Second valve system in lymphatics: endothelial microvalves

lymphatic mapping, Fluorescence lymphography, lymphatic alterations, lymphatic vessel abnormalities, clinical trials, lymphatic capillary regeneration, orbital lymphatics, lymph fluid physiology, lymph fluid, cutaneous periarteritis nodosa, lymph detox, Lymphatic Vessel Hyperplasia, puffy hand syndrome, lymphatic valves, lymph flow, lymphedema roadshow, Thoracic lymphatics, Abdominal Lymphatics

Moderators: jenjay, Cassie, patoco, Birdwatcher, Senior Moderators

Second valve system in lymphatics: endothelial microvalves

Postby patoco » Tue Oct 24, 2006 12:56 pm

Evidence for a second valve system in lymphatics: endothelial microvalves

Department of Bioengineering, The Whitaker Institute for Biomedical Engineering, University of California, San Diego, La Jolla, California 92093-0412, USA

1Correspondence: Department of Bioengineering, The Whitaker Institute for Biomedical Engineering, 9500 Gilman Dr., Engineering Bldg. I, Room 5606, University of California San Diego, La Jolla, CA 92093-0412, USA. E-mail:


The mechanism for interstitial fluid uptake into the lymphatics remains speculative and unresolved. A system of intralymphatic valves exists that prevents reflow along the length of the lymphatic channels. However, these valves are not sufficient to provide unidirectional flow at the level of the initial lymphatics. We investigate here the hypothesis that initial lymphatics have a second, separate valve system that permits fluid to enter from the interstitium into the initial lymph channels but prevents escape back out into the tissue. The transport of fluorescent microspheres (0.31 µm) across endothelium of initial lymphatics in rat cremaster muscle was investigated with micropipette manipulation techniques. The results indicate that microspheres can readily pass from the interstitium across the endothelium into the lumen of the initial lymphatics. Once inside the lymphatic lumen, the microspheres cannot be forced out of the lumen even after elevation of the lymphatic pressure by outflow obstruction. Reaspiration of the microspheres inside the lymphatic lumen with a micropipette is blocked by the lymphatic endothelium. This blockade exists whether the aspiration is carried out at the microsphere entry site or anywhere along the initial lymphatics. Nevertheless, puncture of the initial lymphatic endothelium with the micropipette leads to rapid aspiration of intralymphatic microspheres. Investigation of lymphatic endothelial sections fixed during lymph pumping shows open interendothelial junctions not found in resting initial lymphatics. These results suggest that initial lymphatics have a (primary) valve system at the level of the endothelium. In conjunction with the classical (secondary) intralymphatic valves, the primary valves provide the mechanism that facilitates the unidirectional flow during periodic compression and expansion of initial lymphatics.—Trzewik, J., Mallipattu, S. K., Artmann, G. M., DeLano, F. A., Schmid-Schönbein, G. W. Evidence for a second valve system in lymphatics: endothelial microvalves.

Key Words: initial lymphatics • lymphatic endothelium • lymphatic valves • rat • cremaster muscle • interstitium • permeability


The initial lymphatics consist of a highly attenuated endothelial lining and serve to collect interstitial fluid, proteins, and cells. Initial lymphatics have no smooth muscle media and feed into a set of contractile lymphatics, which in turn carries fluid toward the lymph nodes and the central thoracic ducts. Lymph flow can be generated by local enhancement of interstitial fluid pressure above the pressures inside the lymphatics (1 , 2) . Lymph pumping can be enhanced by cyclic compression and expansion of the initial lymphatics, a process that depends on periodic motion of tissue structures positioned in immediate proximity to the initial lymphatics (3) . Rhythmic tissue deformations such as walking (4) , arterial pulse and vasomotion (5 6 7) , intestinal peristalsis (8) , muscle movement (9) , or skin massage (10 11 12) serve to enhance lymph flow.

An important unresolved issue in lymphology is the mechanism that provides the unidirectional transport of fluid from the interstitium into the initial lymphatics. Although reversal of motion of lymph fluid inside the initial or collecting lymphatics is prevented by a set of well-described intralymphatic valves (13 , 14) , these valves are insufficient to prevent fluid return into the interstitium. They form a barrier only to reflow inside a lymphatic lumen. We postulate here that a second valve system exists. We shall refer to it as the primary valve system, since interstitial fluid needs to first pass across these valves on its way into the initial lymphatics. The traditional intralymphatic valves will be referred to as the secondary valves, since fluid passes through them after entry into the lymphatic lumen. The primary valve system is required to prevent fluid escape from the initial lymphatics back into the interstitial space.

We hypothesize that the primary valve system is located at the level of endothelial cells in the initial lymphatics. It permits easy entrance of fluid from the interstitial space into the lymphatics, but no significant escape back into the interstitial space. We provide here evidence for the existence of this primary lymph valve system in individual initial lymphatics of rat cremaster skeletal muscle by use of a set of microinjection and aspiration experiments with a fluorescent tracer.

Materials and Methods

All in vivo experiments were reviewed and approved by the University of California San Diego Animal Subject Committee. After general anesthesia (pentobarbital, 40 mg/kg, intraperitoneal; Abbott Laboratories, North Chicago, IL), the femoral vein of Wistar rats (Charles River Breeding Laboratory, Wilmington, MA; 220–400 g) was cannulated (50 PE tubing, Clay Adams, Parsippany, NJ). Supplemental anesthesia (5 mg/kg) was administered at intervals determined by a tail pinch response. Coagulation inside the catheter was kept to a minimum by use of heparin (10 U/ml in saline).

Muscle preparation
The cremaster muscle was surgically exteriorized according to the method by Baez (15) and placed on a microscope stage for transillumination and intravital microscopy. Throughout the experiments, the muscle tissue was continuously superfused with Krebs-Henseleit bicarbonate-buffered solution at 37°C (pH 7.4, 305 mOsm, saturated with a 95% N2 and 5% CO2).

Microinjection of fluorescent microsphere
The cremaster muscle was visualized by an intravital microscope (Ploempak, Leitz Wetzlar, Stuttgart, Germany) with a 3.5x objective lens (Leitz). The transport of lymph fluid was monitored with fluorescent microspheres (0.31 µm, FITC, estapor®, Bang Laboratories) and injected into the cremaster muscle interstitium near (within 50 to 100 µm) larger arterioles (diameter 80 to 100 µm). This site was selected since the majority of arterioles in skeletal muscle are accompanied by initial lymphatics (7) .

The microspheres were injected via a micropipette mounted on a micromanipulator (Narishige Scientific Instrument Laboratory, Tokyo, Japan). Micropipettes with a tip diameter between 10 and 30 µm were prepared from capillary tubes (FHC, Brunswick, ME; 10 mm, OD 1 mm, ID 0.5 mm) with a pipette puller. Photobleaching of the fluorescent probe and oxidative damage to the tissue were minimized by limiting the observation period to four to six light exposures per hour, each less than 5 min. At each time point, bright-field and fluorescence images of the selected muscle areas were videotaped and viewed on a color monitor. Both bright-field and fluorescent images were recorded with a color-coupled charge device (CCD) camera (Model V1479, Optronics, Goleta, CA) and a videocassette recorder (Panasonic AG 1270).

The video images were analyzed off line and digitized in black and white with image analysis software (Optimas 6.5, Media Cybernetics, Silver Spring, MD). Fluorescent intensities were measured in terms of optical density (0–255 gray levels) and expressed in terms of the light intensity
LI=(I-I b) /255
where I refers to the light intensity over the observation site and IB refers to the background intensity outside the area of the fluorescent microspheres. Fluorescent intensities are normalized with respect to peak intensity (255 gray level units).

Lymphatic cross sections
For histological examination, 15 tissue samples from 8 cremaster muscles were fixed by superfusion of 10% formalin for 15 min at the same time that an oscillatory muscle compression was applied. Three tissue samples from three different rats were fixed under the same conditions but without application of an oscillatory compression. The tissues were postfixed in 1% OsO4, dehydrated, and embedded in araldite resin (Polysciences, Warrington, PA); 1 µm-thick sections were stained with toluidine blue and examined by light microscopy (Diavert, Leitz) with and without oil immersion objectives (10x, 40x, and 100x).

Cyclic cremaster muscle compression
The cremaster muscle is a thin, flat muscle of almost uniform thickness (800 µm). During the intravital observations, it rests on a rigid glass slide. To enhance lymph pumping, the muscle tissue was compressed with a glass coverslip placed on its top surface. An inflatable latex tubing (OD 4 mm, ID 2 mm) was mounted on the top coverslip to serve as the compression device. The latex tubing was inflated in a sinusoidal pattern by a pump such that the amplitude and frequency of the coverslip displacement could be adjusted to preselected values. The compression device was mounted on a micromanipulator (Narishige) to permit precise positioning over the muscle. The use of the coverslip permits observation of the tissue during periods of compression. The frequency of compression was set at 1 Hz, a typical physiological order-of-magnitude frequency.

After injection of microspheres into the muscle interstitium, the microvasculature was observed with and without oscillatory compression for selected periods. The position of the glass coverslip was adjusted with the micromanipulator to achieve a surface displacement between 100 and 150 µm. The microvasculature in the cremaster muscle was observed to assure that the tissue compression did not result in a compromise of the circulation through the arterioles and venules in the center of the muscle.

Lymphatic outflow obstruction
In selected experiments, the lymphatic outflow channels draining the cremaster muscle were obstructed with a clamp at the insertion point of the muscle. These occlusion sites were in close proximity to the major arteriolar/venular pair feeding the muscle microcirculation and typically at a distance of 3 cm from the microsphere injection sites.

The measurements are presented as mean ± SD. Group comparisons were carried out by a two-tailed Student’s test for unpaired data. A probability P < 0.05 was considered statistically significant.


Microsphere transport into initial lymphatics
The lymphatics in the cremaster muscle are positioned near arterioles and larger venules. In previous histological sections and network reconstructions, we have not been able to find initial lymphatics in the capillary network of skeletal muscle per se (7 , 9) . The initial lymphatics exhibit no inherent smooth muscle activity. Collecting lymphatics with smooth muscle media are located outside the muscle tissue.

A small volume (1 µl) of fluorescent microspheres was deposited into the interstitium 250 µm from the arterioles and venules (Fig. 1A , B , C ). In the absence of muscle compression, no entry of microspheres into initial lymphatics was observed over a period of 1.5 h (Fig. 1D , E ). Application of oscillatory surface compressions for periods of several minutes led to filling of adjacent initial lymphatics with microspheres (Fig. 1F ). This local entry of the microspheres occurred irrespective of the particular interstitial position along the length of an initial lymphatic where the microspheres were deposited. Thus, the initial lymphatics absorb material anywhere along their length. The filling of initial lymphatics could be further enhanced by microinjection of a larger volume of microsphere suspension (300 µl), a situation that causes significant local swelling of the muscle interstitium. The initial lymphatics are positioned along an arteriolar/venular pair (Fig. 2A ).

Microsphere retention without and with lymphatic pressure elevation
No microspheres could be detected to escape the lumen along the length of the initial lymphatic duct back into the interstitium whether the microspheres had entered the lymphatics by periodic muscle compression or interstitial volume expansion.

The retention of microspheres in the lymphatic lumen was observed even after occlusion of the proximal outflow of the initial lymphatics at the root of the cremaster muscle in combination with oscillatory muscle compression over the microinjection site (Fig. 2B ). No direct micropressure measurements inside the initial lymphatics were carried out. However, previous studies have shown that such a procedure serves to raise the intralymphatic fluid pressures to values of 30 to 40 cmH2O and even higher, depending on the force applied during the periodic tissue compression (16) . Microsphere dispersion in the tissue was measured in terms of the linear dimensions of the microsphere pool over the interstitial injection site, and the linear dimensions within the lymphatic lumen measured normal to the lymphatic long axis (Fig. 2A ). Whereas dispersion of microspheres inside the interstitial space was observed during lymph outflow obstruction and muscle compression (for 30 min), the width of the microsphere column inside the lymphatic did not change significantly (Fig. 3 ).

Microsphere retention in lymphatics during interstitial fluid aspiration
We used two kinds of micromanipulation studies to examine whether the initial lymphatic endothelium may serve as an one-way barrier. First, microspheres were injected via a micropipette (Fig. 4A ) and aspirated shortly afterward (within 5 s) at the same site into the same micropipette by application of a negative fluid pressure (-100 mmHg) (Fig. 4B ). A fraction of microspheres remained in the tissue at the injection site, but a significant amount of microsphere suspension could be reaspirated from the interstitium back into the micropipette (large arrows). In contrast, microspheres that had entered the lumen of the initial lymphatics remained inside these lymph vessels (Fig. 4C , small arrows). This pattern was confirmed by direct fluorescent light intensity measurements over the lymphatic channel and over the adjacent interstitial injection site. There was no change in fluorescence light intensity over the lymphatic vessel during fluid aspiration whereas the light intensity in the interstitium decreased on average 50% (Fig. 5 ). This local valve action could be observed whatever the location along the initial lymphatics.

Second, initial lymphatics were filled by injection of 300 µl of microsphere suspension as described above (Fig. 2 , Fig. 6A ). Next, a fresh micropipette was placed with its tip outside the wall of an initial lymphatic and an aspiration pressure of 10 cm H2O was applied for 5 min (Fig. 6B ). No significant number of microspheres could be collected into the micropipette under these circumstances. We saw the same phenomenon even when the aspiration pressure was lowered to -100 mmHg. But when the pipette tip was advanced to where it punctured the lymphatic vessel, fluorescent microspheres could be removed from a local segment of the initial lymphatic within a fraction of a second (Fig. 6C ). Measurements of fluorescent light intensity (Fig. 7 ) confirm this observation. This evidence indicates that the microspheres maintain their mobility in the lymphatic lumen, an observation in line with the unrestricted transport of microspheres along lymphatic channels.

Lymphatic endothelial flap junctions
Histological sections through the initial lymphatics exhibited no openings when the muscle was fixed in a noncontracting resting state. In contrast, when the muscles were fixed during muscle contraction, we encountered many open junctions (Fig. 8 ). The endothelial openings were frequently larger than 1 µm and serve as open channels for fluid or particulate transport. The density of these openings is significantly increased if the muscles are fixed during active contraction (Table 1 ). No openings could be seen in any of the adjacent microvascular endothelial cells of arterioles, capillaries, or venules.


The results described here support the hypothesis that the endothelium in the initial lymphatics serves as a primary valve mechanism that permits fluid flow from the interstitial tissue with significantly reduced leakage back into the interstitial space. The evidence was obtained by direct observation of a fluorescent tracer in the immediate vicinity of the initial lymphatic endothelium. Although entry of tracer particles into the initial lymphatics could be readily achieved by periodic compression of the muscle tissue or by elevation of the fluid volume in the immediate vicinity of the lymphatic endothelium, no significant loss of tracer particles out of the vessel was observed. Micropipette aspiration of fluid at the point of the initial lymphatic endothelium where the tracer particle had entered into the lymphatic lumen (Fig. 4) did not remove a significant number of microspheres out of the lymphatic lumen. No microspheres inside the lumen could be reaspirated even within seconds after they had entered the initial lymphatics or at any subsequent time, indicating the capacity for rapid opening and closing action by the primary valve mechanism. Neither could microspheres be removed anywhere else along the length of an initial lymph channel by aspiration with a micropipette located adjacent but outside its endothelium (Fig. 6) . All parts of the initial lymphatics possess the valve-like features. In contrast, puncture of the initial lymphatic endothelium rapidly led to aspiration of tracer microspheres from the lymph lumen. Thus, the initial lymphatic endothelium may serve as a unidirectional transport barrier.

Periodic compression and expansion of the initial lymphatics serves as a key mechanism for lymph pumping (7 , 9) . Pulse pressure oscillations and/or vasomotion, for example, provide such periodic motion. Lymph formation can be enhanced by application of oscillatory pressure on the muscle surface, low enough not to interrupt the muscle microcirculation. Such periodic muscle compression may occur under physiological conditions of tissue movement, as during walking, skin massage, intestinal peristalsis, or respiration.

Two phases make up a single pump cycle in the initial lymphatics. During expansion, an initial lymphatic channel is filled with interstitial fluid. This is a period when the primary lymphatic valves need to be open and the secondary valves inside the lymphatics are closed to prevent reflow of fluid inside the lymphatic. During lymphatic channel compression, fluid is transported along the lumen of the initial lymphatics in a proximal direction toward the contractile lymphatics and nodes. The primary valves need to be closed during this phase of the cycle to prevent escape of fluid back into the interstitium while the secondary valves along the lymphatic lumen are open. In this respect, the lymphatic system may work not unlike a manual pair of bellows at a fireplace. Bellows also require cyclic expansion and compression and two valves for unidirectional airflow.

What mechanism may constitute a primary lymphatic valve system? The paucity of tight cell junctions and adhesion molecules such as VE-catherins (17 18 19) between lymphatic endothelial cells suggests that the junctions may have the form of cellular flaps that could act as valves. Neighboring lymphatic endothelial cells have at their junctions overlapping cytoplasmic extensions (20 21 22 23 24) that may be separated from each other due to the lack of interendothelial adhesion molecules. We hypothesize that such overlapping junctions should be opened during expansion of an initial lymphatic when the intralymphatic pressure falls below the ambient tissue fluid pressure. During compression, however, when the intralymphatic pressure rises, the overlapping endothelial junctions in initial lymphatics are compressed and thereby become sealed. Thus, lymphatic endothelium is leaky when fluid enters from the interstitial space but is tight for fluid transport in the reverse direction.

Cross sections through the initial lymphatics show open lymphatic endothelial flaps in muscle that has been periodically compressed during tissue fixation. No open junctions can be observed on cross sections of any initial lymphatic fixed in a resting state without oscillatory muscle compression (Table 1) .

We have observed endothelial openings in histological sections with dimensions of up to 2 µm. Thus, colloidal material can readily enter into the initial lymphatics across the primary valves. Transport studies with larger tracers show that even cells the size of lymphocytes are carried in large numbers during periodic tissue compression along the lymphatics (25) . This may suggest that the junctions between endothelial cells in initial lymphatics may open to even larger dimensions than could be demonstrated with the tissue fixation technique described here.

The interendothelial junctions can readily be separated if tension is applied to the lymphatic wall, e.g., by overinflation of the lymphatic lumen (21) . There are local regions along the interendothelial junctions where neighboring endothelial cells can be separated when tension is applied to them. When these junctions are stretched open, one can identify endothelial attachment to the underlying basement membrane and to anchoring filaments, and detect large openings between endothelial cells that permit free fluid entry in and out of the initial lymphatics (21) . The primary lymphatic valves are expected to cease functioning under such edematous conditions.

In the presence of a primary and secondary valve systems, the initial lymphatics serve as an efficient unidirectional transport system that depends on periodic compression and expansion provided by surrounding tissue structures. The primary lymphatic valves appear to be provided by the specialized junctions in the endothelium of initial lymphatics that are depleted of interendothelial adhesion molecules. Further work is under way to analyze the ultrastructure of these junctions and their biomechanical properties.


This research was supported by NSF grant IBN 9876379, the German Akademischer Austausch Dienst, and in part by grant 516/40001798 (Ministerium für Schule, Wissenschaft und Forschung des Landes Nordrhein-Westfalen, Germany).

Received for publication February 9, 2001. Revision received April 5, 2001.


Aukland, K., Reed, R. K. (1993) Interstitial-lymphatic mechanisms in the control of extracellular fluid volume. Physiol. Rev. 73,1-78[Abstract/Free Full Text]
Swartz, M. A., Kaipainen, A., Netti, P. A., Brekken, C., Boucher, Y., Grodzinsky, A. J., Jain, R. K. (1999) Mechanics of interstitial-lymphatic fluid transport: theoretical foundation and experimental validation. J. Biomech. 32,1297-1307[Medline]
Schmid-Schönbein, G. W. (1990) Mechanisms causing initial lymphatics to expand and compress to promote lymph flow. Arch. Histol. Cytol. 53,107-114[Medline]
White, J. C., Field, M. E., Drinker, C. K. (1933) On the protein content and normal flow of lymph from the foot of the dog. Am. J. Physiol. 103,34-44
Intaglietta, M., Gross, J. F. (1982) Vasomotion, tissue fluid flow and the formation of lymph. Int. J. Microcirc. Clin. Exp. 1,55-65[Medline]
Parsons, R. J., McMaster, P. D. (1938) The effect of the pulse upon the formation and flow of lymph. J. Exp. Med. 68,353-376
Skalak, T. C., Schmid-Schönbein, G. W., Zweifach, B. W. (1984) New morphological evidence for a mechanism of lymph formation in skeletal muscle. Microvasc. Res. 28,95-112[Medline]
Simmonds, W. J. Q. (1957) The relationship between intestinal motility and the flow and rate of fat output in thoracic duct lymph in unanaesthetized rats. J. Exp. Physiol. 42,205-221
Mazzoni, M. C., Skalak, T. C., Schmid-Schönbein, G. W. (1990) The effect of skeletal muscle fiber deformation on lymphatic volume. Am. J. Physiol. 259,H1860-H1868[Abstract/Free Full Text]
Brace, R. A., Taylor, A. E., Guyton, A. C. (1977) Time course of lymph protein concentration in the dog. Microvasc. Res. 14,243-249[Medline]
Ikomi, F., Schmid-Schönbein, G. W. (1996) Lymph pump mechanics in the rabbit hind leg. Am. J. Physiol. 271,H173-H183[Abstract/Free Full Text]
Ohhashi, T., Yokoyama, S., Ikomi, F. (1991) Effects of vibratory, stimulation and mechanical massage on micro- and lymph-circulation in the acupuncture points between the paw pads of anesthetized dogs. Niimi, H. Zhuang, F. Y. eds. Recent Advances in Cardiovascular Diseases ,125-133 Natl. Cardiovasc. Ctr Osaka, Japan.
Casley-Smith, J. R. (1977) Lymph and lymphatics. Kaley, G. Altura, B. M. eds. Microcirculation ,423-502 University Park Baltimore, MD.
Mazzoni, M. C., Skalak, T. C., Schmid-Schönbein, G. W. (1987) The structure of lymphatic valves in the spinotrapezius muscle of the rat. Blood Vessels 24,304-312[Medline]
Baez, S. (1973) An open cremaster muscle preparation for the study of blood vessels by in-vivo microscopy. Microvasc. Res. 5,384-394[Medline]
Ikomi, F., Zweifach, B. W., Schmid-Schönbein, G. W. (1997) Fluid pressures in the rabbit popliteal afferent lymphatics during passive tissue motion. Lymphology 30,13-23[Medline]
Casley-Smith, J. R. (1972) The role of the endothelial intercellular junctions in the functioning of the initial lymphatics. Angiologica 9,106-131[Medline]
Sauter, B., Foedinger, D., Sterniczky, B., Wolff, K., Rappersberger, K. (1998) Immunoelectron microscopic characterization of human dermal lymphatic microvascular endothelial cells. Differential expression of CD31, CD 34, and type IV collagen with lymphatic endothelial cells vs. blood capillary endothelial cells in normal human skin, lymphangioma, and hemangioma in situ. J. Histochem. Cytochem. 46,165-176[Abstract/Free Full Text]
Schmelz, M., Moll, R., Kuhn, C., Franke, W. W. (1994) Complex adhaerentes, a new group of desmoplakin-containing junctions in endothelial cells: II. Different types of lymphatic vessels. Differentiation 57,97-117[Medline]
Azzali, G. (1999) The lymphatic vessels and the so-called ‘lymphatic stomata’ of the diaphragm: a morphologic ultrastructural and three-dimensional study. Microvasc. Res. 57,30-43[Medline]
Castenholz, A. (1984) Morphological characteristics of initial lymphatics in the tongue as shown by scanning electron microscopy. Scanning Elect. Microsc. 1984,1343-1352
Collan, Y., Kalima, T. V. (1974) Topographic relations of lymphatic endothelial cells in the initial lymphatic of the intestinal villus. Lymphology 7,175-184[Medline]
Ohtani, O. (1987) Three-dimensional organization of lymphatics and its relationship to blood vessels in rat small intestine. Cell Tissue Res 248,365-374[Medline]
Wenzel-Hora, B. I., Berens von Rautenfeld, D., Majewski, A., Lubach, D., Partsch, H. (1987) Scanning electron microscopy of the initial lymphatics of the skin after use of the indirect application technique with glutaraldehyde and MERCOX as compared to clinical findings. Lymphology 20,126-144[Medline]
Ikomi, F., Hunt, J., Hanna, G., Schmid-Schönbein, G. W. (1996) Interstitial fluid, protein, colloid and leukocyte uptake into interstitial lymphatics. J. Cell. Physiol. 81,2060-2067

* * * * * *

Lymphedema People
Site Admin
Posts: 2175
Joined: Thu Jun 08, 2006 9:07 pm

Return to Lymphedema Articles

Who is online

Users browsing this forum: No registered users and 1 guest