Res Exp Med (1998) 198: 23–35 Springer-Verlag 1998 1996 1997 © Springer-Verlag

Quantitative analysis of small intestinal microcirculation in the mouse Steffen Massberg, Simone Eisenmenger, Georg Enders, Fritz Krombach, Konrad Messmer Ludwig-Maximilians University, Institute for Surgical Research, Klinikum Grosshadern, Munich, Germany Received: 10 January 1998 / Accepted: 14 April 1998

Abstract. Impairment of intestinal nutritive perfusion and accumulation of inflammatory cells in the intestinal microvasculature are well-known sequelae of mesenteric ischemia/reperfusion, sepsis, and shock. However, the molecular mechanisms underlying these alterations are still not fully understood. The mouse is particularly suitable for the study of these mechanisms since in this species the involvement of, for example, adhesion receptors or pro-/anti-adhesive mediators can be selectively investigated by the use of monoclonal antibodies or gene-targeted strains. The aim of our present study was, therefore, to establish a model to investigate the microcirculation in the mouse small intestine. Under anesthesia by inhalation of isoflurane-N2O, Balb/c mice (n=16) were laparotomized, and a segment of the jejunum was exteriorized for intravital fluorescence microscopy. Using FITC-dextran (MW 150,000) as a plasma marker, functional capillary density (FCD) of both the intestinal mucosa and muscle layer was analyzed. Nutritive perfusion was homogeneous in both compartments with values for FCD of 512±15 cm–1 in mucosa and 226±21 cm–1 in the muscle layer. No significant changes were observed throughout the observation period of 2 h (FCD values at the end of the observation period: 524±31 cm–1 and 207±7 cm–1 in mucosa and muscle, respectively). Besides capillary perfusion, leukocyte-endothelial cell interaction was analyzed in postcapillary venules of the intestinal submucosa using rhodamine-6G as an in vivo leukocyte stain. Under physiological conditions only a few white blood cells were found rolling along or firmly adherent to the microvascular endothelium (number of rolling leukocytes 1±0.2 cells/mm per second; number of adherent leukocytes: 18±7 cells/mm2). In a separate group rhodamine-6G-labeled syngeneic platelets were infused to analyze platelet-endothelial cell interactions quantitatively in vivo. Platelets rolled along or attached to the endothelium in a manner similar to leukocytes. However, in contrast to leukocytes the interactions were not restricted to venules, but were also observed Supported by research grant Biomed 2 Contract No. BMH4-CT95-0875 (DG12-SSMA) Correspondence to: S. Massberg, e-mail: [email protected], Tel.: +4989-7095-4357, Fax: +49-89-7095-8897

24

in small arterioles. The newly established model allows for the visualization and quantitative assessment of both nutritive perfusion and platelet/leukocyteendothelial cell interactions within the distinct layers of the mouse small intestine. Using this model in combination with gene-targeted mice or monoclonal antibodies it is possible to investigate the molecular mechanisms of intestinal inflammation reactions. Key words: Leukocytes – Platelets – Endothelial cells – Fluorescence microscopy – Angioarchitecture Introduction The microcirculation of the intestine represents a primary site of injury in sepsis, shock and normo- as well as hypothermic ischemia [9, 10, 16]. Microvascular injury is characterized by an impairment of nutritive capillary perfusion as well as cell-cell interactions (leukocytes, platelets, endothelial cells) within the microcirculatory networks supplying the different compartments of the bowel wall [16, 19]. However, the exact humoral and molecular mechanisms underlying microvascular alterations of the small intestine remain unclear. Therefore, an appropriate model is needed to investigate intestinal microcirculation under both physiological and pathological conditions. The mesentery has been frequently used as a model for the intestinal microvasculature [2]. However, mesenteric and intestinal microvascular networks have very little in common. Consequently, we have established a model in the rat which allows the simultaneous investigation of the microcirculation within all layers of the small intestine, i.e. subserosa, smooth muscle, submucosa and mucosa [10, 19]. However, the potential of the rat model is limited when information concerning distinct molecular mechanisms of microcirculatory phenomena is required. In contrast, the mouse is immunologically well characterized. A variety of monoclonal antibodies (mAb) directed against mouse antigens has been developed and only recently have gene-targeted mouse strains become available (e.g. disruption of the genes encoding inducible nitric oxide synthase [iNOS], interleukin [IL]-1, IL-2, P-, E-selectin, or ICAM-1) [11, 13]. Both functionblocking mAb and the use of gene-targeted animal strains are important tools in biological research, to obtain insight into distinct cellular, humoral, and/or molecular mechanisms encountered under physiological or pathological conditions. Therefore, it was our aim to establish an in vivo model for microvascular studies in the small intestine of the mouse allowing for the visualization and quantitative assessment of microvascular perfusion and cellular phenomena, including leukocyte- and platelet-endothelial cell interactions within the different tissue compartments intestinal muscle, submucosa, and mucosa. Materials and methods Animals Female Balb/c mice (Charles River, Sulzfeld, Germany), aged 5–7 weeks, were housed in an environmentally controlled room with a 12-h light-dark cycle (n=16). The animals were

25 fed with standard laboratory pellet food (18000 I. E./kg Vit. A, 1280 I. E./kg Vit. D3, 120 mg/kg Vit. E; ssniff Spezialdiäten, Soest, Germany) and water ad libitum. All experimental procedures performed were approved by the German legislation on protection of animals.

Anesthesia and surgical procedure Spontaneously breathing mice were anesthetized by inhalation of isoflurane-N2O (FiO2 0.35, 5 Vol % isoflurane for induction followed by 1.5 Vol % isoflurane for maintenance of anesthesia; Forene, Abbott, Wiesbaden, Germany). The animals were placed in supine position on a heating pad (Effenberger, Munich, Germany) for maintenance of rectal body temperature between 36.5°C and 37°C (Fig. 1A). Polyethylene catheters (PE 50, ID 0.28 mm, Portex, Hythe, UK) were inserted into the left carotid artery and jugular vein under a Leitz dissecting microscope (Leitz, Wetzlar, Germany). The arterial catheter was connected to a pressure transducer (Viggo-Spectramed, DTX/Plus, Oxnard, USA) for continuous recording of mean arterial blood pressure. Throughout the experiment, the animals received an intravenous infusion of Ringer′s lactate at a rate of 50 ml/kg per hour to maintain isovolemia. Following catheterization, the abdomen was opened transversely. A segment of the jejunum (approx. 3 cm from the ileo-cecal valve) was gently exteriorized and placed on a specially designed adjustable stage. A longitudinal incision (approximately 10 mm) was performed antimesenterically with a microcautery to open the distal, aboral part of the segment. The entire exposed segment was then covered with a glass coverslide and constantly superfused with 37°C Ringer’s lactate solution to avoid temperature changes and drying by exposure to ambient air.

Intravital fluorescence microscopy Technical setup: The animals were allowed to stabilize for 30 min after surgery. Subsequently, the intestinal microcirculation within the exposed segment was analyzed by epi-illumination technique, using a modified Leitz-Orthoplan microscope with a 100 W mercury lamp (HBO, Osram, Munich, Germany), attached to a Ploemo-Pak illuminator with I2/3 (excitation: 450–490 nm, emission >515 nm) and N2 (excitation: 530–560 nm, emission: >580) filter blocks (Leitz, Wetzlar, Germany). With ×10 (long distance, ×10/0.3, Leitz, Wetzlar, Germany) and ×25 objectives (water immersion, W ×25/0.6, Leitz, Wetzlar, Germany), the magnifications on the video screen (PVM-1442 QM, diagonal 33 cm, Sony, Munich, Germany) were ×180 and ×450, respectively. To facilitate evaluation of dynamic video images, the signal of a time generator (VTG-33, For-A-Company, Tokyo, Japan) was superimposed on the video screen (Fig. 1B). The microvasculature of the smooth muscle and of the submucosa was assessed from the serosal side in the oral closed part of the segment by changing the depth of focus of the microscope. Villous microcirculation was visualized from the mucosal surface through the longitudinal incision in the aboral part of the segment. The microscopic images were recorded by a charge-coupled device video camera (FK 6990, Cohu, Prospective Measurements, San Diego, USA) and transferred to a video cassette recorder (U-matic Video cassette recorder VO-5800 PS, Sony, Tokyo, Japan) for off-line evaluation. Fluorescent dyes: The intestinal microcirculation was visualized after intravenous administration of 0.05 ml 5% fluorescein isothiocyanate (FITC)-labeled dextran per animal (MW 150,000; Sigma-Aldrich Chemie GmbH, Deisenhofen, Germany). Leukocytes were stained in vivo by intravenous injection of 0.1 ml 0.05% rhodamine-6G (MW 479; Sigma-Aldrich Chemie, Deisenhofen, Germany) per animal. In a separate group (5 donors and 5 recipients) 50×106 platelets labeled with rhodamine-6G ex vivo (for description of fluorescent labeling see below) were infused via the jugular vein with a syringe pump (WPI, S250i Pump, Sarasota, USA) for 10 min starting 5 min prior to intravital microscopy. After i.v. administration of 0.05 ml 0.05% acridine orange (MW 302, Merck, Darmstadt, Germany), both

26

A

B

Fig. 1. A Animal preparation and microscopic setup. Spontaneously breathing mice are anesthetized by inhalation of isoflurane-N2O and placed in supine position on a heating pad (for details see Materials and methods). Polyethylene catheters are inserted into the left carotid artery and jugular vein for continuous recording of mean arterial blood pressure, for volume replacement, and for injection of fluorescent dyes. Through a transverse laparotomy a segment of the jejunum is exteriorized and placed on an adjustable stage. B The intestinal microcirculation within the exposed segment is analyzed by epi-illumination technique, using a Leitz-Orthoplan microscope. The microscopic images are recorded by a charge-coupled device video camera and transferred to a video system with a video tape recorder (VTR) and a monitor. The signal of a time generator is superimposed on the video screen. (CCD charge-coupled device, RL Ringer’s lactate, VTR video tape recorder)

27 platelet- and leukocyte-endothelial cell interactions were visualized in the submucosa of the same animal using the two different excitation and emission filter sets described above.

Quantification of microcirculatory parameters Quantitative assessment of microcirculatory parameters was performed off-line by frameto-frame analysis of the videotaped images using a computer-assisted image analysis system (CAP IMAGE, Dr. Zeintl, Heidelberg, Germany) [14]. Functional capillary density (FCD) of muscle and mucosa. In the intestinal smooth muscle layer, FCD, i.e. the length of red blood cell-perfused, nutritive capillaries per observation area (cm–1), was measured. The exposed segment was scanned at low magnification (×180) from the oral to the aboral section. Using high magnification (×450) FCD in the muscle layer was determined in five non-overlapping regions of interest. Quantitative analysis of FCD was performed off-line by means of CAP IMAGE [14]. Similarly, functional capillary density, length of red blood cell-perfused capillaries per villus area (cm–1), was determined in five non-overlapping fields of the exposed mucosa. The index of capillary heterogeneity (HI, distribution of local capillary density) was calculated from these 5 areas by HI = (FCDmax - FCDmin) / FCDmean, where FCDmax, FCDmin, and FCDmean represent the highest, the lowest, and the mean FCD, respectively. Leukocyte-/platelet-endothelial cell interactions. For the investigation of leukocyte-/platelet-endothelial cell interactions, arterioles and venules were visualized in the intestinal submucosa by adjusting the focus level of the microscope. Platelet-endothelial cell and leukocyte-endothelial cell interactions were analyzed each within 5–7 second order arterioles (mean diameter 40 µm) and postcapillary venules (mean diameter 60 µm) per animal, respectively. Platelets and leukocytes were classified according to their interaction with the endothelial cell lining as free-flowing, rolling, and adherent cells. Rolling platelets or leukocytes were defined as cells crossing an imaginary perpendicular line through the vessel at a velocity significantly lower than the centerline velocity in the microvessel; their numbers are given as cells per second per vessel diameter. Adherent cells were defined in each vessel segment as cells that did not move or detach from the endothelial lining within an observation period of 30 s. Adherence is quantified as number of cells per mm2 of endothelial surface, calculated from the diameter and length of the vessel segment observed, assuming cylindrical geometry.

Platelet preparation and flow cytometry To visualize platelet-endothelial cell interactions, platelets were isolated from whole blood and labeled with rhodamine-6G ex vivo. For each experiment 1 ml blood was harvested from a separate Balb/c mouse by cardiac puncture and collected in 1-ml polypropylene tubes containing 0.1 ml volume of 38 mM citric acid/75 mM trisodium citrate/100 mM dextrose. 50 µl 0.05% rhodamine-6G per ml whole blood (MW 479; Sigma-Aldrich) was added to label platelets in vitro, and the blood was centrifuged at 250 × g for 10 min. Platelet-rich plasma was gently transferred to a fresh tube, containing 1 ml Dulbecco‘s phosphate-buffered saline (D-PBS, PAN Systems, Aidenbach, Germany), and the centrifugation was repeated at 2000 × g for 10 min in the presence of 2 µmol/l PGE1 (Serva, Heidelberg, Germany). The resulting platelet pellet was resuspended in 0.5 ml D-PBS (PAN Systems). The platelet count of the sample was measured using a Sysmex F-800 Microcell counter (Sysmex, Tokyo, Japan). A total of 50×106 platelets labeled with rhodamine-6G were transfused. With a physiological platelet count in mice of approximately 600×103/µl whole blood and a total blood volume of 5.85 ml/100 g [21], the labeled fraction was 5–10% of all circulating platelets.

28

Fig. 2. Flow cytometry of separated platelets. A Representative dot plot of separated platelets. Platelets are identified by their characteristic forward and sideward scatter light. Few cells besides platelets are detected in the sample. B After incubation with 0.05% rhodamine-6G (50 µl/ml whole blood), 99% of the cells are stained. C Whereas P-selectin expression is low after separation, thrombin (5 U/ml) markedly enhances surface P-selectin expression on separated platelets. Mean±SEM, n=4 per group, *P<0.05 vs. Control, Mann-Whitney rank sum test

Flow cytometric (FACS) analyses were performed to confirm the purity of each platelet suspension prior to infusion. In addition, P-selectin expression by platelets was determined to assess activation due to the sampling procedure. Expression of P-selectin (CD62P) on platelets was investigated by direct immunofluorescence using rat anti-mouse fluorescein-isothiocyanate (FITC)-coupled monoclonal IgG (Pharmingen, Hamburg, Germany). Fluorochrome-coupled IgG1 isotype-matched control antibodies were used to exclude unspecific binding. The cells were incubated with saturating amounts of mAb for 30 min at 4°C. After incubation, the cells were washed twice with D-PBS (PAN Systems). Analysis of 10,000 events was performed on a FACSort flow cytometer (Becton Dickinson, Heidelberg, Germany). Platelets were selectively analyzed for their fluorescence properties using a Lysis II data handling program (Becton Dickinson). The relative fluorescence intensity of a given sample was calculated by subtracting the signal obtained when cells were incu-

29 bated with the corresponding isotype specific control from the signal generated by cells incubated with the test antibody. Platelet separation by differential centrifugation yielded a platelet suspension with negligible amounts of other cellular components. A representative dot plot of separated platelets is demonstrated in Fig. 2 A. More than 99% of all platelets are labeled with rhodamine6G (Fig. 2B). Platelet preparation did not activate the cells since expression of P-selectin (mean fluorescence intensity) was not up-regulated after differential centrifugation when compared to blood sampled without additional separation (P<0.788). In contrast, after incubation of separated platelets with thrombin (5 U/ml), a tenfold increase of P-selectin expression was observed (P<0.05, Fig. 2C).

Experimental protocol After surgery and exposure of the jejunal segment, the animal was allowed to stabilize for 30 min. Intravital microscopy was performed initially as well as at 60 min and 120 min after the stabilization period. Platelet-endothelial cell interaction was studied in a separate group 60 min after stabilization. At the end of the experiment, the animals were sacrificed by exsanguination.

Statistics Data analysis was performed with a statistical software package (SigmaStat for Windows, Jandel Scientific, Germany). The Kruskal-Wallis One Way Analysis of Variance on Ranks and the Student-Newman-Keuls test were used for the estimation of stochastic probability in intergroup comparisons. The Friedman test was applied in the case of repetitive measurements. Mean values ± standard error of the mean (SEM) are given. P values less than 0.05 were considered significant.

Results Macrohemodynamic parameters Mean arterial blood pressure (MAP) remained constant throughout the entire experiment (Table 1). MAP was 91±3 mmHg after laparotomy and 82±3 mmHg after the last intravital fluorescence microscopy, respectively. Rectal body temperature remained constant between 36.5°C and 37°C. Nutritive perfusion of mucosa and muscle The microcirculation of the mouse villus was found to be similar to what has been previously reported for the rat [10, 19]. A single arteriole in the villous core supplies a dense meshwork of capillaries, drained by several postcapillary venules (Fig. 3). Mucosal perfusion was homogeneous throughout the experiment (Table 1) with a low HI of 0.3±0.1 and a FCD of 512±15 cm–1 (baseline recordings). There were no significant changes during the entire observation period (Table 1). The intestinal smooth muscle layers also demonstrated a homogeneous capillary perfusion. Both the longitudinal and circular muscle layers were found to be perfused by a single, more-or-less two-dimensional network of capillar-

30 Table 1. Mean arterial blood pressure, functional capillary density and index of capillary heterogeneity. Capillary perfusion within the small intestinal mucosa and muscle was visualized using intravital fluorescence microscopy. Mean±SEM, n=6 Baseline Mean arterial blood pressure [mmHg] Mucosa: Functional capillary density [cm–1] Heterogeneity index Muscle: Functional capillary density [cm–1] Heterogeneity index

60 min

120 min

91±3

89±4

82±3

512±15 0.30±0.1

515±12 0.21±0.02

524±31 0.22±0.02

226±21* 0.27±0.02

204±6* 0.21±0.03

207±7* 0.16±0.07

* P<0.05 vs. Mucosa, Student-Newman-Keuls test

Fig. 3. Photomicrograph of mucosal (upper panel) and muscle capillaries (lower panel), visualized by intravital fluorescence microscopy. Contrast enhancement was achieved by intravenous administration of the plasma marker FITC-dextran. Monitor magnification: ×450. Bars represent 50 µm

31 Table 2. Leukocyte-endothelial cell interactions in postcapillary venules of the intestinal submucosa. Leukocyte interactions with the microvascular endothelium were visualized using intravital fluorescence microscopy. Mean±SEM, n = 6 per group, Friedman Test

Number of rolling leukocytes [1/s/mm] Number of adherent leukocytes [mm–2]

Baseline

60 min

120 min

1±0.2 19±7

1±0.1 12±8

1±0.1 19±14

Fig. 4. Platelet-endothelial cell interactions under physiological conditions. Platelet-endothelial cell interactions were investigated in arterioles (white columns) and venules (gray columns) using intravital fluorescence microscopy. According to their interaction with the endothelial cell lining, platelets were classified as rolling or firmly adherent cells. Rolling platelets (left) are presented as number of cells per second and mm vessel diameter; adherent platelets (right) are given per mm2 vessel surface. Under physiological conditions both platelet rolling and firm adherence were rarely observed. Mean±SEM, n=5, *P<0.01 vs arteriole, Student-Newman-Keuls test

ies. The capillaries were characterized by their rectangular course (Fig. 3). FCD values in the intestinal muscle were shown to be significantly (approx. 50%) lower than in the mucosa. Neither HI nor FCD changed during the observation period of 120 min (Table 1). It is noteworthy that non-perfused capillaries were rarely observed in either mucosa or muscle layer. Furthermore, rhythmic intermittent perfusion or vasomotion were not detected. Leukocyte-/platelet-endothelial cell interactions in vivo Blood cell dynamics were investigated in the submucosal microvasculature. Both leukocytes and platelets labeled with rhodamine-6G were easily detected within the microvessels. Similarly, staining of leukocytes by i.v. administration of acridine orange resulted in an intensive fluorescence. In contrast to rhodamine-6G, the use of acridine orange provided the possibility to study independently both leukocyte-endothelial and platelet-endothelial cell interactions within the same animal. The values obtained for leukocyte rolling and firm adhesion did not depend on whether leukocytes were visualized by acridine orange (data not shown) or by rhodamine-6G. In our current study, leukocyte rolling and firm adhesion were rarely observed throughout the entire

32

experimental procedure. However, if leukocyte-endothelial cell interactions were present, they were restricted to postcapillary venules (Table 2) and did not occur in submucosal arterioles. Similar to leukocytes, platelets rarely interacted with the endothelial surface under physiological conditions. Whenever platelet-endothelial cell interactions were seen they revealed striking parallels to leukocyte-endothelial cell interactions: both intermittent (rolling) and stationary adhesion of platelets were observed. In contrast to leukocytes, platelet-endothelial cell interactions were not confined to postcapillary venules, but were prominent also within submucosal arterioles (Fig. 4). Formation of platelet aggregates was not detected. Discussion The microcirculation of the small intestine appears to be a primary target site for inflammation reactions observed in shock, sepsis, and mesenteric ischemia/reperfusion but also in chronic inflammatory diseases such as Morbus Crohn [1, 9, 10, 16, 22]. An impairment of intestinal nutritive perfusion as well as the accumulation of inflammatory cells within intestinal microvasculature represent key events in such pathologies [9, 10, 16]. However, the underlying cellular and molecular mechanisms that contribute to the microvascular injury of the intestine are less well understood. This is mainly due to the lack of an adequate model for the investigation of the microvasculature of the small intestine. The mesentery has been frequently used to assess microcirculatory alterations during intestinal pathologies [2]. However, since the mesenteric and the intestinal angioarchitecture are very different, observations made in the mesentery cannot be expected to correlate with the actual pathology of the gut itself. Other models have been established allowing for a quantitative analysis of the microcirculation within the different tissue compartments of the intestine [3–5, 10, 19]. However, because of the species used in these studies (in particular rat and hamster), it was no possible to make use of antibodies or genetically altered animals. Therefore, it was our aim to establish a model to study the intestinal microcirculation in the small intestine of the mouse quantitatively. Using intravital fluorescence microscopy both nutritive perfusion and cellular phenomena in distinct microvascular segments within different layers of the small intestine of the Balb/c mouse were visualized and quantitatively assessed. The mouse species was chosen because: (1) it is immunologically well characterized, (2) a variety of monoclonal antibodies has been developed directed against specific mouse antigens, and (3) several gene-disrupted (IL-1-, P-selectin-, ICAM-1, or E-selectin-deficient mice) and transgenic strains (e.g. mice overexpressing HSP72) have become available [6, 11–13, 17]. Both immunoinhibition and gene-targeting are important tools in biological research. They make it possible to isolate selectively and study the involvement of distinct humoral or molecular mechanisms in the pathogenesis of microcirculatory alterations of the intestine. Different microvascular parameters were measured from the mouse intestinal microvasculature. Using FITC-dextran as a plasma marker, we characterized the density and distribution of nutritive capillary perfusion within mu-

33

cosa and muscle in vivo. The density of red blood cell-perfused capillaries, i.e. the functional capillary density, has been demonstrated to be a reliable indicator of tissue perfusion [20]. We found that the angioarchitecture of the mouse villus is similar to what has been described previously for the rat [10, 19]. A single arteriole supplies a capillary bed, which is drained by several postcapillary venules. In the intestinal muscle layer, capillaries are arranged in a rectangular pattern. Both the longitudinal and the circular muscle layer share the same blood supply. This is in contrast to the rat, where both layers are perfused via distinct capillary networks [10, 19]. Both mucosa and muscle layer demonstrated a homogeneous, regular microvascular perfusion, as indicated by a heterogeneity index (HI) below 0.3. The mucosa, a tissue characterized by its active metabolism, revealed high FCD values of approximately 500 cm–1 throughout the observation period. In line with previous data from the rat, the intestinal muscle had a FCD significantly (approx. 50%) lower than that of the mucosa [10, 19]. Tissue-specific metabolic factors, such as a higher oxygen demand of the mucosa, are likely to be the reason for the difference in FCD of the mucosa and the muscle layer. Whereas no species differences were observed concerning the FCD within the intestinal muscle layer, mucosal FCD in the mouse was found to be lower than that in Sprague-Dawley rats (approx. 500 cm–1 in the mouse vs. 800–1000 cm–1 in the rat) [4, 18]. In addition to an altered capillary blood flow, the accumulation of inflammatory cells within the small intestinal microvasculature is a typical indication of intestinal pathologies such as ischemia/reperfusion injury. Therefore, leukocyte-endothelial cell interactions were quantitatively studied in submucosal arterioles and postcapillary venules. At baseline conditions only few leukocytes interacted with the endothelial cell lining. The values obtained for both the number of rolling and adherent leukocytes were similar to those reported previously for the rat [4, 10]. No significant changes in the number of rolling and firmly adherent leukocytes were observed throughout the observation period of 2 h. This indicates that no endothelial cell and/or leukocyte activation occurred, underlining the stability of the preparation. There is recent evidence that, besides leukocytes, the platelets might play an important role in the pathogenesis of acute inflammatory responses of splanchnic organs [7, 8, 15]. Since to date platelet-endothelial cell interaction has not been investigated in a systematic way in vivo, we have developed and tested a method for the isolation and labeling of platelets with rhodamine-6G. By determination of P-selectin surface expression we could assure that no significant platelet activation occured due to the separation and labeling procedure. Using tagged platelets it was possible to visualize and assess quantitatively platelet-endothelial cell interactions in vivo. We found that platelets, like leukocytes, are able to roll along or firmly attach themselves to the endothelial cells within the intestinal microcirculation. As was seen in the case of leukocytes, under physiological conditions platelets also rarely interacted with the endothelial surface. Whereas leukocyte-endothelial cell interaction was found to be restricted to postcapillary venules, platelets also interacted with the endothelium of small arterioles. In conclusion, our newly established model allows a thorough investigation of the dynamics of the intestinal microcirculation of the mouse under both

34

physiological and pathophysiological conditions. By intravital fluorescence microscopy, both nutritive perfusion of mucosa and muscle layer and interactions of leukocytes and platelets with the endothelial surface can be visualized and quantitatively analyzed. This model can be used to characterize the molecular and humoral determinants involved in the pathogenesis of microvascular injury in response to acute or chronic inflammatory disorders of the small intestine. Acknowledgments. The authors wish to thank Sylvia Münzing for her excellent and skillful technical assistance.

References 1. Anthony A, Dhillon AP, Thrasivoulou C, Pounder RE, Wakefield AJ (1995) Pre-ulcerative villous contraction and microvascular occlusion induced by indomethacin in the rat jejunum: a detailed morphological study. Aliment Pharmacol Ther 9:605–613 2. Arndt H, Bolanowski MA, Granger DN (1996) Role of interleukin 8 on leucocyte endothelial cell adhesion in intestinal inflammation. Gut 38:911–915 3. Bohlen HG, Gore RW (1977) Comparison of microvascular pressures and diameters in the innervated and denervated rat intestine. Microvasc Res 14:251–264 4. Boros M, Massberg S, Baranyi L, Okada H, Messmer K (1998) Endothelin-1 induces leukocyte adhesion in submucosal venules of the rat small intestine. Gastroenterology 114:103–114 5. Boyd AJ, Sherman IA, Saibil FG (1994) Intestinal microcirculation and leukocyte behavior in ischemia-reperfusion injury. Microvasc Res 47:355–368 6. Bullard DC, Qin L, Lorenzo I, Quinlin WM, Doyle NA, Bosse R, et al (1995) P-selectin/ICAM-1 double mutant mice: acute emigration of neutrophils into the peritoneum is completely absent but is normal into pulmonary alveoli [see comments]. J Clin Invest 95:1782–1788 7. Collins CE, Rampton DS (1995) Platelet dysfunction: a new dimension in inflammatory bowel disease. Gut 36:5–8 8. Collins CE, Rampton DS (1997) Review article: platelets in inflammatory bowel disease-pathogenetic role and therapeutic implications. Aliment Pharmacol Ther 11: 237–247 9. Fukumura D, Miura S, Kurose I, Higuchi H, Suzuki H, Ebinuma H, et al (1996) IL-1 is an important mediator for microcirculatory changes in endotoxin-induced intestinal mucosal damage. Dig Dis Sci 41:2482–2492 10. Gonzalez AP, Sepulveda S, Massberg S, Baumeister R, Menger MD (1994) In vivo fluorescence microscopy for the assessment of microvascular reperfusion injury in small bowel transplants in rats. Transplantation 58:403–408 11. Grassl G, Pummerer CL, Horak I, Neu N (1997) Induction of autoimmune myocarditis in interleukin-2-deficient mice. Circulation 95:1773–1776 12. Hutter JJ, Mestril R, Tam EK, Sievers RE, Dillmann WH, Wolfe CL (1996) Overexpression of heat shock protein 72 in transgenic mice decreases infarct size in vivo. Circulation 94:1408–1411 13. Kelly KJ, Williams WW Jr, Colvin RB, Meehan SM, Springer TA, Gutierrez Ramos JC, et al (1996) Intercellular adhesion molecule-1-deficient mice are protected against ischemic renal injury. J Clin Invest 97:1056–1063 14. Klyscz T, Jünger M, Jung F, Zeintl H (1997) CAP IMAGE: a newly developed computer aided videoframe analysis system for dynamic capillaroscopy. Biomed Tech 42: 168–175 15. Kuroda T, Shiohara E, Homma T, Furukawa Y, Chiba S (1994) Effects of leukocyte and platelet depletion on ischemia-reperfusion injury to dog pancreas. Gastroenterology 107:1125–1134

35 16. Kurose I, Granger DN (1994) Evidence implicating xanthine oxidase and neutrophils in reperfusion-induced microvascular dysfunction. Ann N Y Acad Sci 723:158–179 17. Ley K, Bullard DC, Arbones ML, Bosse R, Vestweber D, Tedder TF, et al (1995) Sequential contribution of L- and P-selectin to leukocyte rolling in vivo. J Exp Med 181: 669–675 18. Massberg S, Boros M, Leiderer R, Baranyi L, Okada H, Messmer K (1998) Endothelin-1-induced mucosal damage in the rat small intestine: role of ETA receptors. Shock 9:177–183 19. Massberg S, Gonzalez AP, Leiderer R, Menger MD, Messmer K (1998) In vivo assessment of the influence of cold preservation time on microvascular reperfusion injury after experimental small bowel transplantation. Br J Surg 85:127–133 20. Nolte D, Zeintl H, Steinbauer M, Pickelmann S, Messmer K (1995) Functional capillary density: an indicator of tissue perfusion? Int J Microcirc 15:244–249 21. Tangelder GJ, Slaaf DW, Arts T, Reneman RS (1988) Wall shear rate in arterioles in vivo: least estimates from platelet velocity profiles. Am J Physiol 254:H1059–H1064 22. Wakefield AJ, Sankey EA, Dhillon AP, Sawyerr AM, More L, Sim R, et al (1991) Granulomatous vasculitis in Crohn’s disease. Gastroenterology 100:1279–1287