Volume measurement variability in three-dimensional high-frequency ultrasound images of murine liver metastases.

The identification and quantification of tumour volume measurement variability is imperative for proper study design of longitudinal non-invasive imaging of pre-clinical mouse models of cancer. Measurement variability will dictate the minimum detectable volume change, which in turn influences the scheduling of imaging sessions and the interpretation of observed changes in tumour volume. In this paper, variability is quantified for tumour volume measurements from 3D high-frequency ultrasound images of murine liver metastases. Experimental B16F1 liver metastases were analysed in different size ranges including less than 1 mm3, 1-4 mm3, 4-8 mm3 and 8-70 mm3. The intra- and inter-observer repeatability was high over a large range of tumour volumes, but the coefficients of variation (COV) varied over the volume ranges. The minimum and maximum intra-observer COV were 4% and 14% for the 1-4 mm3 and <1 mm3 tumours, respectively. For tumour volumes measured by segmenting parallel planes, the maximum inter-slice distance that maintained acceptable measurement variability increased from 100 to 600 microm as tumour volume increased. Comparison of free breathing versus ventilated animals demonstrated that respiratory motion did not significantly change the measured volume. These results enable design of more efficient imaging studies by using the measured variability to estimate the time required to observe a significant change in tumour volume.

Solitary cancer cells as a possible source of tumour dormancy?

Metastasis can occur many years after primary tumour treatment. However, the status of the tumour during this period of dormancy is poorly understood. As part of our ongoing experimental studies on mechanisms of metastasis, we have discovered that large numbers of disseminated single cells may persist in secondary sites for extended time periods. Identification of these cells was facilitated by in vivo techniques developed to quantify the fate of individual cells during the metastatic process. Here we review these in vivo techniques and findings. We also discuss the potential clinical implications if dormant solitary cells exist in appreciable numbers in cancer patients.

Critical steps in hematogenous metastasis: an overview.

Metastasis is responsible for most cancer deaths. A better understanding of the process provides opportunities to develop new treatments to prevent metastasis. This article summarizes findings from experimental in vivo videomicroscopy and quantitative studies on metastatic inefficiency, which indicate that early steps in hematogenous metastasis may be quite efficient, but that regulation of cancer cell growth in secondary sites determines metastatic outcome. The authors have identified three key stages of this growth regulation: survival of a subset of single cells, proliferation of a subset of these cells to form preangiogenic micrometastases, and persistence of growth of a subset of these to form vascularized metastases. Formation of clinically relevant metastases is determined by the proportion of cells that proceeds successfully through each stage, and surviving single cells and preangiogenic micrometastases both represent possible sources of tumor dormancy.

Temporal progression of metastasis in lung: cell survival, dormancy, and location dependence of metastatic inefficiency.

Cancer metastasis is an inefficient process. The steps in metastasis responsible for this inefficiency and how metastatic inefficiency can vary in different locations within an organ remain poorly understood. B16F10 cells were injected to target mouse lung, and at sequential times thereafter we quantified in lung the time course of: (a) overall cell survival and metastatic development; and (b) local cell survival and growth with respect to the lung surface and specific interior structures. We found high rates of initial survival of cells trapped in the lung circulation, extravasation into lung tissue, and subsequent survival of extravasated solitary cells (74% at day 3) before metastasis formation. However, at the time of initial replication of metastatic cells a major loss of cells occurred. Although only a small proportion of injected cells started to form metastases, most of these developed into macroscopic tumors. Solitary cells found at later times were dormant. Thus, overall metastatic inefficiency was largely due to postextravasation events affecting solitary cells. Regionally within the lung, cells and metastases were randomly distributed to day 4, but by day 10 preferential tumor growth was found along the lung surface and around arterial and venous vessels. Thus, trapping and early growth of injected cells was unaffected by location within the lung, whereas subsequent metastatic growth was enhanced in specific microenvironments. This study: (a) quantifies early temporal and spatial progression of metastasis in lung; (b) documents persistence of solitary dormant cells; and (c) shows that metastatic inefficiency depends on the initiation of growth in a subset of extravasated cells, whereas continued growth of metastases occurs preferentially in specific tissue environments.

Clinical targets for anti-metastasis therapy.

Metastasis is responsible for most cancer deaths. Therapeutic strategies to prevent development of metastases thus have potential to impact on cancer mortality. Development of these therapies requires a better understanding of the biology and molecular events of the metastatic process. Metastasis is usually defined, clinically and experimentally, by evidence of the endpoint of the process, that is, the presence of metastatic tumors. Endpoint assays are suitable for determining if a therapeutic approach is effective, but can provide little information on how a treatment works in vivo and what steps in metastasis are affected. We describe here two methodological advances in the study of metastasis as a process: in vivo videomicroscopy, which permits direct observation of steps in metastasis, and a "cell accounting" technique that permits quantification of the fate of cells over time. These procedures have provided new and unexpected insights into the biology of the metastatic process. Based on these insights, we consider which steps in the metastatic process are biologically and clinically most appropriate as therapeutic targets for development of anti-metastasis therapies. We conclude that the most promising stage of the metastasis process for therapeutic targeting is the growth phase, after cancer cells have arrested in the microcirculation in secondary sites and have completed extravasation. Earlier phases in the process are either biologically inappropriate or clinically inaccessible, except in specific cases (e.g., chemoprevention strategies). The role of "seed" and "soil" in determining organ-specific metastasis is also discussed. The metastatic growth phase fortunately is a clinically broad target, and any treatment that limits growth of metastases prior to their causing irreversible harm to the patient has the potential to be clinically useful. A variety of therapeutic approaches to target this phase are under active development, including inhibition of angiogenesis or signal transduction pathways needed to support the growth of metastatic cells.

The matrix metalloproteinase inhibitor batimastat inhibits angiogenesis in liver metastases of B16F1 melanoma cells.

Matrix metalloproteinases (MMPs) have been shown to contribute functionally to tumor metastasis. MMP inhibitors are thus being assessed for clinical utility as anti-metastatic therapeutics. Batimastat (BB-94) is a synthetic MMP inhibitor that has been shown to inhibit tumor growth and metastasis in mice. Here we assessed the ability of batimastat to inhibit liver metastases of murine B16F1 cells, after injection of cells in mice via mesenteric vein to target the liver. We then determined which of the sequential steps in metastasis were affected by batimastat, in order to identify its mechanism of action in vivo. Intravital videomicroscopy was used to assess the effect on extravasation, and a 'cell accounting' procedure was used to determine the effect on initial survival of cells. Stereological quantification of functional blood vessels was used to determine the effect on tumor vascularity, thereby avoiding problems associated with immunohistochemical detection of liver sinusoidal endothelial cells. We found that batimastat (50 mg/kg i.p. 5 h prior to and after cell injection, daily thereafter) resulted in a 23% reduction in mean diameter of liver metastases (equivalent to a 54% reduction in tumor volume), while not reducing the number of metastases. Extravasation of cells from the liver circulation was not affected: at 8, 24 and 48 h after injection of cells, the same proportion of cells had extravasated from treated vs. control mice. Batimastat also did not inhibit early survival of cells. However, batimastat-treated mice had a significantly reduced percentage vascular volume within liver metastases, indicating inhibition of angiogenesis. This study demonstrates in vivo that the mechanism by which batimastat limits growth of B16F1 metastases in liver is not by affecting extravasation, but by inhibiting angiogenesis within metastases. This finding suggests that MMP inhibitors may be appropriate for use in patients with metastatic cells that have already extravasated in secondary sites.

Cellular expression of green fluorescent protein, coupled with high-resolution in vivo videomicroscopy, to monitor steps in tumor metastasis.

High resolution intravital videomicroscopy has provided a powerful tool for directly observing steps in the metastatic process, and for clarifying molecular mechanisms of metastasis and modes of action of anti-metastasis therapeutics. Cells previously have been identified in vivo using exogenously added fluorescent labels, limiting observations to a few cell divisions, or by natural markers (e.g. melanin) expressed only by specific cell types. Here we tested the utility of stable green fluorescent protein (GFP)-transfected cells for monitoring and quantifying sequential steps in the metastatic process. Using CHO-K1 cells that stably express GFP, we document the visualization and quantification by intravital videomicroscopy of sequential steps in metastasis within mouse liver, from initial arrest of cells in the microvasculature to the growth and angiogenesis of metastases. Individual, non-dividing cells, as well as micro- and macrometastases could clearly be detected and quantified, as could fine cellular details such as pseudopodial projections, even after extended periods of in vivo growth. We quantified the size distribution of micrometastases and their locations relative to the liver surface using 50 micrometer thick formalin-fixed tissue sections. The data suggest preferential growth and survival of micrometastases near the liver surface. Furthermore, we observed a small population of single cells that persisted over the 11 day observation period, which may represent dormant cells with potential for subsequent proliferation. This study demonstrates the advantages of GFP-expressing cells, coupled with real-time high resolution videomicroscopy, for long-term in vivo studies to visualize and quantify sequential steps of the metastatic process.

Tumour metastasis to the liver, and the roles of proteinases and adhesion molecules: new concepts from in vivo videomicroscopy.

Most preclinical studies of tumour metastasis and effects of molecular interventions have been based on end point assays, and little is known about the fate of cells at sequential steps in the metastatic process. In vivo videomicroscopy permits direct observations of sequential steps in hematogenous metastasis as they occur in living animals over time. These steps include initial arrest of cells in the microcirculation, extravasation, postextravasation migration and growth in the target organ. In the mouse liver model, cells are arrested in periportal sinusoids based on size restriction, survive in the circulation and extravasate into the tissue by 48 to 72 h regardless of metastatic potential. Thereafter, cells may migrate to preferred sites for growth. Critical steps responsible for cell losses and metastatic inefficiency occur at the level of postextravasation cell growth. Many extravasated cells may remain dormant, and growth to form micrometastases is initiated in only a small subset of cells. Most early micrometastases may disappear after a few days, and only a small subset continue growth into macroscopic tumours. Angiogenesis is a prerequisite for continued growth of metastases, as shown previously by others. Integrin based interventions can modulate postextravasation cell migration and cell growth. Matrix metalloproteinase inhibitors can inhibit tumour angiogenesis and thus reduce growth. Key targets against which future therapeutic strategies should be directed include the initiation and maintenance of growth of micrometastases, and the activation of dormant solitary cells.

Multistep nature of metastatic inefficiency: dormancy of solitary cells after successful extravasation and limited survival of early micrometastases.

In cancer metastasis, only a small percentage of cells released from a primary tumor successfully form distant lesions, but it is uncertain at which steps in the process cells are lost. Our goal was to determine what proportions of B16F1 melanoma cells injected intraportally to target mouse liver 1) survive and extravasate, 2) form micrometastases (4 to 16 cells) by day 3, 3) develop into macroscopic tumors by day 13, and 4) remain as solitary dormant cells. Using in vivo videomicroscopy, a novel cell accounting assay, and immunohistochemical markers for proliferation (Ki-67) and apoptosis (TUNEL), we found that 1) 80% of injected cells survived in the liver microcirculation and extravasated by day 3, 2) only a small subset of extravasated cells began to grow, with 1 in 40 forming micrometastases by day 3, 3) only a small subset of micrometastases continued to grow, with 1 in 100 progressing to form macroscopic tumors by day 13 (in fact, most micrometastases disappeared), and 4) 36% of injected cells remained by day 13 as solitary cancer cells, most of which were dormant (proliferation, 2%; apoptosis, 3%; in contrast to cells within macroscopic tumors: proliferation, 91%; apoptosis/necrosis, 6%). Thus, in this model, metastatic inefficiency is principally determined by two distinct aspects of cell growth after extravasation: failure of solitary cells to initiate growth and failure of early micrometastases to continue growth into macroscopic tumors.

Inhibition of angiogenesis in liver metastases by carboxyamidotriazole (CAI).

Carboxyamidotriazole (CAI), an inhibitor of calcium-mediated signal transduction, is a promising new cytostatic anti-cancer drug which has entered Phase II clinical trials, and for which multiple modes of action have been proposed. We tested the hypothesis that CAI can inhibit tumor angiogenesis in vivo. The ability of orally administered CAI to inhibit experimental metastases of B16F1 melanoma cells in mouse liver was assessed. A computer-assisted stereological technique was then used to analyze images from histological sections of CAI-treated vs. control livers; the vascular volume percentage (percentage of tumor volume consisting of functional microvessels) was determined to assess the effect of CAI on tumor angiogenesis. CAI treatment significantly reduced the size (8 x reduction in volume; P = 0.02) but not the number of metastases. In association with this reduction in tumor size, CAI significantly decreased the vascular volume percentage within metastases by at least a factor of two (P = 0.001). A reduction in both number of microvessels/mm2 and microvessel size (cross-sectional area) was found to contribute to this decrease. CAI treatment did not affect the vascular volume percentage of normal liver tissue surrounding metastases (P = 0.8). This study documents for the first time that CAI can inhibit tumor angiogenesis within metastases in vivo.

Preclinical assessment of anti-cancer therapeutic strategies using in vivo videomicroscopy.

Preclinical in vivo studies of agents targeted against metastasis have to date been based primarily on end-point assays. Such assays can determine whether a treatment affects the number or size of metastases in an organ at a given time, but are poorly suited to determining how and at what stage in the process the treatment affected the end point. High resolution in vivo videomicroscopy permits direct observation of the process of metastasis as it occurs in living animals over time. Studies based on this technique and a cell accounting procedure we have devised, have shown that early steps in the metastatic process (survival in the circulation, extravasation) contribute relatively little to cell loss and metastatic inefficiency. Steps that occur after extravasation appear to be primarily responsible for the significant losses that result in metastatic inefficiency, and these steps may represent good targets for the design of new antimetastatic therapies. Matrix metalloproteinases have been implicated functionally in metastasis, and are viewed as an appropriate target in the development of inhibitors of metastasis. Using both endogenous and synthetic exogenous metalloproteinase inhibitors, we have shown that the inhibition of metastasis which these agents produce is not due to inhibition of cell extravasation from the circulation into the tissue, but to reduction of angiogenesis within metastases. A similar conclusion was reached concerning the mechanism of action, on metastasis, of carboxyamidotriazole, an inhibitor of calcium-mediated signal transduction which is currently in Phase II single agent clinical trials. In vivo videomicroscopy of sequential steps in metastasis, coupled with methods that allow precise quantification of cell loss at specific steps in the metastatic process, as well as standard histological assessment at stages identified as crucial, allow characterization of the details of metastasis as an ongoing process. This provides a powerful complement to end-point assays, for it allows mechanistic information to be obtained from in vivo experiments, an approach which provides better understanding of how and when a drug may function in vivo to inhibit metastasis.

Sequential steps in hematogenous metastasis of cancer cells studied by in vivo videomicroscopy.

Understanding metastatic spread of cancer is of upmost importance to developing successful strategies to treat this disease. In this review, we present a picture of the process of hematogenous metastasis from the initial arrest of cancer cells, their extravasation, postextravasation migration, and their replication to form tumors, based on experimental results using in vivo videomicroscopy. The cancer cells are initially arrested by size constraints within minutes of entering the circulation and with little hemodynamic destruction. Within 24-48 h >80% of these cancer cells extravasate as single cells by adhesion to and spreading along the vessel wall, often using pseudopodial projections to move into the surrounding tissue without disrupting the microcirculation. Some of the extravasated cells also use pseudopodial projections to migrate to specific structures in the tissue where they can replicate. Many cancer cells can persist as dormant cells, neither dividing nor undergoing apoptosis. Only a small fraction of extravasated cells begin to divide to form micrometastases, and only a very small fraction of these micrometastases continue to grow to form tumors. Possible clinical implications are that (1) initial arrest and extravasation may be difficult to prevent and thus may be poor therapeutic targets; (2) dormant single cells will not be affected by conventional cancer therapies which are designed to treat actively growing cells; and (3) regulation of growth of cells after extravasation is key to determining whether clinically evident metastases form - this stage of metastasis thus offers promising targets for new antimetastasis drugs.

Independence of metastatic ability and extravasation: metastatic ras-transformed and control fibroblasts extravasate equally well.

Escape of cancer cells from the circulation (extravasation) is thought to be a major rate-limiting step in metastasis, with few cells being able to extravasate. Furthermore, highly metastatic cells are believed to extravasate more readily than poorly metastatic cells. We assessed in vivo the extravasation ability of highly metastatic ras-transformed NIH 3T3 cells (PAP2) versus control nontumorigenic nontransformed NIH 3T3 cells and primary mouse embryo fibroblasts. Fluorescently labeled cells were injected intravenously into chicken embryo chorioallantoic membrane and analyzed by intravital videomicroscopy. The chorioallantoic membrane is an appropriate model for studying extravasation, since, at the embryonic stage used, the microvasculature exhibits a continuous basement membrane and adult permeability properties. The kinetics of extravasation were assessed by determining whether individual cells (n = 1481) were intravascular, extravascular, or in the process of extravasation, at 3, 6, and 24 h after injection. Contrary to expectations, our results showed that all three cell types extravasated with the same kinetics. By 24 h after injection > 89% of observed cells had completed extravasation from the capillary plexus. After extravasation, individual fibroblasts of all cell types demonstrated preferential migration within the mesenchymal layer toward arterioles, not to venules or lymphatics. Thus in this model and for these cells, extravasation is independent of metastatic ability. This suggests that the ability to extravasate in vivo is not necessarily predictive of subsequent metastasis formation, and that postextravasation events may be key determinants in metastasis.

Integrin VLA-2 (alpha2beta1) function in postextravasation movement of human rhabdomyosarcoma RD cells in the liver.

It is now known that members of the selectin and integrin families are critical in the initial interaction of cells in circulation with endothelial surfaces. Also, platelet/endothelial cell adhesion molecule-1 has been shown to be involved in transendothelial migration of extravasating cells. Little is known about adhesion molecules involved in subsequent postextravasation events. In this study, the significance of VLA-2 (alpha2beta1) integrin in the movement of human rhabdomyosarcoma RD cells in the liver was characterized by in vivo videomicroscopy. Results show that after extravasation, the mock-transfected RDpF cells were able to migrate to the subcapsular region of the liver. Although the RDX2C2 transfectant expressing VLA-2 integrin extravasated equally well, a majority of RDX2C2 cells remained in close proximity to blood vessels and failed to reach the subcapsular region. The functional involvement of VLA-2 in affecting the ability of RD cells to reach the subcapsular region was verified by the preparation of an RD transfectant [RDX2C2(I-)] expressing a nonfunctional variant of VLA-2 lacking the inserted (I)-domain of alpha2 subunit. In vivo microscopy showed that RDX2C2(I-) cells migrated in a manner similar to control RDpF cells. To demonstrate that RDX2C2 cells that remained in dose proximity to blood vessels were due to VLA-2 function, a blocking monoclonal antibody against VLA-2 (BHA2.1) was prepared. Mice were injected with BHA2.1 or control monoclonal antibody P3 at the time when RDX2C2 cells completed their extravasation. Treatment with BHA2.1 increased the number of RDX2C2 cells that reached the subcapsular region and subsequently formed tumor foci. Therefore, VLA-2 integrin expression has major roles in postextravasation movement and affects tumor foci formation at the liver surface.

Steps in tumor metastasis: new concepts from intravital videomicroscopy.

Metastases are responsible for the majority of failures in cancer treatment. Clarifying steps in metastasis and their molecular mechanisms will be important for the development of anti-metastasis therapeutic strategies. Considerable progress has been made in identifying molecules involved in metastasis. However, because of the nature of assays that have been available, conclusions about steps in metastasis and their molecular bases have been drawn primarily from inference. In order to complete the picture of how metastases form, a technique is needed to directly watch the process in vivo as it occurs over time. We have developed an intravital videomicroscopy (IVVM) procedure to make such observations possible. Results from IVVM are providing us with new conceptual understanding of the metastatic process, as well as the nature and timing of the contributions of molecules implicated in metastasis (e.g. adhesion molecules and proteinases). Our findings suggest that early steps in metastasis, including hemodynamic destruction and extravasation, may contribute less to metastatic inefficiency than previously believed. Instead, our results suggest that the control of post-extravasation growth of individual cancer cells is a significant contributor to metastatic inefficiency. Thus, this stage may be an appropriate target for design of novel strategies to prevent metastases.

Intermittence of blood flow in liver sinusoids, studied by high-resolution in vivo microscopy.

Kupffer cell migration and leukocyte-vessel wall interactions cause temporary slowing and/or stoppage of blood flow through individual liver sinusoids. Such temporal heterogeneity of flow was quantified in anesthetized mice and rats. Video recordings of red blood cell flow in 44 networks containing 8-16 sinusoids each were analyzed for 5- to 10-min periods. Flow was graded "fast," "slow," "stopped," or "reversed" based on red blood cell velocity. The mean numbers of flow changes (between grades) per minute in zone 1 vs. zone 3 were 1.39 vs. 0.78 (mouse) and 1.25 vs. 0.09 (rat). The mean percentage of time for each flow grade differed significantly between zones 1 and 3 and between species. For example, fast flow was present in zone 1 sinusoids for 51% of the time in mice and for 74% in rats; in zone 3 the corresponding numbers were 76 and 95%. Flow stasis was present in zone 1 sinusoids for 19% of the time in mice and for 7% in rats; in zone 3 the corresponding numbers were 2 and 0%. Thus considerable intermittence of perfusion exists, and the flow conditions create very different microenvironments for hepatocytes in zone 1 vs. zone 3.

Capillary network morphology and capillary flow.

This paper examines the authors' research on capillary network morphology and the heterogeneity of capillary red cell (RBC) perfusion in skeletal muscle with the aim of demonstrating that capillary network structure plays a major role in determining flow distribution. Capillary network morphology was examined by quantifying the heterogeneity of capillary diameters, path and segment lengths, as well as the changes in configuration that occur as vessels accommodate themselves to continual changes of fiber length. Because of the network complexity and the two-phase nature of the perfusing blood, both spatial (i.e. among capillaries) and temporal heterogeneity of capillary perfusion were predicted to result. By means of computer analysis of video images of the microcirculation in vivo, we have demonstrated that more than 70% of the total spatial heterogeneity of capillary RBC perfusion arises from the capillary network as opposed to the arterioles, and that RBC flow continuously redistributes among capillaries. The spatial heterogeneity increases substantially as the arteriolar input to the network falls, and the data predict that during low-flow states, the network will fail to distribute blood properly among its constituent vessels. Thus passive rheological mechanisms and capillary network morphology are important determinants of functional capillary density.

Effects of the disintegrin eristostatin on individual steps of hematogenous metastasis.

Adhesion molecules, including integrins, are important for interactions of cancer cells with vessel walls, a step leading to cancer metastasis. Disintegrins block the action of integrins by binding to them. We tested the hypothesis that the disintegrin eristostatin would block metastasis by interfering with cancer cell adhesion to vessel walls, thus preventing extravasation. Experimental metastasis assays, in which B16F1 melanoma cells (controls vs eristostatin-treated, 25 micrograms/ml) were injected via mesenteric veins of anesthetized C57BL/6 mice, showed that eristostatin reduced (P < 0.05) the mean number of liver metastases from 14.4 to 0.6 at 11 days postinjection (p.i.). We examined three different steps in metastasis at which eristostatin could have exerted its effect, namely, cell arrest, extravasation, and migration. Control and eristostatin-treated B16F1 cells were fluorescently labeled and examined by videomicroscopy in liver microcirculation in vivo at various times up to 14 days p.i. Measurements of vessel size in which cell arrest occurred and length/width ratio of arrested cells showed only small differences between control and eristostatin-treated cells. Eristostatin treatment did not prevent extravasation, and the timing and process of extravasation were similar for both treated and control cells; by 3-4 days p.i. more than 90% of the cells had extravasated or were in the process. Eristostatin also did not affect the ability of extravasated cells to migrate through the extracellular matrix to the subcapsular region where tumors later form. Therefore, we conclude that eristostatin exerted its primary effect by regulating the number of individual cancer cells that grow after extravasation.

Fate of melanoma cells entering the microcirculation: over 80% survive and extravasate.

Metastasis is an inefficient process; only a few cancer cells are able to form tumors after being released into the circulation. We studied the fate of cancer cells after injection into the circulation, quantifying their survival and ability to extravasate by 1 day later. B16F10 cells, parental or transfectants overexpressing tissue inhibitor of metalloproteinases 1, were injected i.v. into chorioallantoic membrane of chick embryos and analyzed by intravital videomicroscopy. Cell survival was quantified in two ways: (a) 15-microns microspheres were injected with cancer cells, and proportions of viable cells to microspheres were compared before and after injection; and (b) individual cancer cells were monitored continuously for 0.5-8-h intervals covering the first 24 h. Both methods showed virtually no destruction of cells. Greater than 80% of injected cells survived and extravasated by 24 h, indicating that growth after extravasation is a key stage of metastatic control.

Luminal constrictions due to endothelial cells in capillaries of mouse exocrine pancreas.

During our recent studies of the capillaries in exocrine pancreas of mouse, numerous local constrictions which reduced the luminal diameter were observed both by scanning electron microscopy of corrosion casts and by in vivo microscopy. In the present study we have identified the features responsible for the constrictions and compared the diameters of vessels and constrictions measured using the two methods. A simple theoretical model was used to predict the effects of such constrictions on blood flow in the acinar capillaries of the pancreas. Intravital observations revealed that bulging endothelial cells were primarily responsible for the constrictions. For samples of 100 measurements, good agreement was found between the mean capillary diameters from casts (6.3 microns +/- 0.50 SD) and in vivo (6.2 microns +/- 0.53 SD), but the mean diameter measurement at constrictions was greater (P < 0.01) in casts (3.9 microns +/- 0.84 SD) than in vivo (3.5 microns +/- 1.05 SD). Topical application of norepinephrine caused endothelial nuclear regions to bulge into the capillary lumen, decreasing the mean diameter at these locations to 3.3 microns +/- 0.9 (SD, n = 21). Based on the 100 in vivo measurements, the theoretical model predicted that, on average, the constrictions would reduce flows to 51% of those in fully open vessels. It is unlikely, however, that the constrictions observed in acinar capillaries of the pancreas of mouse would result in significant blockage of the vessels by red blood cells.