The present protocol describes a detailed surgical protocol for isolating flowing intestinal lymph in response to intraduodenal nutrient infusions in mice. This allows for the physiological determination of total intestinal lipid absorption and chylomicron synthesis and secretion in response to various experimental nutrients.
Intestinal lipoproteins, especially triglyceride-rich chylomicrons, are a major driver of metabolism, inflammation, and cardiovascular diseases. However, isolating intestinal lipoproteins is very difficult in vivo because they are first secreted from the small intestine into the mesenteric lymphatics. Chylomicron-containing lymph then empties into the subclavian vein from the thoracic duct to deliver components of the meal to the heart, lungs, and, ultimately, whole-body circulation. Isolating naïve chylomicrons is impossible from blood since chylomicron triglyceride undergoes hydrolysis immediately upon interaction with lipoprotein lipase and other lipoprotein receptors in circulation. Therefore, the original 2-day lymph fistula procedure, described by Bollman et al. in rats, has historically been used to isolate fresh mesenteric lymph before its entry into the thoracic vein. That protocol has been improved upon and professionalized by the laboratory of Patrick Tso for the last 45 years, allowing for the analysis of these critical lipoproteins and secretions from the gut. The Tso lymph fistula procedure has now been updated and is presented here visually for the first time. This revised procedure is a single-day surgical technique for installing a duodenal feeding tube, cannulating the mesenteric lymph duct, and collecting lymph after a meal in conscious mice. The major benefits of these new techniques include the ability to reproducibly collect lymph from mice (which harnesses the power of genetic mouse models); the reduced anesthesia time for mice during the implantation of the duodenal infusion tube and the lymph cannula; the ability to continuously sample lymph throughout the feeding and post-prandial period; the ability to quantitatively measure hormones and cytokines before their dilution and enzymatic hydrolysis in blood; and the ability to collect large quantities of lymph for isolating intestinal lipoproteins. This technique is a powerful tool for directly and quantitatively measuring dietary nutrient absorption, intestinal lipoprotein synthesis, and chylomicron secretion.
The physiological importance of the mesenteric lymphatic system
The mesenteric lymphatics have been described in some form since ~300 BC when Herophilos described the hepatic portal system and all the "absorptive veins in the intestines"1,2,3,4,5. For more than 1,700 years after this initial description, a defining characteristic of intestinal lymphatics was the presence of milky fluid in the mesenteric lymph shortly after a meal3. It is now known that milky, chylomicron-containing lymph ("chylous" lymph) does not drain into the portal vein and liver but instead travels through the cisterna chyli into the thoracic duct and, ultimately, joins the blood at the left subclavian vein6. Due to this anatomical arrangement, chylous lymph first travels through the heart and lungs before circulating through the rest of the body. This means that the heart and lungs get a "first-pass" at the secretions into the mesenteric lymph7.
A major role of mesenteric lymphatics is their transport of dietary lipids from the small intestine8,9,10. Anatomically, chylomicrons, intestinal immune cells11,12,13,14, gut hormones15,16,17,18, antigens19,20,21, non-chylomicron lipophilic compounds22,23, and excess fluid enter the lacteals underlying the enterocyte basement membrane and are then concentrated through the lymphatic capillaries to the mesenteric lymph nodes. There are likely many unknown mesenteric lymphatic components, including metabolites, antigens, environmental contaminants, nutrients, and signaling molecules.
The components of mesenteric lymph have not been systematically identified, largely due to the difficulty of isolating mesenteric lymph. Accessing the mesenteric lymph has always been a serious challenge because lymph ducts are colorless, and except for after a fatty meal when they become chylous and milky, the mesenteric lymphatics contain colorless lymphatic fluid6,8,9,10.
Current and commons methods for isolating intestinal lymph
Mesenteric lymph cannot be accessed in humans (except for rare and extreme circumstances where serious GI trauma has occurred)24. In vivo lymph collection is surgically complex and demanding. The original 2 day lymph fistula procedure was described by Bollman et al.25 and has been improved upon and professionalized by the laboratory of Patrick Tso for the last 45 years26,27. The lymph fistula procedure allows investigators to collect flowing lymph from conscious animals during a 6 h duodenal lipid infusion period.
The lymph fistula model has mainly been used in rats to measure lymph flow rate, the output of triglyceride and cholesterol (or other duodenal-infused compounds), chylomicron composition, and intestinal hormone concentrations. To a lesser extent, this technique can also be used in mice, though surgical survival and lymph volume suffer. Due to the difficulties in seeing mesenteric lymph ducts, historical methods have included cannulating mesenteric lymph in larger animals like mini-pigs28, sheep29, pigs30, dogs31, and rats17. Working with these larger models is resource intensive and does not allow for studies in knock-in or knock-out models.
Alternative approaches have also been used. Chylomicrons can be isolated from blood in the post-prandial state (though these will be partially hydrolyzed by plasma lipoprotein lipase)32,33,34,35. The thoracic duct can also be cannulated, though the lymph collected there contains an admixture of mesenteric lymph and extra-intestinal lymph drainage26,36. In vitro, Caco-2 cells secrete a chylomicron-like particle in response to fatty acid treatment, and these cells can be co-cultured with relevant lymphatic endothelial or vascular cells37,38. Human and mouse intestinal organoid cultures have been shown to process apical lipids and secrete chylomicrons39,40,41,42. These models are highly advantageous and enable mechanistic insights into small intestinal physiology, but they cannot replicate the complexity, physio-chemical gradients, or dynamic lymph flow of in situ mesenteric lymphatics.
Advantages of the 1 day mouse lymph fistula model presented here
With respect to these other methods of isolating intestinal lipoproteins, the Tso Lab lymph fistula technique has traditionally been considered the gold-standard technique for measuring the absorption of dietary nutrients into mesenteric lymph. This in vivo technique has the advantage of capturing key physiological aspects of dietary lipid absorption-the dynamic appearance of compounds over the absorption period, which requires the repeated sampling of flowing mesenteric lymph in live animals with duodenal nutrients. This surgical technique also measures gut hormones and cytokines directly in their physiological compartment rather than in blood, where they are diluted and enzymatically degraded17,43.
If the experimental question requires an understanding of lipid secretion dynamics or the dynamic absorption and metabolism of any hydrophobic GI compound or drug, then this technique is not only appropriate but is also the only approach that interpolates the movement of luminal contents from the proximal to the distal gut (stomach to colon) and from the apical to the basolateral surface (luminal contents through enterocyte to lacteals and portal circulation). As this technique employs the luminal delivery of nutrients through the intraduodenal catheter, and because the flowing mesenteric lymph is diverted and collected, the entire absorption apparatus is under experimental control and can be used to qualitatively assess small intestinal absorption profiles.
Presented visually here for the first time is a major update to the Tso Lab lymph fistula model, which (1) reduces the experimental time to a 1 day surgical implantation and experimental collection period; (2) improves mouse survival and animal welfare considerations; and (3) increases the reproducibility of the approach in mice to harness the power of mouse genetic models. This technique must be considered a gold standard for all experimental questions of intestinal secretions, lipoproteins, or dietary lipid absorption and is the best technique for the high-fidelity determination of lipid absorption kinetics and chylomicrons.
All surgical procedures were approved by the University of Pittsburgh Internal Animal Care and Use Committee [Protocol # 20047008] and comply with the NIH Guide for the Care and Use of Laboratory Animals. C57BL6/J male mice, 8-14 weeks of age, were used for the present study. The mice were obtained from a commercial source (see Table of Materials). All mice were housed on a 12 h light/dark cycle with ad libitum access to standard chow and water.
1. Animal preparation
2. Surgery and experimental design
3. Collection of luminal contents and mucosal tissue
4. Thin-layer chromatography (TLC)
5. Chylomicron isolation and characterization
Figure 2 shows the triglyceride secretion in mesenteric lymph and the average lymph flow rates in the most recent n = 17 wild-type (WT) mice who underwent the single-day lymph fistula procedure described here. As shown in Figure 2A, the triglyceride concentration in lymph increases in response to an intraduodenal bolus of 300 µL of SMOFLipid. Peak triglyceride concentration is reached at ~2-3 h post-bolus and decreases steadily through the 6 h time (immediately before euthanasia). In parallel with triglyceride secretion, lymph flow rate increases from Time 0 bolus infusion through the end of the experiment (Figure 2B).
These results show that the surgical implantation of both the mesenteric lymph duct and intraduodenal cannula have occurred and are representative of a positive control to which other lymphatic contents (hormones, peptides, nutrients) could be benchmarked. If there is no change in triglyceride concentration in lymph in response to bolus intraduodenal lipids, this signals that the surgery has caused significant damage to the small intestine so that lipid absorption capacity is absent, or, in previously unphenotyped genetic models, this would suggest that the mouse has a defect in lipid absorption that is physiologically meaningful.
Figure 1: Proposed timeline for the lymph fistula mouse model. T-4, T-3, T-2: Double-cannulation takes approximately 2-4 h, followed by placement of the mice in recovery chambers. T-1: Once the mice are in the recovery period, they receive a continuous duodenal infusion of 5% glucose-0.9% saline. T0: the continuous intraduodenal infusion is switched from 5% glucose-0.9% saline to an infusion of experimental nutrients. T0-T6: Mice receive continuous intraduodenal 5% glucose in saline or, alternatively, continuous nutrients. Lymph is collected hourly during this period. Endpoint: Mice are euthanized, and tissues can be collected. The figure was created with BioRender.com. Please click here to view a larger version of this figure.
Figure 2: Mesenteric lymph triglyceride concentration and flow rate. Wild-type mice were fitted with an intraduodenal feeding tube and a mesenteric lymph cannula. Mice received a bolus infusion of 300 µL of lipid. Lymph was collected for 6 h in hourly aliquots and kept on ice. (A) Lymph triglyceride content was determined by a chemical assay in each hourly aliquot of lymph. (B) In the 6 h after bolus infusion, the lymph flow is plotted as grams of lymph secreted per hour. Points are means ± SEM. Please click here to view a larger version of this figure.
Lipid Delivery Methods | Recommendations/Instructions | |||||
Mixed lipid emulsion bolus | A 0.3 mL dose of either (A) Liposyn III 20% lipid injectable or (B) SMOFlipid 20% lipid injectable emulsion, USP, via intra-duodenal infusion tube can be used. These emulsified lipids are useful because they do not need to be prepared in advance, are liquid at room temperature, are sterile, and because they contain an “easily digestible” mix of free fatty acids, triglyceride, phospholipids, and sodium taurocholate. This is a good starter infusion for mice that may have pancreatic or biliary insufficiencies. | |||||
Triglyceride bolus | 0.3 mL gently warmed olive oil (or purified triolein) as a neutral lipid that reflects a diet rich in unsaturated fat, lard (reflective of a diet with high saturated fat content) or coconut oil (reflective of a diet with high medium chain triglyceride content) can all be given by intra-duodenal infusion tube or oral gavage. If the experimental triglyceride is saturated, it must be warmed to be liquid at room temperature. | |||||
Mixed meal bolus | 0.3 mL Ensure with the formulation containing 0.075 g fat (21.6%), 0.5 g carbohydrate (64.0%), and 0.1125 g protein (14.4%) and reflects a low-fat mixed meal, can be administered via intra-duodenal infusion. | |||||
Radiolabeled lipid bolus | 5.0 µCi 3H-labeled glycerol trioleate and/or 1.0 µCi 14C-cholesterol in 0.3 mL of olive oil can be given via intra-duodenal infusion. 14C-cholesterol: Cholesterol-[4-14C] with specific activity of 55Ci/mmol and a concentration of 0.1mCi/ mL. 3H-TG Triolein [9,10-3H(N)] (3H-TG) with specific activity of 60Ci/mmol and concentration of 1mCi/mL | |||||
Continuous dose | The infusion of any of the above experimental nutrient formulations at a rate of 0.3 mL/h (rather than 0.3 mL total bolus), will result in a steady output of triglyceride into the lymph. This differs from a bolus infusion, where triglyceride concentration in lymph peaks at ~2–3 h, and then returns to baseline concentrations at ~6–8 h. |
Table 1: Table of lipid infusions.
The original 2 day lymph fistula procedure was described by Bollman et al.25 and practiced by the laboratory of Patrick Tso for the last 45 years26,27. The protocol presented here is a powerful addition to this classic, gold-standard method for identifying, quantifying, and understanding the unique chylous secretions of the small intestine, as well as the in vivo dynamics of dietary nutrient absorption and metabolism, gut hormones, and intestinal immunity.
The advantages of this model include (1) the ability to continuously sample mesenteric lymph throughout the feeding and post-prandial period rather than static sampling at one time point during either absorption, digestion, or secretion; (2) the measurement of gut hormones and cytokines directly in their physiological compartment rather than in blood, where they are diluted and enzymatically degraded17,44; (3) the ability to isolate, quantitate, and characterize the lipoproteins secreted by the small intestine following the ingestion of a lipid meal and the absence of endothelial lipases in the mesenteric lymph, which preserves chylomicron triglyceride concentrations and native chylomicron structure46,47; (4) the ability to directly measure lymph flow rate, the output of triglyceride and cholesterol (or other duodenal-infused compounds), chylomicron composition, and intestinal hormone concentrations. Finally, this protocol allows for collecting relatively large quantities of >50 µL of lymph every hour over a 6 h period. As the fluid is replenished with intra-duodenal saline and glucose infusion, the lymph fistula model is significantly improved over other lymph sampling techniques and results in flowing rather than static pools of mesenteric lymph. As volume is a major hurdle for lipidomic, proteomic, and metabolomic approaches, this is a major strength.
The 1 day mouse lymph fistula protocol described here has several advantages over the original lymph fistula protocol, including a reduction in total experimental animal number from previously described 2 day lymph fistula protocols26,27 because of a higher survival rate after surgery; a reduction in the overall experimental time from 2 days to a single day; and, finally, a reduction in the recovery period for mice from overnight (>18 h), where breakthrough pain or poor post-surgical outcomes may occur, to a more manageable ~6 h post-surgery period.
A feature of this 1 day protocol is the focus on humane considerations and endpoints. These must take the highest priority: (1) animals must receive intraduodenal or IV replacement fluids; (2) they must be kept warm and as pain-free as possible (with post-operative Buprenex and/or carprofen, depending upon the experimental design and the need for avoiding anti-inflammatory effects); (3) bleeding, shaking, diarrhea, or signs of distress are all compelling reasons for a humane endpoint. Per strict IACUC guidelines, isoflurane followed by cervical dislocation is a good endpoint. Surgical survival rates are ~70% for a single day (compared to ~40% for the original 2 day surgery), but investigators should not hesitate to end the experiment at a sign of distress. This should be taken into consideration when planning animal numbers.
In terms of troubleshooting this technique, successful placement of the lymph cannula is the major bottleneck in this surgical procedure. While practicing the surgery, it helps to gavage the mouse with 0.3 mL olive oil approximately 2 h prior to surgery. This will cause the secretion of chylomicrons into the mesenteric lymph duct, making it appear "milky" and more visible. Methylene blue can also be used but is often less obvious than the milky lymph duct. If after the placement of the lymph cannula and its placement with glue the mesenteric lymph is successfully flowing through the cannula into a collection vessel, then one can proceed with the placement of a duodenal infusion tube. Occasionally, the lymph may not flow continuously but may start flowing again when the animal is placed on the rotating table. Critically, watch out for clots within the lymph tubing. These should be massaged out of the tubes to prevent backflow pressure on the lymph duct.
In addition to triglyceride secretion and lymph flow rate, this technique can be used to determine the following lipid absorption kinetics and chylomicron characteristics:
Chylomicron secretion45,48,49
Immediately following a meal containing fat, there is a transient rise in circulating plasma triglyceride. As triglycerides are inherently hydrophobic, they must first be emulsified to be soluble in blood50. Small intestinal enterocytes carry out this role and package dietary triglyceride into chylomicron emulsion particles51. Chylomicrons contain cholesterol and dietary triglyceride in their core, surrounded by phospholipids and apolipoproteins, including apoB-48, apoA-I, apoA-IV, and apoC-III52,53,54,55. ApoB-48 is the essential structural protein, and the other apolipoproteins have various functions required for chylomicron metabolism and clearance from the blood. To determine the key characteristics of chylomicrons, including their triglyceride and apolipoprotein content, the 1 day lymph fistula technique shown here should be used. The chylomicron secretion rate is the percent of infused 3H-triglyceride that is secreted into the lymph and measured by scintillation by counting hourly lymph samples. This can be combined with detailed chylomicron characterization48,49,56,57,58. Hourly lymph samples can be combined or kept separately. Lymph is transferred to ultracentrifuge tubes, mixed with 0.9% NaCl, and then carefully overlaid with 300-500 µL of 0.87% NaCl. Samples are then ultra-centrifuged at 110,000 x g at 4 °C for 16 h. The top fractions containing isolated chylomicrons are collected and tested for triglyceride concentrations using the triglyceride assay kit. Briefly, 2 µL of the chylomicron (1:10 dilution) is incubated with 200 µL of enzyme reagent at room temperature for 10 min in a 96-well plate. The plate is read by a plate reader at 500 nm, and the standards and blanks are used for the calculation of triglyceride concentrations. Chylomicron size can then be determined by negative staining and transmission electron microscopy (TEM)14,29. Triglyceride and cholesterol can be quantified by chemical assay and apolipoprotein content (apoB-48, A-I, C-II, C-III) by ELISA Kits or Western blot.
Determining the primary site of lipid absorption48,59
By isolating the luminal and epithelial cell compartments of the duodenum, jejunum, and ileum at 6 h after the infusion of 3H-triglyceride or a radio-labeled mixed meal, the contents are Folch extracted to determine how much of the 3H-triglyceride is absorbed across the epithelial cell membrane (normal) or is retained in the lumen (abnormal) along the length of the small intestine48. These studies are particularly impactful if there are potential differences in GI motility60,61,62, if there is a hypothesis regarding bile acids (highly active throughout lipid absorption and themselves reabsorbed in the ileum)63,64,65,66, or if there is a concern that nutrients are being absorbed in the wrong anatomical location (ileum or even colon)67,68,69,70.
Identifying mechanisms of 3H-triglyceride trafficking into absorptive epithelial cells48
This is performed by calculating the percent of 3H-triglyceride hydrolyzed to 3H-free fatty acid in the intestinal lumen, absorbed into the mucosa, and re-esterified into intracellular 3H-triglyceride prior to secretion as chylomicrons. This is a powerful marker of absorption/secretion defects since it can be traced to show the movement of dietary triglyceride into its breakdown products and subsequent packaging into chylomicrons. mRNA expression of the fatty acid absorption machinery (CD36, FABPs, ACSLs), re-esterification pathway (MGAT, DGAT, MTTP, apoB), and apolipoproteins (apoC-III, B-48, C-II, A-I, A-IV) can be further quantitated by RT-PCR.
Future applications of this technique are only limited by interest in gut-organ crosstalk, metabolism, immunity, nutrient absorption, environmental dietary contaminants, or any other disease with a role in the GI system. It is likely that many compelling experiments and hypotheses have been stalled because of the difficulty in accessing the critical mesenteric lymph system, and the goal of this visualized protocol is to make this technique more readily available. Isolating chylomicrons and the mesenteric lymph in which they initially reside is a critical part of understanding whole-body metabolism; the 1 day mouse lymph fistula model is a powerful physiological model for studying these events.
The authors have nothing to disclose.
We are extremely grateful to the Cystic Fibrosis Foundation (Pilot and Feasibility Award 1810, to ABK), the Rainin Foundation (Synergy Award, to PI GJ Randolph and Co-I ABK), and the National Institutes of Health (R01DK118239, R03DK116011 to ABK) for their support of these studies.
0.9% Sodium Chloride Injection, USP sterile | Hospira, Lake Forest, IL, US | NDC 0409-4888-06 | |
1.7 mL Eppendorf tubes | Fisher Scientific, Waltham, MA, US | 7200184 | |
14C-cholesterol: Cholesterol-[4-14C] (0.1mCi/ mL) | American Radiolabeled Chemicals Inc., St. Louis, MO, US | ARC 0857 | |
18 G needle | Becton Dickinson, Franklin Lakes, NJ, US | 305199 | |
2 Dumont micro-dissecting forceps | Fine Science Tools, Foster City, CA, US | 11251-35 | |
2 Forceps | ROBOZ Surgical Instrument Co., Gaithersburg, MD, US | RS-5130 | |
3H-TAG: Triolein [9,10-3H(N)] (3H-TG) (1mCi/ mL) | American Radiolabeled Chemicals Inc., St. Louis, MO, US | ART0199 | |
3H-TG Triolein [9,10-3H(N)] (3H-TG) | American Radiolabeled Chemicals, INC. St. Louis | MO 63146 | |
50% Dextrode Injection, USP 25grams/50 mL sterile | Hospira, Lake Forest, IL, US | NDC-0409-6648-16 | |
Analtech TLC Uniplates: silica gel matrix Z265500-1PAK | Fisher Scientific, Waltham, MA, US | 11-101-0007 | |
BD CareFusion ChloraPrep Swabstick | Fisher Scientific, Waltham, MA, US | 14-910-501 | |
Betadine surgical scrub | Dynarex Corp., Orangeburg, NY, US | 1201 | |
Bevel-cut cannula | Braintree Scientific., Braintree , MA, US | MRE025 | |
Buprenorphine HCl Injection Carpuject PFS 0.3mg/mL 10/Bx (Buprenex) | HenrySchein, Warrendale, PA, US | 1278184 | |
C57BL6/J male mice | Jackson Laboratory, Bar Harbor, Maine | ||
ChloraPrep | Becton Dickinson, Franklin Lakes, NJ, US | 260100 | |
Cholesterol Assay Kit | FujiFilm Healthcare, Lexington, MA, US | 99902601 | |
Cholesterol-[4-14C] | American Radiolabeled Chemicals, INC. St. Louis | MO 63146 | |
Clear plexiglass box L43cm X W26 X H 21 with temperature-controlled heating pad and humidification | our own design and modifications | ||
commercially available amphibian/reptile heating pad | ShenZhen XingHongChang Electric CO., LTD. ShenZhen, China | XHC-F035D | |
Cotton tip applicators | Fisher Scientific, Waltham, MA, US | 22363156 | |
Duodenal infusion tube – canuala | Braintree Scientific, Braintree , MA, US | MRE037 | |
Ensure | Abbott Nutrition, Columbus, OH | ||
Heating pad surgical platform with circulating warm water pump combination | Patterson Scientific, Waukesha, WI, US | Gaymar T/Pump Classic | |
Hetarin Sodium Injection, USP 1,000 units/mL sterile | Mylan, Morgantown, WV, US | NDC-67457-384-31 | |
Image J Software | National Institute of Health, Bethesda, Maryland, | ||
Iris scissors | ROBOZ Surgical Instrument Co., Gaithersburg, MD, US | RS-5602 | |
Isoflurane | Piramal Pharma Solutions, Riverview, MI, US | NDC 66794-017-25 | |
Isoflurane induction apparatus and Anesthesia Apparatus | Patterson Scientific, Waukesha, WI, US | mouse induction chamber | |
Krazy glue | Elmer's products Inc., Columbus, OH, US | KG484 | |
Liposyn III 20% lipid injectable | Hospira Inc. Lake Forest, Illinois, USA | ||
LS 6500 Multi-Purpose Scintillation Counter | Beckman Coulter, Brea, CA | ||
Micro-dissecting Spring Scissors | ROBOZ Surgical Instrument Co., Gaithersburg, MD, US | RS-5602 | |
Mouse apoB ELISA Kit | ABCAM Inc., Waltham, MA, US | ab230932-1 | |
Needle Holder | Fine Science Tools, Foster City, CA, US | 12002-12 | |
Retractors | Kent Scientific Co., Torrington, CT, US | SURGI-5001 | |
Rimadyl (Carprofen) | Zoetis Inc., Kalamazoo MI, US | 4019449 | |
Rotating table Barnstead Thermolyne | Labquake, Zürich, Switzerland | Barnstead Thermolyne | |
SMOFlipid 20% lipid injectable emulsion, USP | Fresenius Kabi, Warrendale, PA, US | NDC-63323-820-01 | |
Snuggle | Lomir Biomedical Inc., Notre-Dame-de-l'Île-Perrot, QC J7V 7M4, Canada | MS 02.5PM | |
Surgical Scissors | ROBOZ Surgical Instrument Co., Gaithersburg, MD, US | RS-5912SC | |
Suture (5-0 silk with needle) | DemeTECH, Miami Lakes, FL, US | DT-719 | |
Transmission Electron Microscope (JEOL 1400-FLASH 120KV ) | JEOL, Peabody, MA | ||
Triglyceride Assay Kit | Randox Laboratories, Crumlin, United Kingdom | TR210 | |
ULTIMA GOLD XR Scintillation Fluid | Perkin Elmer, Hebron, KY, US | 6013119 | |
Ultracentrifuge, rotor S100AT4-497 | SORVALL RC M120 GX |