BIOCHEMICAL
Vol. 181, No. 3, 1991 December
31, 1991
DIFFERENT PROLIFERATIVE
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 1201-1207
RESPONSES OF
PERIPORTAL AND PRRIVENOUS
HEPATOCYTES TO EGF
Rolf
Gebhardt'
Physiologisch-chemisches D-7400 Received
November
14,
and Dirk Institut Tubingen,
Jonitza der FRG
UniversitZt,
1991
Stimulation of DNA synthesis by EGF was compared in and perivenous hepatocyte populations. periportal Periportal hepatocytes responded to EGF more sensitive (IC,values 20 vs 75 rig/ml) and with a higher maximal stimulation hepatocytes with respect to (420 vs 290%) than perivenous both ['Hlthymidine incorporation and labeling index. The glutamine synthetase-positive hepatocytes responded much less to EGF than did the perivenous cells in general. The simultaneous presence of insulin increased the sensitivity for EGF predominantly in the periportal hepatocytes. These inherent differences in the growth potential of hepatocytes from different acinar localizations may contribute to different growth patterns across the lobules in normal and regenerating liver.
suNNARy:
cultured
0 1991
Academic
Press,
Inc.
Regenerative growth of hepatocytes following partial hepatectomy is known to proceed in several sequential waves of DNA synthesis and mitosis starting in the periportal and ending in the perivenous zone of the liver acinus (l-3). Although the characteristic liver microcirculation may play some role in this wave-like progression, evidence is accumulating that considerable differences in the growth potential exist between hepatocytes arranged along the porto-central axis of the sinusoids (4). Using glutamine synthetase (EC 6.3.1.2, GS) as a marker for the most distal (perivenous) hepatocytes (5) strong differences were noted in the growth response to various growth promoting conditions including exposure to EGF between
To whom reprint requests Abbreviations: GS, glutamine perivenous. l
should be synthetase;
1201
addressed. pp, periportal;
pv,
0006-291x/91 $1.50 Copyrighr 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol.
181, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
GS-negative and GS-positive hepatocytes in vitro (6) and in vivo (7). EGF has been found to be a potent mitogen for cultured hepatocytes (for review: 8) and after injection in vivo (9). Recently, a striking heterogeneity in the acinar distribution of the EGF receptor was reported (10) characterized by a steep porto-central gradient of low-affinity receptors and a periportal localization only of high-affinity receptors suggesting the different sensitivity of these cell populations to EGF and TGFa. In the present study we have investigated the different sensitivity of periportal and perivenous hepatocytes isolated by technique (11,12) with respect to the digitonin/collagenase the stimulation of DNA synthesis by EGF in primary culture. MATERIALS
AND METHODS isolation and cultivation: Adult male Sprague-Dawley rats with body weights between 220 and 290 g were used in these experiments. Hepatocytes from whole livers were isolated as described (6). Periportal and perivenous hepatocyte subpopulations were isolated by the digitonin/collagenase perfusion technique of Lindros and Penttill (11) modified according to Burger et al. (12). Hepatocytes were suspended in Williams medium E containing 2 mM glutamine, penicillin (5 U/ml) and streptomycin (50 pg/ml) and seeded on collagen-coated Petri dishes ( 35 mm) or six-well plates (Greiner, Niirtingen) at low cell density (25,000 cells/cm2). Collagen-coated cover-slips were used as support in experiments using autoradiography. Newborn calf serum (5%) was present only during the first 2 h of cultivation. The culture medium was changed after 2 h, and subsequently every 24 h. EGF (from mouse submaxillary glands, grade II, Boehringer, Mannheim) was used at concentrations between 0.5 and 200 rig/ml. Insulin was used at a concentration of 5*10-' M. All growth factors were added after 2 h and were present throughout cultivation. AnalvtiCal Drocedures: In experiments assessed by autoradiography or scintillation counting [3H]thymidine (specific acti40 to 60 Ci/mmol) was added to the culture medium at a vity, concentration of 10 or 5 mCi/ml, respectively. Scintillation counting was performed as described (6). For autoradiography, cells were fixed on cover-slips with ice-cold 3.5% paraformaldehyde in PBS, followed by 98% ethanol. Further processing was performed as described (6). Immunocytochemical localization of GS in conjunction with autoradiography was performed using a rabbit anti-rat liver GS antiserum (1:350) as described (6).
HeDatOCYte
RESULTS Incorporation of [3H]thymidine into DNA was low in control cultures (low cell density (13)) of all cell populations after 48 for periportal hepatocytes higher h, but was significantly 1202
Vol.
BIOCHEMICAL
181, No. 3, 1991
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
TABLE 1.
Effect of EGF on ['Hlthymidine incorporation of different hepatocyte populations
Culture Time (h)
EGF
72
DNA
['H]Thymidine Incorporation (dpm/pg protein)a
(w/ml) Wholeb
48
into
PP
PV
None
150 f
22
187 +
34
117 f
19
2
188 f
37
234 f
43
137 +
25
80
616 +
59
804 +
68
255 f
63
None
389 f
76
448 f
67
304 +
53
405 f
48
2
n.d.
622 _+ 53
80
1971 + 177
2508 f 221
1124 + 138
a values represent means +- SD of duplicate determinations from 3 to 4 experiments. b Hepatocyte preparations: whole, all hepatocytes; PP, periportal cells; PV, perivenous cells.
for perivenous cells (Table 1). Addition of EGF to the culture medium led to a marked stimulation preferentially in periportal hepatocytes. As should be expected, whole hepatocyte preparations showed values in between those of the subpopulations. After 72 h, thymidine incorporation in controls was about 2.5-fold higher (Table 1) compared to 48 h. Stimulation in response to EGF was even more enhanced. At all concentrations incorporation was significantly higher in periportal hepatocytes. Detailed analysis of the concentration dependence in the range between 0.5 and 200 rig/ml carried out after 48 h revealed sigmoidal dose-response curves on a log-scale than
for both, the periportal and the perivenous hepatocyte population (Fig. 1). However, the periportal hepatocytes responded more sensitive (IC,,-values: 20 vs 75 rig/ml) and with a higher maximal response (420 + 35% vs 290 + 25%) than perivenous hepatocytes. A similar difference was found if the labeling indices were determined (Fig. 2). Most interestingly, the GS-positive hepatocytes which represent the most distal localized perivenous cells showed a very low sensitivity for EGF (Fig. 2). This is in line with earlier observations concern1203
Vol.
181,
No.
BIOCHEMICAL
3, 1991
PP
J’
AND
,’F P,’
_--_
4
#’ #’ ,’ 250
8’
-
,’ i
PV
,,g
,’ ;i
_.___s s
-* ,’
BIOPHYSICAL
I
40
g
30-
c2 -0 -c
20
z 5 P 3
lo-
RESEARCH
COMMUNICATIONS
-
-
o-
02
01
EGF
concentration
(rig/ml)
Fig. 1. Btimulation of [%]thymidine incorporation into DNA by ( 0 ) or perivedifferent concentrations of EGF. Periportal nous ( 0 ) hepatocytes were cultured in the presence of different concentrations of EGF for 48 h and were exposed to ['Hlthymidine and processed for scintillation counting as described in Methods. Data represent means f SD of duplicate determinations of 3 to 4 experiments except for the values marked by s which represent means k range of duplicate determinations from a single experiment. of EGF on the labeling index of different Fig. 2. Influence hepatocyte subpopulations. Periportal ( 0 ) or perivenous (0) hepatocytes were cultured on cover slips in the presence of different concentrations of EGF and were fixed after 48 h. GSpositive cells ( n ) were discriminated by immunocytochemistry prior to autoradiography. Data represent means + SD of countings of 200 to 300 cells. In the case of GS-positive cells, at least 250 such cells were counted.
ing the growth potential of this particular hepatocyte population (4,6). As shown in Table 2, the simultaneous presence of insulin considerably enhanced the stimulation of DNA synthesis by EGF apparently by increasing the sensitivity for this growth factor. In case of the periportal hepatocytes, a strong increase was found at EGF concentrations as low as 2 rig/ml and the maxIn case of the perivenous imal response was seen at 10 rig/ml. hepatocytes a continuous increase between 2 and 80 rig/ml was obvious. Again, the maximal response was lower than that of the periportal cells. Interestingly, insulin alone led to a higher stimulation in perivenous hepatocytes (Table 2). DISCUSSION
The ting tes
results
of
inherent
along
this
study
differences
the
porto-central
add in
to
the
axis 1204
the
recent
growth
of
the
potential
liver
findings of
lobules
suggeshepatocy-
(4,6,7)
Vol.
BIOCHEMICAL
181, No. 3, 1991
TABLE 2.
EGF
Synergistic effects ing index of cultured
Insulin
(WW
0 2 10
80
+
AND BIOPHYSICAL
of insulin hepatocyte
Labeling
RESEARCH COMMUNICATIONS
and EGF on the populationsa
label-
index at 48 h (%)b
PP
PV
2+1 4f2
12 + 4
1 f 1
+
5+2 32 2 3
2 + 1 12 + 2
+
17 f 4 56 f 6
34 + 7
+
26 2 5 51 f 8
14 2 3 43 2 5
9+3
a Periportal (PP) or perivenous (PV) hepatocytes were cultured on coverslips in the presence of different concentrations of EGF with and without addition of insulin (5.10.' M) as specified in Methods. Labeling with ['H]thymidine was carried out during the last 24 h. After 48 h the hepatocytes were fixed and processed for autoradiography. bData represent means f SD of countings of 200 to 300 cells.
As found herein, the sensitivity of DNA synthesis to EGF decreases along this axis and even the maximal response is higher in periportal than in perivenous hepatocytes. These results fit nicely to the heterogeneous distribution of EGF-receptor populations reported recently (10). It has to be kept in mind, that receptor types and concentrations on however, the surface of the hepatocytes may change during cultivation (14). On the other hand, the qualitative similarity of the results obtained after 48 h and 72 h indicates stable zonal differences, while the quantitative enhancement during this period reflects the long lag-phase of the cultured cells observed also for other growth factors (6). The synergistic effect of insulin and EGF on DNA synthesis has often been described for cultured hepatocytes (15,16). It appears from this study that the basis of this synergism is the rise in sensitivity against EGF induced by insulin. This effect may reinforce the preferential growth stimulation of periportal hepatocytes at low concentrations of EGF present in normal plasma. This may be of considerable importance for the 1205
Vol.
181, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
initiation of hepatocyte growth, since plasma levels of EGF do not change, for instance, in response to partial hepatectomy (8). Indeed, the different priming of the different hepatocytes after partial hepatectomy seems more important in vivo than other differences in the growth potential among hepatocytes. Since EGF and TGFa act via the same receptor, the observed differences in sensitivity might also hold in the case of TGFa which is believed to be produced in regenerating hepatocytes in vivo and act in an autocrine manner (17). The low but significant higher labeling index of perivenous hepatocytes in response to insulin alone fits to the assumption advanced in the concept of metabolic zonation (18) that these cells might be preferential targets of insulin. However, this does not seem to hold for all perivenous hepatocytes, since the GS-positive hepatocyte subpopulation located at the perivenous end plate behaves differently as emphasized by the determinations of the labeling index of GS-positive hepatocytes herein and in other studies (4,6). In conclusion, the described differences in the stimulation of DNA synthesis in hepatocytes of different origin in the liver lobules may play a role in the different contribution of these cells to normal hepatocyte renewal known from in vivo studies [19-211 as well as in the generation of sequential waves of DNA synthesis and mitosis running across the lobules after partial hepatectomy. The higher sensitivity of periportal cells may allow a much quicker initiation of growth in these cells both, by exogenous and endogenous factors. ACKNOWLEDGMENTS: The excellent H. Fitzke and Mr. C. Mayer is work was supported by a grant meinschaft.
technical assistance of Mrs. gratefully acknowledged. This of the Deutsche Forschungsge-
REFERENCES 1) 2)
3) 4) 5)
6)
7) 8)
Grisham, J.W. (1962) Cancer Res. 22, 842-849. Rabes, H., Wirshing, R., Tuczek, H., et al. (1976) Cell Tissue Kinwet 9, 517-532. Stbcker, E. (1966) verb. Dtsch. Ges. Pathol. 50, 53-74. Gebhardt, R. (1988) Stand. J. Gastroenterol. 23, 8-18. Gebhardt, R. and Mecke, D. (1983) EMBO J. 2, 567-570. Gebhardt, R., Cruise, J., Houck, K.A., Luetteke, N.C., Novotny, A., Thaler, F. and Michalopoulos, G. (1986) Differentiation 33, 45-55. Gebhardt, R. (1990) Cancer Res. 50, 4407-4410. Michalopoulos, G.K. (1990) FASEB J. 4, 176-187. 1206
Vol.
181, No. 3, 1991
9) 10) 11) 12) 13) 14) 15) 16)
17) 18)
19) 20) 21)
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Bucher, N.L.R. (1982) In Cold Spring Harbor Conferences on Cell Proliferation Vo1.9, p.15-26. Marti, U. and Gebhardt R. (1991) Eur. J. of Cell Biol. 55, 158-164. Lindros, K.O. and PenttilB;, K.E. (1985) Biochem. J. 228, 757-760. Burger, H.J., Gebhardt, R., Mayer, C. and Mecke D. (1989) Hepatology 9, 22-28. Nakamura, T., Tomita, Y. and Ichihara, A. (1983) J. Biothem. 94, 1029-1035. Wollenberg, G-K., Harris, L., Farber, E. and Hayes, M.A. (1989) Lab. Invest. 60, 254-259. Tomita, Y., Nakamura, T. and Ichihara, A. (1981) Exp. Cell Res. 135, 363-371. Michalopoulos, G. Cianciulli, H.D., Novotny, A.R., Klingerman, A-D., Strom, S.C. and Jirtle, R.L. (1982) Cancer Res. 42, 4673-4682. Mead, J.E. and Fausto, N. (1989) Proc. Natl. Sci. 86, 1558-1562. Jungermann, K. and Katz, N. (1989) Physiol. Rev. 69, 708764. Grisham J.W. (1962) Cancer Res. 22, 842-849. Zajicek G., Oren, J. and Weinreb, M. (1985) Liver 5, 293300. Savchenko, T.V. and Romanov, I.A. (1989) Biull. Eksp. Biol. Med. 107, 227-229.
1207
Vol. 181, No. 3, 1991 December
31, 1991
DIFFERENT PROLIFERATIVE
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 1201-1207
RESPONSES OF
PERIPORTAL AND PRRIVENOUS
HEPATOCYTES TO EGF
Rolf
Gebhardt'
Physiologisch-chemisches D-7400 Received
November
14,
and Dirk Institut Tubingen,
Jonitza der FRG
UniversitZt,
1991
Stimulation of DNA synthesis by EGF was compared in and perivenous hepatocyte populations. periportal Periportal hepatocytes responded to EGF more sensitive (IC,values 20 vs 75 rig/ml) and with a higher maximal stimulation hepatocytes with respect to (420 vs 290%) than perivenous both ['Hlthymidine incorporation and labeling index. The glutamine synthetase-positive hepatocytes responded much less to EGF than did the perivenous cells in general. The simultaneous presence of insulin increased the sensitivity for EGF predominantly in the periportal hepatocytes. These inherent differences in the growth potential of hepatocytes from different acinar localizations may contribute to different growth patterns across the lobules in normal and regenerating liver.
suNNARy:
cultured
0 1991
Academic
Press,
Inc.
Regenerative growth of hepatocytes following partial hepatectomy is known to proceed in several sequential waves of DNA synthesis and mitosis starting in the periportal and ending in the perivenous zone of the liver acinus (l-3). Although the characteristic liver microcirculation may play some role in this wave-like progression, evidence is accumulating that considerable differences in the growth potential exist between hepatocytes arranged along the porto-central axis of the sinusoids (4). Using glutamine synthetase (EC 6.3.1.2, GS) as a marker for the most distal (perivenous) hepatocytes (5) strong differences were noted in the growth response to various growth promoting conditions including exposure to EGF between
To whom reprint requests Abbreviations: GS, glutamine perivenous. l
should be synthetase;
1201
addressed. pp, periportal;
pv,
0006-291x/91 $1.50 Copyrighr 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol.
181, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
GS-negative and GS-positive hepatocytes in vitro (6) and in vivo (7). EGF has been found to be a potent mitogen for cultured hepatocytes (for review: 8) and after injection in vivo (9). Recently, a striking heterogeneity in the acinar distribution of the EGF receptor was reported (10) characterized by a steep porto-central gradient of low-affinity receptors and a periportal localization only of high-affinity receptors suggesting the different sensitivity of these cell populations to EGF and TGFa. In the present study we have investigated the different sensitivity of periportal and perivenous hepatocytes isolated by technique (11,12) with respect to the digitonin/collagenase the stimulation of DNA synthesis by EGF in primary culture. MATERIALS
AND METHODS isolation and cultivation: Adult male Sprague-Dawley rats with body weights between 220 and 290 g were used in these experiments. Hepatocytes from whole livers were isolated as described (6). Periportal and perivenous hepatocyte subpopulations were isolated by the digitonin/collagenase perfusion technique of Lindros and Penttill (11) modified according to Burger et al. (12). Hepatocytes were suspended in Williams medium E containing 2 mM glutamine, penicillin (5 U/ml) and streptomycin (50 pg/ml) and seeded on collagen-coated Petri dishes ( 35 mm) or six-well plates (Greiner, Niirtingen) at low cell density (25,000 cells/cm2). Collagen-coated cover-slips were used as support in experiments using autoradiography. Newborn calf serum (5%) was present only during the first 2 h of cultivation. The culture medium was changed after 2 h, and subsequently every 24 h. EGF (from mouse submaxillary glands, grade II, Boehringer, Mannheim) was used at concentrations between 0.5 and 200 rig/ml. Insulin was used at a concentration of 5*10-' M. All growth factors were added after 2 h and were present throughout cultivation. AnalvtiCal Drocedures: In experiments assessed by autoradiography or scintillation counting [3H]thymidine (specific acti40 to 60 Ci/mmol) was added to the culture medium at a vity, concentration of 10 or 5 mCi/ml, respectively. Scintillation counting was performed as described (6). For autoradiography, cells were fixed on cover-slips with ice-cold 3.5% paraformaldehyde in PBS, followed by 98% ethanol. Further processing was performed as described (6). Immunocytochemical localization of GS in conjunction with autoradiography was performed using a rabbit anti-rat liver GS antiserum (1:350) as described (6).
HeDatOCYte
RESULTS Incorporation of [3H]thymidine into DNA was low in control cultures (low cell density (13)) of all cell populations after 48 for periportal hepatocytes higher h, but was significantly 1202
Vol.
BIOCHEMICAL
181, No. 3, 1991
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
TABLE 1.
Effect of EGF on ['Hlthymidine incorporation of different hepatocyte populations
Culture Time (h)
EGF
72
DNA
['H]Thymidine Incorporation (dpm/pg protein)a
(w/ml) Wholeb
48
into
PP
PV
None
150 f
22
187 +
34
117 f
19
2
188 f
37
234 f
43
137 +
25
80
616 +
59
804 +
68
255 f
63
None
389 f
76
448 f
67
304 +
53
405 f
48
2
n.d.
622 _+ 53
80
1971 + 177
2508 f 221
1124 + 138
a values represent means +- SD of duplicate determinations from 3 to 4 experiments. b Hepatocyte preparations: whole, all hepatocytes; PP, periportal cells; PV, perivenous cells.
for perivenous cells (Table 1). Addition of EGF to the culture medium led to a marked stimulation preferentially in periportal hepatocytes. As should be expected, whole hepatocyte preparations showed values in between those of the subpopulations. After 72 h, thymidine incorporation in controls was about 2.5-fold higher (Table 1) compared to 48 h. Stimulation in response to EGF was even more enhanced. At all concentrations incorporation was significantly higher in periportal hepatocytes. Detailed analysis of the concentration dependence in the range between 0.5 and 200 rig/ml carried out after 48 h revealed sigmoidal dose-response curves on a log-scale than
for both, the periportal and the perivenous hepatocyte population (Fig. 1). However, the periportal hepatocytes responded more sensitive (IC,,-values: 20 vs 75 rig/ml) and with a higher maximal response (420 + 35% vs 290 + 25%) than perivenous hepatocytes. A similar difference was found if the labeling indices were determined (Fig. 2). Most interestingly, the GS-positive hepatocytes which represent the most distal localized perivenous cells showed a very low sensitivity for EGF (Fig. 2). This is in line with earlier observations concern1203
Vol.
181,
No.
BIOCHEMICAL
3, 1991
PP
J’
AND
,’F P,’
_--_
4
#’ #’ ,’ 250
8’
-
,’ i
PV
,,g
,’ ;i
_.___s s
-* ,’
BIOPHYSICAL
I
40
g
30-
c2 -0 -c
20
z 5 P 3
lo-
RESEARCH
COMMUNICATIONS
-
-
o-
02
01
EGF
concentration
(rig/ml)
Fig. 1. Btimulation of [%]thymidine incorporation into DNA by ( 0 ) or perivedifferent concentrations of EGF. Periportal nous ( 0 ) hepatocytes were cultured in the presence of different concentrations of EGF for 48 h and were exposed to ['Hlthymidine and processed for scintillation counting as described in Methods. Data represent means f SD of duplicate determinations of 3 to 4 experiments except for the values marked by s which represent means k range of duplicate determinations from a single experiment. of EGF on the labeling index of different Fig. 2. Influence hepatocyte subpopulations. Periportal ( 0 ) or perivenous (0) hepatocytes were cultured on cover slips in the presence of different concentrations of EGF and were fixed after 48 h. GSpositive cells ( n ) were discriminated by immunocytochemistry prior to autoradiography. Data represent means + SD of countings of 200 to 300 cells. In the case of GS-positive cells, at least 250 such cells were counted.
ing the growth potential of this particular hepatocyte population (4,6). As shown in Table 2, the simultaneous presence of insulin considerably enhanced the stimulation of DNA synthesis by EGF apparently by increasing the sensitivity for this growth factor. In case of the periportal hepatocytes, a strong increase was found at EGF concentrations as low as 2 rig/ml and the maxIn case of the perivenous imal response was seen at 10 rig/ml. hepatocytes a continuous increase between 2 and 80 rig/ml was obvious. Again, the maximal response was lower than that of the periportal cells. Interestingly, insulin alone led to a higher stimulation in perivenous hepatocytes (Table 2). DISCUSSION
The ting tes
results
of
inherent
along
this
study
differences
the
porto-central
add in
to
the
axis 1204
the
recent
growth
of
the
potential
liver
findings of
lobules
suggeshepatocy-
(4,6,7)
Vol.
BIOCHEMICAL
181, No. 3, 1991
TABLE 2.
EGF
Synergistic effects ing index of cultured
Insulin
(WW
0 2 10
80
+
AND BIOPHYSICAL
of insulin hepatocyte
Labeling
RESEARCH COMMUNICATIONS
and EGF on the populationsa
label-
index at 48 h (%)b
PP
PV
2+1 4f2
12 + 4
1 f 1
+
5+2 32 2 3
2 + 1 12 + 2
+
17 f 4 56 f 6
34 + 7
+
26 2 5 51 f 8
14 2 3 43 2 5
9+3
a Periportal (PP) or perivenous (PV) hepatocytes were cultured on coverslips in the presence of different concentrations of EGF with and without addition of insulin (5.10.' M) as specified in Methods. Labeling with ['H]thymidine was carried out during the last 24 h. After 48 h the hepatocytes were fixed and processed for autoradiography. bData represent means f SD of countings of 200 to 300 cells.
As found herein, the sensitivity of DNA synthesis to EGF decreases along this axis and even the maximal response is higher in periportal than in perivenous hepatocytes. These results fit nicely to the heterogeneous distribution of EGF-receptor populations reported recently (10). It has to be kept in mind, that receptor types and concentrations on however, the surface of the hepatocytes may change during cultivation (14). On the other hand, the qualitative similarity of the results obtained after 48 h and 72 h indicates stable zonal differences, while the quantitative enhancement during this period reflects the long lag-phase of the cultured cells observed also for other growth factors (6). The synergistic effect of insulin and EGF on DNA synthesis has often been described for cultured hepatocytes (15,16). It appears from this study that the basis of this synergism is the rise in sensitivity against EGF induced by insulin. This effect may reinforce the preferential growth stimulation of periportal hepatocytes at low concentrations of EGF present in normal plasma. This may be of considerable importance for the 1205
Vol.
181, No. 3, 1991
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
initiation of hepatocyte growth, since plasma levels of EGF do not change, for instance, in response to partial hepatectomy (8). Indeed, the different priming of the different hepatocytes after partial hepatectomy seems more important in vivo than other differences in the growth potential among hepatocytes. Since EGF and TGFa act via the same receptor, the observed differences in sensitivity might also hold in the case of TGFa which is believed to be produced in regenerating hepatocytes in vivo and act in an autocrine manner (17). The low but significant higher labeling index of perivenous hepatocytes in response to insulin alone fits to the assumption advanced in the concept of metabolic zonation (18) that these cells might be preferential targets of insulin. However, this does not seem to hold for all perivenous hepatocytes, since the GS-positive hepatocyte subpopulation located at the perivenous end plate behaves differently as emphasized by the determinations of the labeling index of GS-positive hepatocytes herein and in other studies (4,6). In conclusion, the described differences in the stimulation of DNA synthesis in hepatocytes of different origin in the liver lobules may play a role in the different contribution of these cells to normal hepatocyte renewal known from in vivo studies [19-211 as well as in the generation of sequential waves of DNA synthesis and mitosis running across the lobules after partial hepatectomy. The higher sensitivity of periportal cells may allow a much quicker initiation of growth in these cells both, by exogenous and endogenous factors. ACKNOWLEDGMENTS: The excellent H. Fitzke and Mr. C. Mayer is work was supported by a grant meinschaft.
technical assistance of Mrs. gratefully acknowledged. This of the Deutsche Forschungsge-
REFERENCES 1) 2)
3) 4) 5)
6)
7) 8)
Grisham, J.W. (1962) Cancer Res. 22, 842-849. Rabes, H., Wirshing, R., Tuczek, H., et al. (1976) Cell Tissue Kinwet 9, 517-532. Stbcker, E. (1966) verb. Dtsch. Ges. Pathol. 50, 53-74. Gebhardt, R. (1988) Stand. J. Gastroenterol. 23, 8-18. Gebhardt, R. and Mecke, D. (1983) EMBO J. 2, 567-570. Gebhardt, R., Cruise, J., Houck, K.A., Luetteke, N.C., Novotny, A., Thaler, F. and Michalopoulos, G. (1986) Differentiation 33, 45-55. Gebhardt, R. (1990) Cancer Res. 50, 4407-4410. Michalopoulos, G.K. (1990) FASEB J. 4, 176-187. 1206
Vol.
181, No. 3, 1991
9) 10) 11) 12) 13) 14) 15) 16)
17) 18)
19) 20) 21)
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
Bucher, N.L.R. (1982) In Cold Spring Harbor Conferences on Cell Proliferation Vo1.9, p.15-26. Marti, U. and Gebhardt R. (1991) Eur. J. of Cell Biol. 55, 158-164. Lindros, K.O. and PenttilB;, K.E. (1985) Biochem. J. 228, 757-760. Burger, H.J., Gebhardt, R., Mayer, C. and Mecke D. (1989) Hepatology 9, 22-28. Nakamura, T., Tomita, Y. and Ichihara, A. (1983) J. Biothem. 94, 1029-1035. Wollenberg, G-K., Harris, L., Farber, E. and Hayes, M.A. (1989) Lab. Invest. 60, 254-259. Tomita, Y., Nakamura, T. and Ichihara, A. (1981) Exp. Cell Res. 135, 363-371. Michalopoulos, G. Cianciulli, H.D., Novotny, A.R., Klingerman, A-D., Strom, S.C. and Jirtle, R.L. (1982) Cancer Res. 42, 4673-4682. Mead, J.E. and Fausto, N. (1989) Proc. Natl. Sci. 86, 1558-1562. Jungermann, K. and Katz, N. (1989) Physiol. Rev. 69, 708764. Grisham J.W. (1962) Cancer Res. 22, 842-849. Zajicek G., Oren, J. and Weinreb, M. (1985) Liver 5, 293300. Savchenko, T.V. and Romanov, I.A. (1989) Biull. Eksp. Biol. Med. 107, 227-229.
1207
