this end, first subcultre of rat brain astrocytes at 1, 9 or 21 days in vitro (DIV) were treated during 2 hs with 1 mM of dBcAMP, a chemical compound known to induce cell differentiation. Following treatment, immunoperoxidase labeling of GFAP, specific marker of astrocyte activation, was carried out. Although total count of GFAP- positive cell foci was greater in treated samples in all times tested, when such cell foci were evaluated by image analysis, differences between perimeter/area ratios of such foci were only statistifically significant at 1 DIV. It may be concluded that while dBcAMP effect is maintained despite astrocyte agining, the morphological pattern of response varies markedly along the observation period.


Cambios morfológicos dependientes de la edad en astrocitos tratados con dBcAMP. Dado que los cambios morfológicos son componente mensurable de la reacción astrocitaria, se recurrió a un procedimiento histométrico para evaluar la influencia de la edad celular en la respuesta a los estimulos diferenciadores.
Asi, subcultivos astrocitarios derivados de encéfalo de rata fueron tratados a los días 1, 9 y 21 de edad con dBcAMP, un compuesto de reconocida acción estimulante sobre ese tipo celular. Consecutivamente a la marcación por inmunoperoxidasa de la proteína gliofibrilar ácida (GFAP), que es marcador específico del astrocito activado, por análisis de imagen se determinó el perímetro y área de los focos celulares positivos. Aunque en todos los tiempos considerados el número total de focos GFAP-positivos fue mayor en los cultivos tratados, la relación perímetro/área de dichos focos sólo arrojó diferencias
significativas al día 1 de desarrollo in vitro. Se concluye que, si bien la acción diferenciadora se mantuvo pese al envejecimiento astrocitario, el perfil morfológico de la respuesta celular varió marcadamente a lo largo del período de observación.

Recibido: 21-VII-1996 Accepted: 3-VII-1996

Postal address: Lic. Diego G. Gicovate, Departamento de Microbiología, Facultad de Medicina, UBA, Paraguay 2155, 1121 Buenos Aires, Argentina


Activated astrocytes, commonly recognized morphologically by their enlarged size and numerous cytoplasmic processes, and immunocytochemically by the massive accumulation of GFAP-positive filaments, arise following any type of CNS injury, whether physical, chemical, immunological, viral or bacterial1. The fact that astroglial responsiveness is maintained in spite of aging, even at times when cell senescence becomes associated with diminished brain functions, is still not understood2.
Since astrogliosis phenomena on the whole organism may somehow be recapitulated by cell cultures affording the advantages of easy manipulation and feasible control, cultured astrocytes has been extensively used notwithstanding the limitations of such in vitro model. In its connection, cell aging has been shown to lead to a gradual increase in spontaneously differentiated cells with a concomitant drop in their division potential3. On the other hand, it is well known that the addition of dibutyryl cyclic AMP (dBcAMP) to the culture medium of immature astrocytes results in their differentiation into reactive-like astrocytes4,
since dBcAMP is capable of inhibiting growth and altering morphology of such cells, thus providing a suitable in vitro counterpart of in situ response to neural injury5, within the limitations of the currently employed criteria to define the reactive state6.
With the aim to evaluating age influence on cell response to dBcAMP, we resorted to astroglial cultures obtained from brains of newborn Wistar rats as previously described7. After 1, 9 or 21 days in vitro (DIV) of first subculture in coverslips of Leighton tubes, cell monolayers were treated with 1 mM dBcAMP in PBS during 2 hs at 37C. Treated samples, as well as matched controls incubated in PBS alone, were harvested, washed 3 times with 0.05 M Tris-saline buffer, and fixed with chilled acid methanol for 20 min at -20C. GFAP labeling was then carried out by the peroxidase-antiperoxidase (PAP method). Accordingly, antisera of rabbit anti-GFAP (Sigma Lab, St Louis, MO, USA) at 1/1600 dilution, goat anti-rabbit (Pell-Freez Biological, Tustin, CA, USA) at 1/200 and PAP-rabbit (Cappel, Cochranville, PA, USA) at 1/200 were consecutively employed. Potential sources of variation in peroxidase development were ruled out by normalized
conditions: 0.03% diaminobenzidine tetrahydrochloride (Fluka Labs, hauppage, NY, USA) plus 0.02% hydrogen peroxide for 10 min, followed by washing with tap water. To evaluate morphological changes in treated cultures vs controls, a suitable software (IMAGE PRO-PLUS 1.1, Media Cybernetics, Silver Spring, MD, USA) was employed. Hardware consisted of a black and white CCD video camara coupled to a Zeiss microscope. Under standardized conditions using a 500/510 nm green filter, the perimeter and area of 20 randomly selected GFAP- positive foci were determined for each sample. Recorded values were entered into an standard spreadsheet in order to calculate means and standard deviations (SD). Regarding two or more stained astrocytes as a labeled focus, such foci were counted in randomly selected microscopic fields at 12 x final magnification. Student t test was used for statistical analysis taking p < 0.05 as significant. As regards histometric analysis of each positive focus, since individual cell coutlines could not clearly be discerned, we resorted to an alternative approach, taking the ratio between perimeter and area of each selected focus. In the case of single GFAP-positive cells, we were able to evaluate growth of astroglial processes by means of a graticule of 13 concentric circles, the firstbeing the most proximal and the 13th the most distal. This graticule was superimposed on the image of isolated GFAP-immunolabed cells visualized on the monitor screen. The perikaryon of each selected stained cell was placed in the center of the graticule and all the intersections produced by stained processes with circles were counted. For each sample, the first 15 single labeled cells to be encountered were measured and data analyzed using a two-way ANOVA.
In this study, we evaluated effects of astrocyte aging on the differentiation- inducing properties of dBcAMP, as expressed by the morphological changes taking place in GFAP-positive cells. Mean total focus count ± SEM was significantly greater in treated cultures (p < 0.01) than in controls at 1 DIV (8 ± 0.9 vs 5 ± 0.6), 9 DIV (10 ± 0.7 vs 6 ± 0.4) and 21 DIV (5 ± 0.4 vs 3 ± 0.4).
As shown in Table 1, the perimeter/area ratio was only significantly greater in treated vs control samples (p < 0.05) at 1 DIV, while at longer times tested,differences failted to reach statistical significance. The increase in perimeter/area ratio recorded in treated cultures at 1 DIV, seems due to cell body retractionrather than changes in focus perimter, since the major changes took place in area alone. The trend towards significance suggested by values recorded at 21 DIV was disregarded, since by that time controls had shown some degree of cell degeneration as evidenced by intracytoplasmic granulation accompanied by an increase in cell detritus.
Although cultures of lower density than employed herein would have allowed a more accurate delimitation of the cells making up GFAP-positive foci, it should not be overlooked that cells grown at high and low densities exhibited distinct morphologies8, and such culture modality may markedly affect astrocyte properties, among them cell response to dBcAMP9. In this connection, when single GFAP-positive cells were evaluated resorting to the graticule and by a two-way ANOVA (Fig. 1), at all times  assayed the single labeled astrocytes from treated samples invariably displayed a greater number of cell processes intersecting with concentric graticule circles 2 and 3, since the first was disregarded because on occasion it was occupied by the cell body itself.
Interestingly enough, cell process branching as depicted by intersections with circles 4 to 10 was greater in controls at 1 DIV but not at later times. It should be pointed out that in treated samples, particularly at 1 DIV, most cell processes intersecting with circles 2-3 readily reached and even surpassed the outer circle. On the basis of results achieved, cell aging failed to exert inhibitory effect on astrocyte response to dBcAMP, as disclosed by the number of GFAP-positive cell foci, although the perimeter/area ratio of such foci was significantly greater at 1 DIV alone. As regards single labeled cells from treated cultures, their morphological pattern was also significantly different at 1 DIV, since by this time
dBcAMP had induced longer and more numerous cell processes but less branching. It may be concluded that while dBcAMP effect is maintained despite astrocyte aging, the morphological pattern of response varies markedly.
Although GFAP expression has proven reduced in many long-term cultures to the extent that their identification becomes difficult10, the capability of such aged cells to re-enter the division cycle11, suggests that cultured astrocytes may undergo a transition in a manner partly resembling reactive astrogliosis in intact brain. Abundant evidence indicates that, in the aging whole organism, the neuroglial proliferative ability arises in a previously quiescent population of cells which may not have divided for prolonged periods of time and which display few markers of cytodifferentiation [for references see2, 11].
However, there are also examples of astrocyte reaction solely disclosed by an hypertrophy occurring in absence of cell division14. At any rate, signals which  rigger aged astrocyte activation remain to be elucidated. In this sense, the in vitro system may represent a valid tool to discern the stereotypic events involved in the aging process. Thus, the long-term astroglial culture becomes a useful substrate to develop procedures intended to speed up, slow down or arrest a glial reaction which, in the whole organism, may behave as beneficial or detrimental depending on the timing13.

Acknowledgements: To Dr. Joaquín García-Estrada, a former fellow at Instituto Cajal, Madrid, who kindly provided the graticule employed as test grid. This work was partly supported by grants from CONICET (Consejo Nacional de Investigaciones Científicas y Técnicas) and University of Buenos Aires.



1. Eng LF. Regulation of glial intermediate filaments in astrogliosis. In: Norenberg MD, Hertz L, Schusboe A, (eds). The biochemical pathology of astrocytes. New York: Alan R Liss, 1988; 79-90.
2. Kozlova M, Kentroti S, Vernadakis A. Influence of culture substrata on the differentiation of advanced passage glial cells in culture from aged mouse cerebral hemispheres. Int J Devl Neurosc 1993; 11: 513-9.
3. Ponten J, Stein WD, Shall S. A quantitative analysis of the aging of human glial cells in culture. J Cell Phys 1983; 117: 342-52.
4. Federoff S, Mc Auley WAJ, Houle JD, Devon RM. Astrocyte cell lineage. V. Similarity of astrocytes that form in the presence of dBcAMP in cultures to reactive astrocytes. J Neurosci Res 1984; 12: 15-27.
5. Federoff S, Ahmed I, Wang E. The relation ship of expression of statin, the nuclear protein of nonproliferating cells to the differentiation and cell cycle of astroglia in cultures and in situ. J Neurosci Res 1990; 26: 1-15.
6. Malhotra SK, Aldskoguis H, Shnitka TK. Biochemical diversity among reactive astrocytes. Dendron 1992; 1: 55-68.
7. Berría MI, Lascano EF. Astrocyte differentiation induced by Junin virus in rat brain cell cultures. Acta Neuropathol 1985; 66: 233-8.
8. Goldman JE, Chiu F-C. Growth kinetics, cell shape, and the cytoskeleton of primary astrocyte cultures. J Neurochem 1984; 42: 175-84.
9. Juurlink BHJ, Hertz L. Plasticity of astrocytes in primary cultures: an experimental tool and a reason for methodological caution. Develop Neurosci 1985; 7: 263-77.
10. Szigeti V, Miller RH. A cell surface antigen expressed by astrocytes and their precursors. Glia 1993; 8: 20-32.
11. Langan TJ, Slater MC. Astrocytes derived from long-term primary cultures recapitulate features of astrogliosis as they re-enter the cell division cycle. Brain Res 1992; 577: 200-9.
12. Hatten ME, Liem RHK, Shelanski ML, Mason CA. Astroglia in CNS injury. Glia 1991; 4: 233-43.
13. Mucke L, Eddleston M. Astrocytes in infectious and immune- mediated diseases of the central nervous system. FASEB J 1993; 7: 1226-32.


Fig. 1. Vertical bars depict the number of intersections of single GFAP- positive astrocytes with concentric circles (from 1 the most proximal to 13 the most distal) of the test grid used to evaluate cell morphology. Each determination corresponds to 15 single labeled cells to be encountered. DIV: days in vitro.

TABLE 1.-- Morphological changes of GFAP-positive foci in dBcAMP treated vs untreated astroglial cultures.

Days in Perim./Nucleus ratio Area/Nucleus ratio Perim./Area ratio

vitro control dBcAMP control dBcAMP control dBcAMP

1 51 ± 31 63 ± 66 518 ±
464 275 ± 124* 0.104 ± 0.1 0.204 ± 0.1*
9 57 ± 37 110 ± 304 781 ±
1005 1535 ± 2874 0.122 ± 0.1 0.210 ± 0.4
21 70 ± 53 185 ± 377 749 ±
686 1016 ± 398 0.097 ± 0.1 0.161 ± 0.3

* Significant differences (p < 0.05) between ratios calculated from
mean values (± SD).