methods for lymphatic vessel culture and gene transfection

Methods for Lymphatic Vessel Culture and GeneTransfection

ANATOLIY A. GASHEV,* MICHAEL J. DAVIS,$ OLGA Y. GASHEVA,* ZHANNA V.NEPIUSHCHIKH,* WEI WANG,* PATRICK DOUGHERTY,* KATHERINE A. KELLY,*

SHIJIE CAI,% PIERRE-YVES VON DER WEID,§ MARIAPPAN MUTHUCHAMY,*

CYNTHIA J. MEININGER,* AND DAVID C. ZAWIEJA*

*Department of Systems Biology and Translational Medicine, College of Medicine, CardiovascularResearch Institute, Division of Lymphatic Biology, Texas A&M Health Science Center, Temple,

Texas, USA; $Department of Medical Pharmacology & Physiology, University of Missouri, Columbia,Missouri, USA; %Department of Clinical Pharmacology, University of Oxford, Oxford, UK;

§Department of Physiology and Biophysics, University of Calgary, Calgary, Alberta, Canada

ABSTRACTObjective: To develop the techniques needed for the specific gene/protein targeting transfectionexperiments in isolated lymphatic vessels, we completed two major tasks: 1) optimize theexperimental conditions to maintain the viability of isolated rat lymphatic vessels in culture forsufficiently long periods of time to permit knockdown or overexpression of selected proteins/genesand 2) develop effective transfection protocols for lymphatic muscle and endothelial cells in intactlymphatic vessels without nonspecific impairment of lymphatic contractile function due to thetransfection protocol itself.Methods: Experimental protocols were developed for the maintenance of isolated lymphatic vesselsunder nonpressurized and pressurized conditions for 3�12 days in culture and for adenoviral genetransfection of the lymphatic muscle and endothelial cells.Results: The data demonstrate the effectiveness of the newly developed experimental protocols forthe maintenance of isolated rat mesenteric lymphatic vessels and thoracic duct in culture up to 3�12days without significant impairment of the parameters of their pumping and effective adenoviral/GFP transfection of lymphatic endothelial and muscle cells in isolated rat mesenteric lymphaticvessels.Conclusions: These experimental techniques will extend the set of the modern experimental toolsavailable to researchers investigating the physiology of lymphatic function.Microcirculation (2009) 16, 615�628. doi:10.1080/10739680903120778

KEY WORDS: lymphatic vessel, tissue culture, adenoviral transfection

INTRODUCTION

It is an accepted paradigm of modern lymphologythat all major functions of the lymphatic system,namely, the maintenance of fluid and macromole-cular homeostasis, lipid uptake from gut, andimmune cell trafficking, require an effective trans-port of lymph (containing fluid, macromolecules,and cells). These processes cannot be maintained

without the normal contractility of the muscularcollecting lymphatic vessels [2,15,20,33,39,40].Lymphatic muscle cells regulate lymph flow byinfluencing both the resistance to lymph flow andthe generation of lymph propulsion. Lymphaticvessel resistance is determined largely by the levelof lymphatic muscle tone, whereas the force oflymph propulsion critically depends on the strengthand frequency of the rapid phasic contractionsof lymphatic muscle cells (see recent reviewsin [15,20,31]). Both of these tonic and phasiccomponents of lymphatic muscle contractilityare regulated by multiple mechanisms, therebyconstantly adjusting lymphatic pumping ability tothe moment-by-moment combinations of extrinsicand intrinsic influences [15,20,31]. To date,only a few publications (reviewed in [31]) provide

Address correspondence to Anatoliy A. Gashev, Department ofSystems Biology and Translational Medicine, Collegeof Medicine, Cardiovascular Research Institute, Division ofLymphatic Biology, Texas A&M Health Science Center,702 SW H.K. Dodgen Loop, Temple, TX 76504, USA.E-mail: [email protected]

Received 17 February 2009; accepted 16 June 2009.

Microcirculation, 16: 615�628, 2009Copyright # 2009 Informa UK Ltd.ISSN: 1073-9688 print / 1549-8719 onlineDOI: 10.1080/10739680903120778

sparse and incomplete information on molecularregulatory mechanisms of tonic and phasic lympha-tic contractions.

To further investigate these mechanisms, over thepast several years, we have invested significant effortin developing methodologies to study the regulatorymechanisms of tonic and phasic lymphatic contrac-tions. Because lymphatic muscle cells possess amixture of smooth and striated contractile proteinisoforms different from other muscle-containingtissues [30], it is not easy to perform the same typesof studies that have previously been conducted onother tissue types. In addition, the potential ofgenetically modified mouse models, which arewidely used in modern biological studies, is criticallydiminished for studies on regulatory mechanisms ofthe lymphatic muscle contractility because: 1) theliterature [28,29,32,37] and our own unpublisheddata suggest that murine vessels are not highly orconsistently contractile, 2) spontaneous contractionshave only been demonstrated in the DDY strain[28,29,32,37], which is not typically used fortransgenic crosses, and 3) genetically modifiedmouse models suffer from the problem that genesother than the targeted one may contribute tocompensatory phenotypic alterations [7,8,11,21,27,42,43]. Recently, a new method was describedto image intrinsic lymph pumping in mice [26] to:‘‘(1) assess lymphatic function in transgenic mousemodels to better understand the role that specificgene expression has on lymphatic function; and (2)investigate pharmacologic agents that stimulate theformation of functional lymphatics as well asstimulate the contractile apparatus of existing lym-phatics’’ [41]. Unfortunately, careful review of thosestudies raises several serious questions and concernsabout their conclusions. These data and the accom-panying online video clips [26] were time com-pressed three- to five-fold above real time for‘‘demonstration purposes’’ and would be moreaccurately interpreted as the slow tonic contractionsof mouse lymphatic vessels. Additionally, the possi-bility of overdistension of mouse lymphatic vesselsby abnormal volume expansion could not be ex-cluded in that study. Moreover, it is well known thatnear infrared fluorophores, including indocyaninegreen (ICG), can distribute freely in the body [13]and have particularly high rates of cellular uptakeand laser-induced photooxidation [1]. This seemslikely to be important given the high sensitivity oflymphatic muscle cells to biologically active sub-stances, in particular to free radicals/oxidative

agents [45,46] that can be produced during photo-oxidation. While the laser irradiation in studies byAbels et al. [1] was more intense than that in thework by Kwon and Sevick-Muraca [26], no phar-macological tests were done to evaluate or excludethis potential influence of ICG on their ability torecapitulate normal functioning lymphatic vessels.We conclude that this work [26] did not providestrong support for the principle of using the mousemodels to study phasic lymphatic contractility.Because of all currently available data describedabove, the potential power and specificity ofusing lymphatic preparations from geneticallymodified mice for studies of the regulation oflymphatic function has been unattainable, despitethe expenditure of significant time and effort.

Therefore, in an effort to attain alternatives togenetically modified mouse lymphatic models, webegan to develop techniques in which protein/genefunction is altered over the course of up to two weeksin lymphatic muscle and/or endothelium, usingisolated rat lymphatic vessels. Rat vessels are parti-cularly suited for gene transfection, since the musclelayer is only one to two cell layers thick. To developthese techniques, we focused on two major tasks: 1)optimize the experimental conditions to maintain theviability of isolated ‘‘normal’’ rat lymphatic vesselsin culture for sufficiently long periods of time topermit effective knockdown or overexpression ofselected proteins/genes and 2) develop effectivetransfection protocols for lymphatic muscle cells inintact lymphatic vessels without nonspecific impair-ment of lymphatic contractile function due to thetransfection protocol per se. In the present article, wedescribe the details of experimental techniques thatmeet these goals and which now are available for usein studies on molecular mechanisms of regulation oflymph transport.

METHODS

Animals and Surgery

Mesenteric lymphatic vessels and thoracic ductswere isolated from male Sprague-Dawley rats(weighing between 300 and 400 g). The animalfacilities used for these studies were accredited bythe Association for the Assessment and Accreditationof Laboratory Animal Care and adhered to theregulations, policies, and principles detailed in thePublic Health Service Policy for the Humane Careand Use of Laboratory Animals (PHS Policy, 1996)

Lymphatic vessel culture and gene transfection616 AA Gashev et al.

and the U.S. Department of Agriculture’s AnimalWelfare Regulations (Animal Welfare Act, AWA,9CFR, 1985, 1992). All animal proceduresperformed for this study were reviewed andapproved by our Institutional Animal Care and UseCommittee.

To isolate the thoracic duct, rats were euthanizedwith pentobarbital (120 mg/kg body weight intra-peritoneal, i.p.). The animal was positioned on itsback, the ventral chest wall was opened by lateralincision, and the sternum and approximately half ofthe ribs were excised. The inferior vena cava wasligated and cut close to the diaphragm. The lungsand heart were pulled to the left side of the animal soas to expose the thoracic duct between the aorta andvertebral column. The thoracic duct was then care-fully cleared of all surrounding tissue using adissecting microscope. Extreme caution was usedto prevent direct contact with the thoracic duct at alltimes, thereby reducing the likelihood of damage.The segment of interest was kept moist for theperiod of dissection by using standard Dulbecco’sphosphate-buffered saline (DPBS) (catalog no.14040-133; Invitrogen Corporation, Carlsbad, CA,USA). Sections of thoracic duct 1�2 cm long weredissected and used for these experiments.

To isolate mesenteric lymphatic vessels, the rat wasanesthetized with a solution containing a combina-tion of fentanyl/droperidol (0.3 mL/kg intramuscu-lar, i.m.) and diazepam (2.5 mg/kg i.m.). A 4-cm-long midline abdominal incision was made throughthe skin, underlying fascia, and muscle layers. Asmall loop of intestine (6�7 cm in length) wasexteriorized through the incision. A section of themesentery containing lymphatic vessels was posi-tioned in a dissection chamber within the field ofview of a dissecting microscope and continuouslysuffused with DPBS. Suitable mesenteric lymphaticvessels were identified and cleared of all surroundingtissue. Sections of mesenteric lymphatic vessels1.0�1.5 cm in length were carefully dissected andused for experiments. After isolating mesentericlymphatic vessels, the rat was euthanized withpentobarbital (120 mg/kg body weight i.p.).Throughout the experiments, we measured thediameters of the thoracic duct and mesentericlymphatic vessel segments used for these studies.At a transmural pressure of 3 cm H2O, the outerdiastolic diameters of thoracic duct segmentswere�500�600 mm; the outer diastolic diametersfor mesenteric lymphatic vessel segments, at

a transmural pressure of 5 cm H2O, were �130�160 mm.

Isolated Lymphatic Vessel Procedures for FunctionalTests

Once the lymphatic segment was exteriorized, it wasused for one of three protocols: 1) acute functionaltests, 2) controlled-pressure, isolated vessel cultureexperiments, or 3) transfection experiments followedby subsequent functional tests. For all functionaltests, the lymphatic segment was transferred to anisolated vessel chamber (a modified Living SystemsInstrumentation single-vessel chamber, model CH/1, with a homemade multichannel temperaturecontroller system), filled with room-temperatureDulbecco’s modified Eagle medium (DMEM: nutri-ent mixture F12 (D-MEM/F12) (catalog no. 11039;Invitrogen). The isolated lymphatic vessel wascannulated and tied onto glass pipettes. Thein- and outflow pipettes were connected to indepen-dently adjustable pressure reservoirs filled with D-MEM/F12. Great care was taken to ensure that therewere no air bubbles in the tubing or the pipettes.Once the vessel was cannulated, a slight positivetransmural pressure (2�3 cm H2O) was applied todetect leaks and to ensure that the vessel wasundamaged and untwisted. The vessel was set toits approximate in situ length and positioned justabove the glass coverslip comprising the chamberbottom. The chamber was transferred to the stage ofa microscope. The vessel was set to an equilibrationtransmural pressure of 3 cm H2O and warmed to 38o

C over 15�20 minutes. Once tone or spontaneouscontractions developed, the vessel was allowed toequilibrate at 3 cm H2O for another 30 minutes. Avideo camera, high-resolution monitor and DVD/HDD recorder were used to observe and recordactivity of the lymphatic segments continuously inall experiments.

For every experiment requiring functional tests, weevaluated the pressure-induced contractile responsesof the lymphatic segments. Segments of mesentericlymphatic vessels were exposed to a range oftransmural pressures: 1, 3, 5, and 7 cm H2Ofor five minutes at each pressure; thoracic ductsegments were exposed to pressures of 1, 2, 3, and 5cm H2O. We chose these sets of transmural pressuresbecause isolated rat mesenteric lymphatic vesselsand thoracic duct display maximal active pumpingat 5 cm H2O and 3 cm H2O, respectively[14,16,18,20]. At the end of each experiment, the

Lymphatic vessel culture and gene transfection617AA Gashev et al.

passive (i.e., relaxed) diameter was measured ateach level of transmural pressure after the vessel wasexposed to a nominally calcium-free, ethylenedia-minetetraacetic acid (EDTA) (3.0 mM)-supplemen-ted physiological salt solution (PSS) (in mM: 145.0NaCl, 4.7 KCl, 1.2 MgSO4, 1.2 NaH2PO4, 5.0dextrose, 2.0 sodium pyruvate, and 3.0 MOPS) for15 minutes.

Data Analysis and Statistics for Isolated VesselExperiments

Lymphatic diameters were tracked from the DVDrecords of experiments using ‘‘vessel track’’ software,developed previously [9,10]. We used cardiac pumpanalogies to define systole and diastole in reference tothe lymphatic contractile cycle [3,16,23,46], and theend-diastolic and end-systolic points in the diametertracings were recorded for each five-minute intervalfor each set of pressures. To normalize changes indiameter from vessel to vessel, diastolic and systolicdiameters were normalized to the passive lymphaticdiameter in Ca-free PSS at the correspondingtransmural pressure. The anatomy and basal con-traction frequencies of lymphatic vessels in manyspecies (e.g., rats, dogs, guinea pigs, humans, etc.)are quite variable from vessel to vessel, even whensimilar vessels from age-, weight-, and sex-matchedanimals are studied [4,22,44]. The normalization oflymphatic diameters minimizes animal-to-animalvariability in basal lymphatic parameters and allowsfor a more sensitive investigation of the controlledparameters by minimizing the effect of randomvariation in vessel size between animals [17,20].From the lymphatic and end-systolic and -systolicdiameters, the following lymph pump parameterswere calculated: lymphatic tone index (the differencebetween the passive lymphatic diameter in Ca-freePSS and end-diastolic diameter expressed as apercentage of the passive lymphatic diameter inCa-free PSS), contraction amplitude (the differencebetween the normalized diastolic and systolic dia-meters), contraction frequency (number of contrac-tions per minute), ejection fraction (the fraction ofend-diastolic volume ejected during a single phasiclymphatic contraction), and fractional pump flow(an index of minute lymph pump flow, calculated asthe ejection fraction times the contraction fre-quency).

Statistical differences were determined by two-wayanalysis of variance (ANOVA, SAS Institute Inc.,Cary, NC, USA), regression analysis, and Student’s

t-test (JMP software, version 5.0.1.2. for Windows),as appropriate, and considered significant at PB

0.05. In the Results section below, the numbers oflymphatic vessels used in the reported data areshown separately for each group of experiments,where n depicts the number of lymphatic segmentsused for each experimental protocol.

Culture of Isolated Lymphatic Vessels

Culture of Lymphatic Vessels under NonpressurizedConditions. Following dissection, isolated segmentsof rat thoracic duct and mesenteric lymphaticvessels, sufficiently clean to perform subsequentfunctional tests after vessel culture, were gentlyrinsed ablumenally in sterile DPBS. This procedurewas performed in standard 35-mm plastic Petridishes filled with DPBS under a dissecting micro-scope to reduce possible vessel contamination. Afterthis procedure, all subsequent manipulations wereperformed using sterile instruments, dishes, andsolutions under a laminar flow hood. After tworinses in new Petri dishes, each lymphatic segmentassigned for organ culture was transferred to an-other sterile 35-mm Petri dish filled with prewarmed(388C) D-MEM/F12 (catalog no. 11039; InvitrogenCorp.), supplemented with antibiotic mixture(catalog no. 15140; Invitrogen) to achieve a con-centration of 100 units of penicillin/100 mg ofstreptomycin per milliliter of D-MEM/F12 (here-after referred to as D-MEM/F12/a/b). Each lym-phatic segment was placed (unpressurized) in aseparate Petri dish and gently pressed to the bottomof the dish by lightly touching the edges with forcepsto keep the segment in a nearly straight position onthe dish bottom. Afterward, the Petri dishes weretransferred into a cell-culture incubator with 10%CO2 at 378C and maintained for 3�16 days. Duringculture, all segments were checked daily and disheswith visible signs of contamination were discarded.The culture media was not changed. After the vesselculture period was complete, the lymphatic vesselswere used for functional tests.

Culture of Lymphatic Vessels under PressurizedConditions and the Development of a Model SimulatingChronically High Lymph Pressure. In separate proto-cols, isolated lymphatic segments were maintained inorgan culture (up to 12 days) under conditions ofcontrolled intralymphatic pressure. Rat mesentericlymphatic vessel segments were isolated, cannulated,and maintained in an isolated vessel chamber similar


to that used routinely for functional tests [19,20].The chamber, was sterilized by overnight exposure to100% alcohol, and filled with prewarmed sterile(388C) D-MEM/F12/a/b. The isolated lymphaticsegment was cannulated and tied onto two alcohol-sterilized glass pipettes in the same manner asdescribed above for acute functional tests. Thechamber was transferred to the stage of a micro-scope. In one set of experiments, pressurized vesselsegments were maintained for three to four days atan intraluminal pressure of 3 cm H2O (with a singlecase of daily observation for 12 days). In a second setof experiments, vessel segments were maintainedovernight at an intraluminal pressure of 3 cm H2O toconfirm the preservation of normal contractility andthen maintained for three to four days at anintraluminal pressure of 10 cm H2O. In thoseexperiments, elevated intralymphatic pressure wasused to simulate conditions that occur in differentdiseases of the lymphatic system involving partial orcomplete insufficiency of regional lymphatic drai-nage [5,12,34�36,38]. Because pressurized, culturedlymphatic vessels were maintained outside of alaminar flow hood in a standard isolated vesselchamber attached to a microscope specifically de-signed for functional tests, we were able to performfunctional tests (described above) at any time pointduring the culture conditions. We performed suchtests at least two times daily throughout the protocol.

The following preventive measures reduced thelikelihood of contamination in pressurized lympha-tic vessel culture: 1) We used an internal cultureroom without windows, with diminished air flow inthe room and with the room door closed wheneverpossible; 2) everything in contact with the vessel wasalcohol-sterilized (i.e., instruments, sutures, vesselchamber, glass pipettes, and glass reservoirs); 3)plastic tubing, stopcocks, and connectors werereplaced after use in every experiment; 4) the uppercover of the vessel chamber was closed duringculture; 5) the input and output reservoirs werecovered with parafilm to prevent contamination byairborne particles; 6) to eliminate the need torecannulate the vessel when air bubbles formedinside the pipettes or tubing, we continuously heatedthe D-MEM/F12/a/b solution in the input pressureline to 388C; 7) a slow-speed (0.15 mL/min)peristaltic pump was used to constantly replenishD-MEM/F12/a/b at 388C through the isolated vesselchamber, and each opened bottle of freshly preparedD-MEM/F12/a/b was used within a 24-hour periodwith any remaining solution discarded.

Transfections of Isolated Rat Mesenteric LymphaticVessels: Development of Protocol Using a GFPReporter Gene

To develop protocols that would effectively transfectlymphatic muscle cells in intact vessels withoutimpairment of lymphatic contractile function, weused recombinant adenoviral vectors to express afluorescent reporter signal, green fluorescent protein(GFP). Transfected vessels were imaged confocallyto determine the intensity of the signal in cells ofthe lymphatic wall. The segments were observed at40�50X magnification, using a Leica AOBS SP2Confocal-Multiphoton Microscope System (LeicaMicrosystems Heidelberg Gmbh, Mannheim, Ger-many), and confocal images were acquired at0.3-mm intervals via 489 nm of peak excitationwavelength and 508 nm peak emission wavelengthin order to completely image the entire lymphaticvessel. The image stacks were reconstructed usingdifferent three-dimensional (3D) perspectives toproduce the projections, with the Leica confocalsoftware. After preliminary tests to optimize theadenoviral GFP (AdGFP) transfections, we wereable to transfect more than 90% of lymphaticmuscle cells. A subset of the AdGFP-transfectedvessels was checked functionally, as describedabove.

Experimental Protocol for Adenoviral Transfection ofIsolated Mesenteric Lymphatic Vessels by AdGFPConstructs. The GFP gene was cloned into thepShuttleCMV plasmid and used to generate arecombinant adenovirus (AdGFP) under the controlof the cytomegalovirus immediate-early promoter,using the AdEasy system [25], as previouslydescribed [6]. Purified virus was used to transfectlymphatic muscle cells in whole isolated lymphaticvessels, as described below. In total, we performedmore than 50 transfections and are providing here asummary of the optimal experimental protocol. Ourcriteria for successful transfection included finding85�90% GFP-positive muscle cells in the lymphaticwall; 12 of the last 14 transfected lymphatic vesselsmet these criteria.

We performed adenoviral GFP (AdGFP) transfec-tions in the smallest possible volumes as practical.We slightly modified, for these purposes, the stan-dard 0.65-mL plastic tubes (catalog no. 53550-100;VWR, West Chester, PA, USA). Their caps wereseparated; three holes were penetrated throughthem: one hole allowed the insertion of a small-diameter plastic tube�10 cm in length (catalog no.


63010-007; VWR); two other holes helped to relievepressure inside the tube after closure, thereforepreventing the prolapse of the lymphatic wall insidethe micropipette. Plastic tubing from the inner-capside was connected with�1-cm-length glass micro-pipette with a tip diameter of�100�120 mm (madefrom capillary glass tubes; catalog no. 9-000-3000;Drummon Scientific, Drummond Scientific Com-pany, Broomall, PA, USA); the other end of thetubing was connected to a 1-mL syringe. After theplastic tubing and pipette were filled with D-MEM/F12/a/b and positioned in a large plastic Petri dish,the lymphatic segments were cannulated at theupstream end onto the micropipette, and the can-nulated end of the vessel was secured by a 6-0 suture

with the output end remaining open. This method ofhandling the lymphatic segment allowed us toreplace solutions inside the vessel.

The solution containing the AdGFP construct wasprepared by the following method. Immediatelybefore loading the isolated vessels for transfection,the adenoviral constructs were mixed with antenna-pedia leader peptide CT (ALP-CT) (catalog no.61376; Anaspec, Inc., Fremont, CA, USA) by mixingequal volumes of ALP-CT stock solution to non-diluted adenoviral construct stock solution (1*1013

viral particles/mL) for 30 minutes at room tem-perature. Antennapedia leader peptide is known toincrease viral transfection efficiency in intact vessels[24]. In our experience, supplementation of viral

Figure 1. The parameters of lymphatic contractile activity in isolated segments of the rat thoracic duct after cultureunder nonpressurized conditions. Day zero: parameters of lymphatic contractile activity in the control group ofnon-cultured lymphatic segments (n�6); days three to six: parameters of lymphatic contractile activity in a group oflymphatic segments cultured for three to six days (n�7); days seven to nine: similar parameters for segments culturedfor seven to nine days (n�8). *Indicates significant differences (PB0.05) within each group between parameters at aninitial level of transmural pressure of 1 cm H2O versus all other levels of transmural pressure during the acutefunctional tests.


particles with this peptide increased AdGFP effi-ciency of lymphatic muscle cells from �10�15% to�85�90%. After pretreatment of AdGFP constructswith ALP-CT, we diluted this mixture by D-MEM/F12/a/b to produce a final titer of viral particlesranging between 1.1 and 1.5*1010 viral particles/mL.

After the vessel segment was cannulated and securedat its upstream end with a suture to the glassmicropipette (110�130 mm diameter of the tip),the vessel lumen was flushed with�0.2 mL of D-MEM/F12/a/b, using a 1-mL syringe. For denuda-tion of the endothelium, an air bolus was insertedinto the vessel lumen to replace all the fluid inside ofthe vessel with air (the syringe was disconnectedfrom the opened input tubing and �0.05 mL of air

was sucked into it). The air bolus was maintainedinside the vessel for 10�12 seconds, and then thevessel was immediately flushed with �0.2 mL of D-MEM/F12/a/b from the syringe to allow the com-plete replacement of air with fluid. Whether or notendothelial denudation was performed in differentvessels, the D-MEM/F12/a/b solution inside of thelymphatic vessel was replaced by ALP-CT/AdGFP-solution from another 1-mL syringe and the vesselwas tied at its output end after it was filled with thevirus suspension. The syringe was disconnected fromthe plastic tubing, leaving the input end of the vesselopen. The 0.65 mL plastic tube was filled byD-MEM/F12/a/b on �2/3 of tube’s volume (withor without virus at 1.1*1010 viral particles/mL), andthen the vessel, pipette, and plastic cap were rapidlytransferred from the Petri dish and inserted into the

Figure 2. The parameters of lymphatic contractile activity in isolated segments of the rat mesenteric lymphaticvessels after culture under nonpressurized conditions. Day zero: parameters of lymphatic contractile activity in thecontrol a group of noncultured lymphatic segments (n�5); days seven to nine: parameters of lymphatic contractileactivity in group of lymphatic segments cultured for seven to nine days (n�10). *Indicates significant differences (PB

0.05) within each group between parameters at initial level of transmural pressure of 1 cm H2O and at all other levels oftransmural pressure during the acute functional tests; 2indicates significant differences (PB0.05) between lymphatictone index values at day zero and at days seven to nine.


plastic tube. Care was taken to avoid puncturing thevessel or bending it at the pipette tip. Next, theplastic tube, containing the cannulated vessel, waspositioned into a tube rack, so that the open end ofthe plastic tubing, filled with fluid, was placed �5cm above the pipette tip to allow the vessel to bepressurized at all times during the subsequentculture. The tube rack was placed into a cell-cultureincubator (10% CO2; 378C) for the required numberof hours for transfection with subsequent confocalimaging and functional tests, as described above.The vessels were checked daily. Any vessels withvisible signs of contamination were discarded.

RESULTS

Cultured Nonpressurized Lymphatic Vessels

Figure 1 summarizes the results of experiments withculture of nonpressurized isolated segments of ratthoracic duct. At day 0 (the control group), thesevessels demonstrated typical contractile responsesthat have been well characterized in the recentliterature [19,20] with optimal lymphatic pumpingobtained at a 2�3-cm H2O intraluminal pressure.Vessel culture through day 9 did not significantlychange any of the measured/calculated parametersof lymphatic contractility, including: lymphatic toneindex, contraction amplitude (amplitude of phasiccontractions), contraction frequency, and fractionalpump flow. At culture days seven to nine, weobserved only a small, but insignificant, trend ofreductions in contraction frequency and pump flow,together with a slight increase in lymphatic tone. Inthe group of vessels cultured for 14�16 days (datanot shown), we observed a significant decline oflymphatic phasic contraction amplitude in thethoracic duct segments, yet without a completecessation of spontaneous phasic contractions.

Figure 2 summarizes the results of similar experi-ments with culture of nonpressurized rat isolatedsegments of the mesenteric lymphatic vessels. At dayzero (the control group), these vessels demonstratedthe features of their contractile responses similar tothose described recently [16], with optimal lympha-tic pumping near 5 cm H2O intraluminal pressure.Vessel-culture conditions up to nine days did notsignificantly affect the contraction amplitude, con-traction frequency, or fractional pump flow incultured, unpressurized mesenteric lymphatic seg-ments. However, after seven to nine days of culture,lymphatic tone in these segments was significantly

decreased*on average, by 34.5% at intraluminalpressures from 1 to 5 cm H2O and by 25.2% at 7 cmH2O. In the group of mesenteric lymphatic vesselscultured for 14�16 days (data not shown), we alsoobserved a significant inhibition of phasic contrac-tion amplitude, yet without a complete cessation ofspontaneous phasic contractions.

Cultured Pressurized Lymphatic Vessels

Figures 3�6 summarize the results of experimentswith culture of isolated rat mesenteric lymphatic

Figure 3. Lymphatic tone in isolated segments of the ratmesenteric lymphatic vessels cultured at normal (A: 3 cmH2O; n�5) or high (B: 10 cm H2O; n�4) intraluminalpressures. *Indicates significant differences (PB0.05)within each group (A and B) between lymphatic toneindex at day zero of culture (control group) and at allother days of culture, shown separately for each level oftransmural pressure during the acute functional tests;2indicates significant differences (PB0.05) betweenlymphatic tone index values in groups A and B for eachgiven pressure/day pair.


vessels under pressurized conditions. In each figure,the data were analyzed in two groups: vesselsmaintained at 3 cm H2O for the entire time inculture (panel A) and vessels initially incubatedovernight at 3 cm H2O (to verify normal contractilefunction) and then subsequently maintained at 10cm H2O (panel B).

None of the parameters of lymphatic function (i.e.,lymphatic tone index, lymphatic phasic contractionamplitude, contraction frequency, and fractionalpump flow) were significantly changed during threedays of culture at a pressure of 3 cm H2O. The

contractile activity of these vessels initially was inthe normal range (compared to previous acutestudies of function [16,17,19]) and was not sig-nificantly different from that of vessels tested at dayzero of culture at an intraluminal pressure of 3 cmH2O (the control group). In contrast, over the firsttwo days of culture at 10 cm H2O, lymphatic tonewas dramatically increased (Figure 3B), comparedto tone at day zero before exposure to the high-pressure conditions. For example, during the firsttwo days of culture at high intraluminal pressure,the vessels demonstrated 2.2- and 2.4-fold increases

Figure 5. Contraction frequency in isolated segments ofthe rat mesenteric lymphatic vessels cultured at normal(A: 3 cm H2O; n�5) or high (B: 10 cm H2O; n�4)intraluminal pressures. *Indicates significant differences(PB0.05) within each group (A and B) between contrac-tion frequency at day zero of culture (control group) andat all other days of culture, shown separately for each levelof transmural pressure during the acute functional tests;2indicates significant differences (PB0.05) betweencontraction frequency values in groups A and B for eachgiven pressure/day pair.

Figure 4. Contraction amplitude in isolated segments ofthe rat mesenteric lymphatic vessels cultured at normal(A: 3 cm H2O; n�5) or high (B: 10 cm H2O; n�4)intraluminal pressures. *Indicates significant differences(PB0.05) within each group (A and B) between contrac-tion amplitude at day zero of culture (control group) andat all other days of culture, shown separately for eachlevel of transmural pressure during the acute functionaltests; 2indicates significant differences (PB0.05) be-tween contraction amplitude values in groups A and Bfor each given pressure/day pair.


in tone at an intraluminal test pressure of 1 cm H2Oat days one and two, respectively. During these firsttwo days, similar trends were observed for the otherlevels of intraluminal pressure. However, lymphatictone declined toward normal levels after three daysin culture at 10 cm H2O (Figure 3).

Figure 4 shows the phasic contraction amplitude forthe same vessel groups pressurized and cultured forthree days at 3 or 10 cm H2O. Contractionamplitude was significantly diminished in lymphaticvessels cultured at 10 cm H2O (Figure 4B). Forinstance, at a test pressure of 1 cm H2O, we observed

75 and 67% decreases in contraction amplitude forhigh-pressure days one and two, respectively, com-pared to their amplitude before culture at highpressure. However, by day three of culture at 10cm H2O, we observed a slight trend toward therestoration of normal contraction amplitude. High-pressure culture, opposite to culture at 3 cm H2O,produced a rise in contraction frequency, as shownin Figure 5. The maximal increases in contraction

Figure 7. Examples of effective adenoviral/GFP trans-fection of lymphatic endothelial and muscle cells inisolated rat mesenteric lymphatic vessels. Vesseldiameters are �120�140 mm. Average two-dimensionalprojections of the stacks of confocal images, taken in stepsof 0.5 mm through the vessels pressurized at 5 cm H2O inthe standard isolated vessel chamber. (A) Transfection ofmostly lymphatic endothelial cells was achieved by usinga solution containing 1.1*1010 viral particles/mL, whichwas injected only inside of nondenuded vessel. (B)Transfection of only lymphatic muscle cells in endothe-lium-denuded vessels by using a concentration of1.5*1010 viral particles/mL only intraluminally. (C)Transfection of both lymphatic endothelial and musclecells required a solution outside the vessel containing1.1*1010 viral particles/mL together with an intraluminalsolution containing 1.5*1010 viral particles/mL.

Figure 6. Fractional pump flow in isolated segments ofthe rat mesenteric lymphatic vessels cultured at normal(A: 3 cm H2O; n�5) or high (B: 10 cm H2O; n�4)intraluminal pressures. *Indicates significant differences(PB0.05) within each group (A and B) between frac-tional pump flow at day zero of culture (control group)and at all other days of culture, shown separate for eachlevel of transmural pressure during the acute functionaltests; 2indicates significant differences (PB0.05) bet-ween fractional pump flow values in groups A and B foreach given pressure/day pair.


frequency were observed after days one and twounder high pressure, whereas at day three of highpressure, this positive chronotropic effect virtuallydisappeared (Figure 5B). As a result of the counter-balancing effect of reduced amplitude, but increasedfrequency, after chronic exposure to high pressure,fractional pump flow did not dramatically declinein overall function up to day three of culture at10 cm H2O (Figure 6B). However, there was aconsistent, albeit not significant, trend toward thedecompensation of lymphatic pumping over timeunder high-pressure culture conditions.

Adenoviral Transfection of Cultured IsolatedMesenteric Lymphatic Vessels by AdGFP Constructs

Figure 7 shows images of AdGFP-treated isolatedrat mesenteric lymphatic vessels, illustrating suc-cessful transfection of both major types of cells inthe lymphatic wall (i.e., endothelial and musclecells) separately as well as together, depending onvariations in the experimental procedure. Afterseveral tests, we found that for optimal AdGFPtransfection, the incubation time was�66�74hours. By varying the concentration of viral particlesinside and outside the cultured vessels and by usingdenuded or nondenuded vessels, we could achievecombined or separate GFP expression in lymphaticendothelial and/or muscle cells. For fairly selectivetransfection of lymphatic endothelial cells in isolatedrat mesenteric lymphatic vessels, it was sufficient toinject D-MEM/F12a/b containing 1.1*1010 viralparticles/mL only into the lumen of a nondenuded

vessel (Figure 7A). To transfect only lymphaticmuscle cells in endothelium-denuded vessels, aconcentration of 1.5*1010 viral particles/mL onlyin the lumen of a denuded vessel was generallysufficient (Figure 7B), whereas transfection of bothlymphatic endothelial, and muscle cells required atleast 1.1*1010 viral particles/mL�containing solu-tion outside of the vessel together with 1.5*1010

viral particles/mL�containing solution into the lu-men of the vessel (Figure 7C).

We also performed functional tests on AdGFP-transfected vessels (n�4), prior to confocal detec-tion of the GFP signal in lymphatic muscle cells.Figure 8 presents an example of the normal phasiccontraction pattern that was observed at threedifferent pressures and the corresponding GFPexpression in the muscle layer of the same vesseltransfected with AdGFP ablumenally. The para-meters of contractile activity of these vessels weresimilar to recent descriptions of acute function [16].

DISCUSSION

In this study, we performed, for the first time, themaintenance of nonpressurized and pressurized ratisolated lymphatic vessels in tissue culture condi-tions for periods of 3�12 days with preservationof normal function, with the option of continu-ously monitoring their contractility at differentintralymphatic pressures. The vessels could bemaintained for up to 10�12 days without substantialimpairment of regular spontaneous contractions.

Figure 8. An example of the normal phasic contraction patterns at three different pressures observed prior todemonstration of GFP expression by lymphatic muscle cells in the same isolated rat mesenteric lymphatic vessel.Average two-dimensional projections of the stacks of confocal images, taken at steps of 0.5 mm through the vessel whilepressurized at 5 cm H2O.


Finally, we also report, for the first time, effectivegene transfection of lymphatic endothelial andmuscle cells in isolated, cultured, intact lymphaticvessels.

During the present studies, we were able, for the firsttime, to develop experimental protocols to performnonpressurized isolated lymphatic vessel culture forup to nine days without substantial compromise ofthe normal contractile patterns. Such conditionsmay be applicable to subsequent transfection ex-periments to overexpress or knock down genesinvolved in the regulation of lymphatic contractility.We believe that daily changes of the outside solutionmay be used to further improve the technique toeffectively maintain the contractile function of thelymphatic vessels cultured for longer periods of time.However, to avoid the potential influence of moder-ate changes in lymphatic tone on phasic contractileactivity of transfected lymphatic segments, weperformed subsequent transfections on pressurizedlymphatic segments (as described in Methods). Weconcluded that culture and transfection protocols ofisolated lymphatic segments should utilize a con-stantly open, pressurized vessel lumen.

We also developed experimental techniques toculture pressurized rat isolated lymphatic vessels,with the option of continuously monitoring theircontractility at different intralymphatic pressures.We found that after one day at a constantly elevatedintralymphatic pressure, a profound restructuring ofthe contractile pattern in lymphatic muscle cells wasunderway. While the mechanisms involved in suchregulatory reactions of lymphatic vessels will be thesubject of a separate research project, we havealready demonstrated the feasibility of this approachto develop an experimental model of the highlymph-pressure conditions that characterize severallymphatic pathologies. In addition, the ability tocontinuously monitor lymphatic diameter and con-tractile function during organ culture of isolatedlymphatic vessels offers the opportunity to simulatedifferent disorders of lymphatic transport that arecharacterized by increases in intralymphatic pres-sure. Such an experimental model would be helpfulto probe mechanisms of pathogenesis in thesediseases to develop potential preventive measuresand treatments for several kinds of lymphedema.

Finally, we performed, for the first time, genetransfection of lymphatic muscle and endothelialcells in isolated whole-vessel lymphatic segments

without significantly altering function via the trans-fection procedure per se.

CONCLUSIONS

Future improvements of our transfection approachwill likely provide powerful tools to lymphologists toprecisely target genes involved in the regulation oflymphatic contractility and provide an alternative tothe use of genetically modified mouse models toinvestigate the mechanisms of lymphatic contractilefunction.

ACKNOWLEDGEMENTS

This work was supported by National Institutes ofHealth (Bethesda, Maryland, USA) grants HL-070308, HL-075199, AG-030578, HL-080526,HL-089784, and KO2 HL-086650. The authorsthank Aleksey Gashev and Sheng (Max) Yu for theirhelp in primary data analysis of functional tests oflymphatic vessels and thank Dr. Andreea TrachePh.D. and the Texas A&M Health Science CenterMicroscopy Imaging Facility for their help with theconfocal imaging studies.

Declaration of interest: The authors report nofinancial conflicts of interest. The authors alone areresponsible for the content and writing of this paper.

REFERENCES

1. Abels C, Fickweiler S, Weiderer P, Baumler W,Hofstadter F, Landthaler M, et al. (2000). Indocya-nine green (ICG) and laser irradiation induce photo-oxidation. Arch Dermatol Res 292:404�411.

2. Aukland K. (2005). Arnold Heller and the lymphpump. Acta Physiol Scand 185:171�180.

3. Benoit JN, Zawieja DC, Goodman AH, Granger HJ.(1989). Characterization of intact mesenteric lym-phatic pump and its responsiveness to acute edema-genic stress. Am J Physiol 257:H2059�H2069.

4. Borisov AV. (1997). The theory of the design of thelymphangion [in Russian]. Morfologiia 112:7�17.

5. Browse N, Burnand KG, Mortimer PS. (2003).Diseases of the Lymphatics. London: Arnold, HodderHeadline Group.

6. Cai S, Alp NJ, McDonald D, Smith I, Kay J, CanevariL, et al. (2002). GTP cyclohydrolase I gene transferaugments intracellular tetrahydrobiopterin in humanendothelial cells: effects on nitric oxide synthaseactivity, protein levels, and dimerisation. CardiovascRes 55:838�849.


7. Cammalleri M, Cervia D, Dal Monte M, Martini D,Langenegger D, Fehlmann D, et al. (2006). Com-pensatory changes in the hippocampus of somatosta-tin knockout mice: upregulation of somatostatinreceptor 2 and its function in the control of burstingactivity and synaptic transmission. Eur J Neurosci23:2404�2422.

8. Cho JH, Klein WH, Tsai MJ. (2007). Compensationalregulation of bHLH transcription factors in thepostnatal development of BETA2/NeuroD1-nullretina. Mech Dev 124:543�550.

9. Davis MJ. (2005). An improved, computer-basedmethod to automatically track internal and externaldiameter of isolated microvessels. Microcirculation12:361�372.

10. Davis MJ, Zawieja DC, Gashev AA. (2006). Auto-mated measurement of diameter and contractionwaves of cannulated lymphatic microvessels. Lym-phat Res Biol 4:3�10.

11. Denofrio JC, Yuan W, Temple BR, Gentry HR, PariseLV. (2008). Characterization of calcium- and integ-rin-binding protein 1 (CIB1) knockout platelets:potential compensation by CIB family members.Thromb Haemost 100:847�856.

12. Foldi M, Foldi E. (2006). Foldi’s Texbook of Lym-phology. Munich, Germany: Elsevier.

13. Frangioni JV. (2003). In vivo near-infrared fluores-cence imaging. Curr Opin Chem Biol 7:626�634.

14. Gashev AA. (2002). Physiologic aspects of lymphaticcontractile function: current perspectives. Ann N YAcad Sci 979:178�187; discussion, 188�196.

15. Gashev AA. (2008). Lymphatic vessels: pressure- andflow-dependent regulatory reactions. Ann N Y AcadSci 1131:100�109.

16. Gashev AA, Davis MJ, Delp MD, Zawieja DC. (2004).Regional variations of contractile activity in isolatedrat lymphatics. Microcirculation 11:477�492.

17. Gashev AA, Davis MJ, Zawieja DC. (2002). Inhibi-tion of the active lymph pump by flow in ratmesenteric lymphatics and thoracic duct. J Physiol540:1023�1037.

18. Gashev AA, Delp MD, Zawieja DC. (2006). Inhibitionof active lymph pump by simulated microgravity inrats. Am J Physiol Heart Circ Physiol 290:H2295�H2308.

19. Gasheva OY, Knippa K, Nepiushchikh ZV, Muthu-chamy M, Gashev AA. (2007). Age-related altera-tions of active pumping mechanisms in rat thoracicduct. Microcirculation 14:827�839.

20. Gasheva OY, Zawieja DC, Gashev AA. (2006).Contraction-initiated no-dependent lymphatic re-laxation: a self-regulatory mechanism in rat thoracicduct. J Physiol 575:821�832.

21. Gerlai R. (2001). Gene targeting: technical confoundsand potential solutions in behavioral brain research.Behav Brain Res 125:13�21.

22. Gnepp DR. (1984). Lymphatics. In: Staub NC,Taylor AE, eds. Edema (pp. 263�298). New York:Raven Press.

23. Granger HJ, Kovalcheck S, Zweifach BW, Barnes GE.(1977). Quantitative analysis of active lymphaticpumping. In: Proceedings of the VII Summer Com-puter Simulation Conference. Simulation Council, LaJolla, California, USA. 562�565.

24. Gratton JP, Yu J, Griffith JW, Babbitt RW, ScotlandRS, Hickey R, et al. (2003). Cell-permeable peptidesimprove cellular uptake and therapeutic gene deliv-ery of replication-deficient viruses in cells and in vivo.Nat Med 9:357�362.

25. He TC, Zhou S, da Costa LT, Yu J, Kinzler KW,Vogelstein B. (1998). A simplified system for gen-erating recombinant adenoviruses. Proc Natl AcadSci U S A 95:2509�2514.

26. Kwon S, Sevick-Muraca EM. (2007). Noninvasivequantitative imaging of lymph function in mice.Lymphat Res Biol 5:219�231.

27. Miosge N, Sasaki T, Timpl R. (2002). Evidence ofnidogen-2 compensation for nidogen-1 deficiency intransgenic mice. Matrix Biol 21:611�621.

28. Mizuno R, Ono N, Nojiri H, Ikomi F, Ohhashi T.(1998). Spontaneous pumping activity of rat andmice isolated lymph microvessels. Microcirc Annu14:69�70.

29. Mizuno R, Ono N, Ohhashi T. (2001). Parathyroidhormone-related protein-(1-34) inhibits intrinsicpump activity of isolated murine lymph vessels. AmJ Physiol Heart Circ Physiol 281:H60�H66.

30. Muthuchamy M, Gashev A, Boswell N, Dawson N,Zawieja D. (2003). Molecular and functionalanalyses of the contractile apparatus in lymphaticmuscle. FASEB J 17:920�922.

31. Muthuchamy M, Zawieja D. (2008). Molecular reg-ulation of lymphatic contractility. Ann N Y Acad Sci1131:89�99.

32. Nojiri H, Ono N, Mizuno R, Ikomi F, Ohhashi T.(1998). Spontaneous contractions in mice mesentericlymph vessels: with special reference to arrangementof smooth muscle layers in the walls. Microcirc Annu14:67�68.

33. Ohhashi T, Mizuno R, Ikomi F, Kawai Y. (2005).Current topics of physiology and pharmacology in thelymphatic system. Pharmacol Ther 105:165�188.

34. Olszewski W. (1981). Surgical pathophysiology ofthe lymphatic system. Z Exp Chir 14:9�20.

35. Olszewski WL. (2002). Contractility patterns ofnormal and pathologically changed human lympha-tics. Ann N Y Acad Sci 979:52�63; discussion, 76�59.

36. Olszewski WL. (2008). Contractility patterns ofhuman leg lymphatics in various stages of obstructivelymphedema. Ann N Y Acad Sci 1131:110�118.

37. Ono N, Mizuno R, Nojiri H, Ohhashi T. (2000).Development of an experimental apparatus for


investigating lymphatic pumping activity of murinemesentery in vivo. Jpn J Physiol 50:25�31.

38. Rockson SG. (2006). Lymphedema. Curr TreatOptions Cardiovasc Med 8:129�136.

39. Schmid-Schonbein G, Zweifach B. (1994). Fluidpump mechanisms in initial lymphatics. News Phy-siol Sci 9:67�71.

40. Schmid-Schonbein GW. (1990). Microlymphaticsand lymph flow. Physiol Rev 70:987�1028.

41. Sharma R, Wendt JA, Rasmussen JC, Adams KE,Marshall MV, Sevick-Muraca EM. (2008). Newhorizons for imaging lymphatic function. Ann N YAcad Sci 1131:13�36.

42. Takahashi E, Ino M, Miyamoto N, Nagasu T. (2004).Increased expression of P/Q-type Ca2� channelalpha 1a subunit mRNA in cerebellum of n-typeCa2� channel alpha 1b subunit gene-deficient mice.Brain Res Mol Brain Res 124:79�87.

43. Wang Q, Hummler E, Maillard M, Nussberger J,Rossier BC, Brunner HR, et al. (2001). Compensatoryup-regulation of angiotensin ii subtype 1 receptors inalpha enac knockout heterozygous mice. Kidney Int59:2216�2221.

44. Yoffey JM, Courtice FC. (1970). Lymphatics, Lymphand the Lymphomyeloid Complex. London, NewYork: Academic Press.

45. Zawieja DC, Davis KL. (1993). Inhibition of theactive lymph pump in rat mesenteric lymphatics byhydrogen peroxide. Lymphology 26:135�142.

46. Zawieja DC, Greiner ST, Davis KL, Hinds WM,Granger HJ. (1991). Reactive oxygen metabolitesinhibit spontaneous lymphatic contractions. AmJ Physiol 260:H1935�H1943.

This paper was first published online on iFirst on 22 July 2009.


methods for lymphatic vessel culture and gene transfection

Documents