Hepatic microvascular development in relation to the morphogenesis of hepatocellular plates in neonatal rats
Article first published online: 23 SEP 2003
Copyright © 2003 Wiley-Liss, Inc.
The Anatomical Record Part A: Discoveries in Molecular, Cellular, and Evolutionary Biology
Volume 275A, Issue 1, pages 1019–1030, November 2003
How to Cite
McCuskey, R. S., Ekataksin, W., LeBouton, A. V., Nishida, J., McCuskey, M. K., McDonnell, D., Williams, C., Bethea, N. W., Dvorak, B. and Koldovsky, O. (2003), Hepatic microvascular development in relation to the morphogenesis of hepatocellular plates in neonatal rats. Anat. Rec., 275A: 1019–1030. doi: 10.1002/ar.a.10117
- Issue published online: 3 OCT 2003
- Article first published online: 23 SEP 2003
- Manuscript Accepted: 7 JUN 2003
- Manuscript Received: 22 APR 2003
- NIH. Grant Number: 5-PO1-HD26013
The development of hepatic microvascular heterogeneity after birth, and its temporal relationship to the development of parenchymal cell plates have received little attention. As a result, the morphogenesis of some of the parameters contributing to this heterogeneity in suckling and weaned rats was studied as a function of time between postpartum days 4 and 30 using in vivo light microscopic, electron microscopic, and immunocytochemical methods. During the early suckling period, the sinusoid network is highly anastomotic, with little evidence of zonation, and the parenchymal cell plates contain multiple cells and are irregularly arranged throughout the lobule. Sinusoidal endothelial fenestration is sparse at 4 days, but phagocytic Kupffer cell (KC) function already exists and exhibits zonal heterogeneity, with more cells located in the periportal zone. With increasing age, endothelial fenestrae increase and organize as sieve plates. Widened centrilobular radial sinusoids form through a loss (“drop-out”) of intersinusoidal sinusoids (ISS). Concomitantly, the associated cell plates straighten and become one cell thick. Hepatocyte DNA synthesis and mitosis are higher in the periportal zone, which retains thickened cell plates and anastomotic sinusoids. The centrilobular sinusoids may widen to accommodate the increased volume of blood that results from the loss of ISS as well as the increased numbers of periportal sinusoids containing flow that feed these vessels. KC phagocytic activity increases during the suckling period concomitant with an increase of gut-derived endotoxin in the portal blood, which suggests that the KCs may be releasing mediators that affect sinusoid diameter, blood flow, endothelial fenestration, and perhaps parenchymal growth either directly or through the stimulation of growth factors. Anat Rec Part A 275A:1019–1030, 2003. © 2003 Wiley-Liss, Inc.
At birth the liver contains a relatively homogeneous population of parenchymal cells. Hepatocellular structural and functional heterogeneity exhibited by the adult liver (Thurman et al., 1986) appears gradually during postnatal development, and distinct zonation is evident by the time of weaning (LeBouton, 1974, 1976; Asada-Kubota et al., 1982; Jungermann and Katz, 1982; Jungermann, 1986; Thurman et al., 1986). Concomitantly, hepatic microvascular heterogeneity develops (McCuskey, 1967a, b, 1968; Wisse et al., 1983, 1985). Although the latter has received little attention, it is likely an important factor influencing the development of the surrounding parenchyma and microvasculature through mediators produced by Kupffer cells (KCs) (McCuskey et al., 1987; Decker, 1990) or perhaps other sinusoidal lining cells. We studied hepatic microvascular development in the suckling rat a function of time to define the morphogenesis of some of the parameters that contribute to heterogeneity. These include differences between the periportal and centrilobular regions of the hepatic lobule in the number of phagocytic KCs, sinusoid network pattern, sinusoid diameter, and endothelial porosity, and their temporal relationship to the morphogenesis of parenchymal cell plates (including DNA synthesis, mitotic index, and ploidy). While data on changes in ploidy, DNA synthesis, and mitosis during postnatal liver development in the rat have been reported previously (LeBouton, 1974, 1976), we performed similar experiments to be able to correlate the previous findings with data obtained from neonatal rats reared and maintained in similar environmental conditions.
MATERIALS AND METHODS
Sprague-Dawley rats were bred in our colony in the Animal Facilities at the University of Arizona. The experimental protocols were approved by the Institutional Animal Care Committee and were performed in accordance with the National Institutes of Health Guidelines on the Use of Laboratory Animals. Suckling rats were removed from their mothers on days 4, 7, 11, and 14 postpartum, as were weaned 30-day-old pups. Once the pups were removed, they were kept together in a cage placed on a heating pad to maintain body temperature. The animals were not fasted and were used within 4 hr. The livers of the suckling pups were studied directly by in vivo light microscopy, as well as by routine light microscopy and transmission and scanning electron microscopy of fixed preparations. The neonatal livers were compared to the livers of weaned 30-day-old pups and adults.
In Vivo Microscopy
After anesthesia was induced with urethane (2 mg/gm bw,ip), the hepatic microvasculature was studied using established high-resolution in vivo microscopy methods (McCuskey, 1986; McCuskey et al., 1997). Briefly, a compound binocular microscope (Leitz, Wetzlar, Germany) adapted for in vivo microscopy was equipped to provide either transillumination or epiillumination, as well as video microscopy using a silicon vidicon or silicon intensified target (SIT) camera (MTI, Michigan City, IN). The liver was exteriorized through a left subcostal incision and positioned over a window of optical-grade mica in a specially designed heated tray mounted on the microscope stage. The tray was equipped with a drain for irrigating fluids, and the window overlaid a long working distance condenser. The liver was covered by a piece of Saran Wrap, which held it in position and limited movement. Homeostasis was ensured by constant suffusion of the organ with Ringer's solution, which was maintained at body temperature and monitored with a thermistor probe connected to a digital thermometer (Physitemp Instruments, Clifton, NJ). With the 80X / 1.0 water immersion objective (Leitz, Wetzlar, Germany) employed for these studies, the maximum resolution was 0.3–0.5 μm. Microvascular events were observed and recorded for at least 30 sec for subsequent offline analysis using a Sony 0.75″ U-matic video tape recorder (Sony Medical Electronics, Park Ridge, NJ).
The relative adequacy of blood perfusion through the sinusoids was evaluated by counting the number of sinusoids containing blood flow (SCF) in 10 standardized microscopic fields (4,125 μm2). KC function was assessed by observing the phagocytosis of fluorescent 1.0-μm latex particles by individual cells. The latex particles (0.1 ml, 4.5 × 108 particles) were injected into a mesenteric vein with a 30-gauge lymphangiography needle. The distribution and relative number of phagocytic KCs were measured by counting the number of cells that phagocytosed latex particles in the same 10 microscopic fields 15 min later. Since reduced perfusion of individual sinusoids can reduce the delivery of the latex particles to KCs in these vessels, the ratio of KCs that phagocytosed latex particles to SCF (KC/SCF) was used as an overall index of KC phagocytic activity.
Routine transmission and scanning electron microscopic methods (Wisse et al., 1983, 1985; McCuskey, 1986; McCuskey et al., 1997) for liver were used to evaluate the ultrastructural features of the microvasculature and parenchymal cells in perfused, fixed livers. The livers were perfused through the portal vein with 0.1 M cacodylate buffer to aid in washing out blood. This was immediately followed with a fixative containing 1.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4. For transmission electron microscopy (TEM), minced pieces of liver were washed in buffer, postfixed with 1% OsO4 in 0.1 M cacodylate buffer at 4°C, dehydrated through graded alcohols, briefly rinsed in propylene oxide, and embedded in Epon. Thin sections were cut on a Reichert Ultracut E microtome (Reichert Optical Works, Vienna, Austria), examined, and photographed with a Philips CM-12S electron microscope (Philips Electron Optics, Eindhoven, The Netherlands). For scanning electron microscopy (SEM), pieces of perfused, fixed livers were fractured, critical-point-dried, sputter-coated with gold, and examined with the Philips CM-12S in the SEM mode or an ETEC AutoScan scanning electron microscope (ETEC, Hayward, CA). In addition, the injection-corrosion cast technique using Batson #17 methyl methylmethacrylate plastic was used to produce microvascular casts for subsequent evaluation of the three-dimensional angioarchitecture by SEM with the ETEC scanning electron microscope (Wisse et al., 1983, 1985; McCuskey, 1986; Ekataksin et al., 1993a, b, c; McCuskey et al., 1997).
Histology and Immunocytochemistry
Liver specimens for analysis of S-phase cells by immunocytochemistry were fixed for 6 hr in 4% phosphate-buffered depolymerized paraformaldehyde. Specimens for analysis of colchicine-arrested mitoses and binuclearity were fixed for 4 hr in a mixture of 40% formaldehyde (commercial) 12 ml, 100% ethanol 33 ml, picric acid 0.1 g, trichloroacetic acid 0.22 g, and mercuric chloride 0.45 g.
In both cases of fixation, standard paraffin sections were obtained at 6 μm thickness. Hematoxylin and eosin (H&E)-stained sections were used for mitotic and binuclear studies. Sections used for S-phase analysis were subjected to an immunocytochemical procedure to detect injected bromodeoxyuridine (BrdU) followed by staining with H&E.
Colchicine (0.25 μg/g) was injected subcutaneously (interscapular) at 10:00 a.m. and the rats were killed at 2:00 p.m. BrdU (0.05 mg/g) was injected intraperitoneally at 7:00 a.m. and the rats were killed 1 hr later.
The immunocytochemical procedure was essentially the same as that described by Johnson et al. (1992). Trypsinization, followed by acid hydrolysis, was used to denature and unwind DNA to expose the incorporated BrdU to the primary antiserum, which was a monoclonal mouse anti-BrdU (Sigma, St. Louis, MO). The secondary antiserum was a peroxidase-conjugated goat anti-mouse IGG (Cappel, Cochranville, PA). The site of the secondary antiserum in hepatocyte S-phase nuclei was revealed by the presence of polymerized diaminobenzidine in the usual manner. The mitotic, binucleated, and S-phase labeled cells were scored as previously described (LeBouton, 1976).
The data were expressed as the mean ± S.E.M. Statistical analyses were performed using analysis of variance (ANOVA) followed by a Student Newman-Keuls test. A 95% confidence level (P < 0.05) was considered significant.
In Vivo Microscopic Observations
At least five breast-fed animals (4, 7, 11, and 14 days old), and 30-day-old weaned pups were successfully studied by in vivo microscopy (Ekataksin et al., 1993a, b, 1994a, b, 1995; McCuskey et al., 1994). The method permitted us to observe the microcirculation of the blood, as well as some of the structure and function of parenchymal and nonparenchymal cells (including phagocytic function of the KCs (Figs. 1–3). The basic patterns of blood flow and angioarchitecture were consistent with the data obtained from the microvascular casts (see below). At 4 days, the sinusoids were arranged in an anastomotic pattern throughout the lobule (Fig. 4) interspersed between irregularly arranged parenchymal cell plates consisting of several cells. With increasing development, the adult pattern of radially arranged sinusoids was formed in the centrilobular zone, apparently as the result of intersinusoidal sinusoids (ISS) becoming attenuated and “dropping out.” These were observed to be short, narrow vessels with limited or no flow in them (Fig. 2). Concomitantly, the intervening parenchymal cell plates straightened and reorganized toward being one cell thick. Additional microvascular heterogeneity was expressed in the progressive increase with age of the diameters of the radial centrilobular sinusoids. These contrasted with the narrower, anastomotic sinusoids in the periportal zone (Fig. 5), which also retained parenchymal cell plates consisting of multiple cells.
Heterogeneity of KC distribution was observed as early as 4 days, at which time there were more phagocytic KCs in the periportal sinusoids than in the centrilobular sinusoids (Fig. 6). At day 7, KC phagocytic activity was significantly increased, especially in the periportal region. This was maintained during the suckling period (14 days), but was reduced after weaning to day-4 levels by 30 days of age (Fig. 7). This transient increase in phagocytic activity was accompanied by an increase in the number of SCF (Fig. 8), as well as increased levels of gut-derived endotoxin in the portal blood (Table 1).
Corrosion Casts of the Microvasculature by SEM
As seen by in vivo microscopy, anatomoses among sinusoids were abundant between days 4 and 10. As a result, sinusoids in all zones of the lobules appeared to be similar (Fig. 9). The average distance between the portal and central venules (PV-CV) at day 4 was 150–200 μm, and this increased gradually with age. With increasing age, regressive changes (attenuating and “drop-out”) of ISS was most pronounced in the immediate vicinity of the central venules (Fig. 10). As a result, in the 30-day-old and adult animals, the radial sinusoids and circumferential sinusoids were clearly distinguishable (Fig. 11). The periportal sinusoids, however, remained tortuous and highly interconnected, while the centrilobular sinusoids were slightly wider and straighter, with only occasional ISS. Typically, these ISS ran at right angles to the radial sinusoids. They were rudimentary, very narrow (1.5–2.0 μm), and short, and traversed only one single hepatic cell plate (Fig. 12). The ISS were usually seen in the intermediate and centrilobular zones, but also were seen occasionally in the peripheral zone (periportal and midseptal regions). Thus, there appears to be both regionality and zonality as regards ISS occurrence. By 30 days the PV-CV distance had expanded to 250–300 μm.
Inlet portal venules supplying a group of sinusoids that formed hepatic micovascular subunits (HMS) (Ekataksin et al., 1993a, b, c, 1994a, b, 1995) were found in all age groups (Fig. 13). Their topographic relations were basically similar to those in adults. When the portal vein was accompanied by an hepatic artery, the arterio-portal anastomosis was likely to be found at an inlet venule or the terminal/preterminal portal venule, rather than more proximally at the larger trunk (Fig. 14). Arterioles that directly and solely supplied sinusoids, without anastomosing with an inlet venule, were not seen. Either they are rare or they are difficult to demonstrate by the microcorrosion-cast technique.
TEM and SEM examinations of the cellular development of the sinusoidal wall indicated that at day 4 sinusoidal endothelial fenestrae were beginning to be organized as sieve plates, which were more abundant by day 30 (Figs. 15–19). However, some fenestrae appeared to be bridged by diaphragms (Fig. 17). KCs were interdigitated into the sinusoidal wall or were attached to the lumenal surface, while ablumenal stellate cells were present in the space of Disse, where little or no collagen fibrils or extracellular matrix was observed (Fig. 16). Between 7 and 14 days, the KCs were enlarged and appeared to be activated, expressing numerous filiopodia and ruffles on their surfaces (Fig. 18). This is consistent with the in vivo microscopic observation of increased phagocytic activity during this time period. Open endothelial fenestrae lacking diaphragms progressively increased in number with age, as did their organization into sieve plates (Figs. 15b and 19). By 7 days, stellate cells appeared to be actively synthesizing material, as indicated by the large amount of rough endoplasmic reticulum in their cytoplasm. Consistent with this was a progressive increase in the amount of collagen fibrils and extracellular matrix in the space of Disse (Figs. 16 and 19). Microvilli on hepatocyte surfaces also increased in size and number with age, with a dramatic increase between the 14th and 30th days (Figs. 16 and 19). Narrowed ISS were observed with progressive age. That these were indeed sinusoids was demonstrated by the presence of fenestrae and stellate cells (Figs. 20 and 21).
Histology and Immunocytochemistry
Based on H&E sections, at day 4 multiple cell plates were predominant. With increasing age, the cell plate was reorganized toward being one cell thick. By day 30, the organization appeared very similar to the adult pattern, with only a few two-cell-thick plates found in the periphery of the lobules. Hematopoiesis was almost completely lost by day 11, although at day 14 a few rudimentary islands were occasionally encountered. The reduction of hematopoietic areas resulted in space for the hepatocytes to expand. The prevailing tendency appeared to be that the one-cell-thick modeling took place from the centrilobular zone toward the periportal zone, and hematopoiesis also disappeared in the same direction. This resulted in the formation of the adult pattern, in which single cell plates predominated. Although double cell plates were uncommon, they were observed occasionally (usually in the peripheral zone).
The percentage of all liver cells in S-phase rose slowly from 4 to 14 days postpartum (Fig. 22). During this time interval, roughly half of all cells in S-phase were located in the periportal zone, while only about 20% were in the centrilobular zone (Fig. 23).
Overall mitotic activity remained essentially low and constant (Fig. 22). As with the S-phase cells, most dividing cells were located in the periportal zone, and their percentage increased from 40% at 4 days to nearly 60% at 14 days (Fig. 24). At the same time, the percentage of dividing hepatocytes in the centrilobular zone was almost 20% at 4 days and slowly decreased to nearly half that value at 14 days (Fig. 24).
The binucleated cells were more prevalent in the periportal zone at 4 days, but by 7 days they were present in equal percentages in both the periportal and centrilobular zones (Fig. 25). Afterwards, the binucleated cells continued to increase in abundance in the centrilobular zones while they decreased in the periportal zones.
The results of this study in suckling and weaned rat pups demonstrate the functional morphogenesis of several features that contribute to hepatic microvascular heterogeneity. These include differences between the periportal region of the hepatic lobule and centrilobular region in numbers of phagocytic KCs, sinusoid network pattern, sinusoid diameter and endothelial porosity, and their temporal relationship to the morphogenesis of parenchymal cell plates (including DNA synthesis, mitotic index, and ploidy).
The finding that all cell plates were more than one cell thick and irregularly arranged in early neonatal liver (4 days), but with time became one cell thick and straight only in the centrilobular zone, agrees with previous results (Elias, 1949; LeBouton, 1974, 1976; Alexander et al., 1997). Consistent with this morphogenetic change in cell plates is our observation that hepatocyte DNA synthesis and subsequent mitosis remain relatively high only in the periportal zone, which has thick, irregular cell plates. Thus, in the postnatal expanding population of liver cells, most growth occurs in the periportal zone, as previously reported (LeBouton, 1976).
Complementing the observed morphogenetic changes in the cell plates were the findings that 1) during the early suckling period, the sinusoid network is highly anastomotic, with little evidence of zonation, and 2) with increasing time after birth, the periportal sinusoids remain tortuous and interconnected by ISS, while those in the pericentral area become wider and straighter, with few ISS as a result of their “drop-out.” Thus, the developing hepatic microvasculature and parenchyma mimic each other. However, whether one or the other initiates and leads the other through this complicated postnatal process cannot be determined with the evidence at hand.
The centrilobular sinusoids may widen to accommodate the increased volume of blood resulting from the loss of ISS, together with the increase in the numbers of periportal sinusoids containing flow that feed these vessels. Since KC phagocytic activity also was observed to increase during the suckling period, it is tempting to speculate that these activated cells, which are principally located in the periportal zone, may be releasing mediators (such as prostaglandins and cytokines) (McCuskey et al., 1987; Wake et al., 1989; Decker, 1990) that may affect downstream sinusoid diameter, blood flow, and morphogenesis. This progressive activation of KCs may be the result of the increased absorption of gut-derived endotoxin demonstrated by day 11 in this study, which is probably due to bacterial colonization of the intestine. Endotoxin is a potent stimulant for the production of tumor necrosis factor alpha (TNFα) and prostaglandin E2 (PGE2) from KCs (McCuskey et al., 1987; Wake et al., 1989; Decker, 1990). In addition to its proinflammatory activities, TNFα also has been reported to prime hepatocytes to the effects of growth factors, including epidermal growth factor (EGF) (Webber et al., 1998). The vasoactive effects of PGE2 are well known. Other substances absorbed from the gut include milk-borne substances such as EGF, which is an hepatic growth factor (Opleta et al., 1987; Berseth and Go, 1988). We have shown that EGF in suckling rat pups enhances KC phagocytic activity while it increases the number of sinusoids containing blood flow and elicits sinusoidal dilatation (McCuskey et al., 1997). Whether or not these substances play a role in the development and heterogeneity of endothelial fenestration remains to be elucidated. It should be noted, however, that vascular endothelial growth factor (VEGF), a related growth factor, has been demonstrated to increase the numbers of fenestrae in cultured sinusoidal endothelial cells (Funyu et al., 2001).
In the fetal liver, the cellular microenvironment presumably is uniform along the sinusoid from the periportal to the pericentral zones due to the composition and volume of umbilical vein blood. However, at birth the umbilical blood flow ceases, which may alter the hepatic microenvironment. Yet, if any area within the liver after birth is reminiscent of fetal conditions, it is the periportal area with its mixture of arterial and portal venous blood. Indeed, it is precisely the periportal zone that retains immature characteristics.
The current observation that, with age, binuclear cells become more prevalent in the centrilobular zone agrees with earlier findings (LeBouton, 1976). It is interesting that the first postnatal overt expression of hepatocyte differentiation (binuclearity) also occurs in the centrilobular zone of the lobule. This also is the zone to first show morphogenetic changes leading to parenchymal and microvascular heterogeneity, i.e., thinning and straightening of cell plates together with the widening and straightening of the sinusoids and loss of ISS.
Polyploidization allows a cell to attain more gene copies, which can yield more transcripts and, ultimately, more protein (Anatskaya et al., 1994). The appearance of increased amounts of protein (especially enzymes) is an accepted indication of cellular differentiation, and when cells increase their differentiation they decrease their multiplication, which is what we observed here to be the situation between the centrilobular and periportal hepatocytes.
At birth, cells that are capable of DNA synthesis are equally distributed among the three acinar zones (LeBouton, 1974). This is considered to be the result of a homogeneous fetal microenvironment. However, in the current study we found that by postpartum day 4 this distribution changes, yielding approximately 50% of all labeled cells in the periportal zone and 20% in the centrilobular zone, in close agreement with the previous studies of LeBouton (1974, 1976). Ultimately, although the total number of hepatocytes that are capable of synthesizing DNA decreases with postnatal age (Terada et al., 1994; Friedman et al., 1995; Nagata, 1995), the percentage distribution of these cells among the various zones remains essentially the same as that noted above. In other words, at any given time after birth, most hepatocytes in S-phase are in the periportal zone, with the least amount in the centrilobular zone.
Dividing hepatocytes were also practically nonexistent in newborn rats immediately after birth, but by 4 days approximately 40% of all dividing cells were localized in the periportal zone and 20% were localized in the centrilobular zone, in close agreement with previous studies (LeBouton, 1974, 1976). With time, the total number of dividing hepatocytes decreased, but their distribution throughout the lobule remained as described. This distribution of mitotic figures throughout the lobule is essentially the same as the previously described distribution for cells that make DNA. Apparently, DNA synthesis and mitosis in the early suckling liver are closely linked in that nearly all hepatocytes that enter S-phase do so in order to prepare for an upcoming division, and these two events are most likely influenced by substances from the intestines.
The percentage of binucleated hepatocytes remains below 2% from birth to 14–16 days postpartum, when it begins to increase rapidly and reaches a value of nearly 30% by 30 days postpartum (Wiest, 1972; LeBouton, 1974; James, 1977). It must be remembered that the percent zonal distribution of binucleated cells is opposite that for mitotic cells, as demonstrated in this and previous studies (LeBouton, 1974). Thus, the appearance of binucleated cells is probably necessary for polyploidization of liver cells, which begins after the third week (Wiest, 1972; LeBouton, 1974; James, 1977).
In newborn rats (a few hours old), cell plates throughout the entire lobule are more than one cell thick and irregular. By 10 days, however, they have attained essentially the adult arrangement, which is thick and irregular in the periportal zone and thin (one cell thick) and straight in the centrilobular zone, in agreement with LeBouton (1974). The thick, irregular plates in the immediate newborn are considered to be the result of a homogeneous microenvironment in the fetus that allows hepatocytes in all zones an equal opportunity to enter mitosis, thus producing thick, irregular plates. However, since in the neonate it is only the periportal zone that has cells capable of much division, it is only this zone that retains the fetal condition of thick irregular plates, whereas (presumably due to hemodynamic factors) the plates in other zones are straightened.
Thus, the first 2 weeks of postnatal life are a time of profound alteration and change, involving both hepatic parenchymal and nonparenchymal cells in a quid pro quo relationship. Liver cells modulate their DNA synthesis and subsequent mitosis, and overtly differentiate into binuclearity accompanied by modulation in cell plates, while at the same time the sinusoids change their course, size, and porosity. And these side-by-side, give-and-take interactions between parenchymal and nonparenchymal elements must occur correctly so as to permit normal function during the entire period.
The authors acknowledge the major contribution of Professor Otakar Koldovsky, M.D., Ph.D., in stimulating us to do this and related projects examining the influence of breast milk-borne substances on the development of the liver and intestine. Unfortunately, Professor Koldovsky died before this project was completed. His mentoring and friendship will always be remembered and appreciated.
- 1997. Morphological changes during hepatocellular maturity in neonatal rats. Anat Rec 248: 104–109. , , .
- 1994. Hepatocyte polyploidy and metabolism/life-history traits: hypotheses testing. J Theor Biol 168: 191–199. , , .
- 1982. Development of ultrastructural heterogeneity among hepatocytes in the mouse. Anat Rec 202: 395–405. , , .
- 1988. Enhancement of neonatal somatic and hepatic growth by orally administered epidermal growth factor in rats. Pediatr Gastroenterol Nutr 6: 889–893. , .
- 1990. Biologically active products of stimulated liver macrophages. Eur J Biochem 192: 245–261. .
- 1993a. HMS, hepatic microcirculatory subunit: three dimensional observations on development and spatial distribution in mammalian livers. Hepatology 18: 153A. , , , , .
- 1993b. Postnatal development of the hepatic microvasculature and microcirculation in rats. Hepatology 18: 157A. , , , , .
- 1993c. Mammalian liver units as revealed by temporal and spatial reconstructions: recognition of the hepatic microcirculatory subunits. Anat Rec Suppl 1: 48. , , , .
- 1994a. Hepatic microcirculatory subunits (HMS) in mammalian liver lobules. FASEB J 8: A547. , , , , .
- 1994b. Hepatic microcirculatory subunits (HMS): a new look at functional units in mammalian liver. FASEB J 8: A1038. , , , , , .
- 1995. HMS, hepatic microcirculatory subunits in mammalian species: intralobular “grouping” of liver tissue with definition enhanced by the “drop-out” sinusoids. In: WisseE, KnookDL, WakeD, editors. Cells of the hepatic sinusoid. Vol. V. Leiden: Kupffer Cell Foundation. p 21–24. , , , , , , , , .
- 1949. Re-examination of the structure of the mammalian liver. Am J Anat 85: 311–334. .
- 1995. Growth patterns in selected organs of the miniature swine as determined by gross macromolecular composition. J Anim Sci 73: 1340–1350. , , , , , ,
- 2001. VEGF can act as vascular permeability factor in the hepatic sinusoids through upregulation of porosity of endothelial cells. Biochem Biophys Res Commun 280: 481–485. , , , , .
- 1977. The genesis of polyploidy in rat liver parenchymal cells. Cytobiology 15: 410. .
- 1992. Immunolocalization of proliferating perisinusoidal cells in rat liver. Histochem J 24: 67–72. , , .
- 1986. Dynamics of zonal hepatocyte heterogeneity. Perinatal development and adaptive alterations during regeneration after partial hepatectomy, starvation and diabetes. Acta Histochem Suppl 32: 89–98. .
- 1982. Functional hepatocellular heterogeneity. Hepatology 2: 385–395. , .
- 1974. Growth, mitosis and morphogenesis of the simple liver acinus in neonatal rats. Dev Biol 41: 22–30. .
- 1976. DNA synthesis and cell proliferation in the simple liver acinus of 10 to 20-day old rats: evidence for cell fusion. Anat Rec 184: 679–688. .
- 1967a. Dynamic microscopic anatomy of the fetal liver. I. Microcirculation. Angiology 18: 648–653. .
- 1967b. Dynamic microscopic anatomy of the fetal liver. II. Effect of pharmacodynamic substances on the microcirculation. Bibl Anat 9: 71–75. .
- 1968. Dynamic microscopic anatomy of the fetal liver. III. Erythropoiesis. Anat Rec 161: 267–280. .
- 1986. Microscopic methods for studying the microvasculature of internal organs. In: BakerCH, NastukWF, editors. Physical techniques in biology and medicine microvascular technology. New York: Academic Press. p 247–264. .
- 1987. Kupffer cell function in host defense. Rev Infect Dis 5: S616–S619. , , , .
- 1994. Microvascular heterogeneity of the hepatic lobule. FASEB J 8: A547. , , , .
- 1997. Effect of milk-borne EGF on the hepatic microcirculation and Kupffer cell function in suckling rats. Biol Neonate 71: 202–206. , , , , .
- 1995. Light and electron microscopic radioautographic studies on macromolecular synthesis in digestive organs of aging mice. Cell Mol Biol 41: 21–38. .
- 1987. Effect of epidermal growth factor on growth and postnatal development of the rabbit liver. Am J Physiol 253: G622–626. , , , , , .
- 1994. Expression of transforming growth factor-alpha and its receptor during human liver development and maturation. Virchows Arch 424: 669–675. , , .
- 1986. Regulation of hepatic metabolism. New York: Plenum Press. 489 p. , , .
- 1989. Cell biology and kinetics of Kupffer cells in the liver. Int Rev Cytol 118: 173–229. , , , , , .
- 1998. Tumor necrosis factor primes hepatocytes for DNA replication in the rat. Hepatology 28: 1226–1234. , , , .
- 1972. The effect of diethylnitrosamine on the distribution of cell classes in the parenchyma of the liver of newborn rats. Eur J Cancer 8: 121–125. .
- 1983. Scanning electron microscopic observations on the structure of portal veins, sinusoids and central veins. Scan Electron Microsc 3: 1441–1452. , , , .
- 1985. The liver sieve: consideration concerning the structure and function of endothelial fenestrae, the sinusoid wall and the space of Disse. Hepatology 5: 683–692. , , , , , .