Tail Vein Transection Bleeding Model in Fully Anesthetized Hemophilia A Mice

Summary

The refined tail vein transection (TVT) bleeding model in anesthetized mice is a sensitive in vivo method for the assessment of hemophilic bleeding. This optimized TVT bleeding model uses blood loss and bleeding time as endpoints, refining other models and avoiding death as an endpoint.

Abstract

Tail bleeding models are important tools in hemophilia research, specifically for the assessment of procoagulant effects. The tail vein transection (TVT) survival model has been preferred in many settings due to sensitivity to clinically relevant doses of FVIII, whereas other established models, such as the tail clip model, require higher levels of procoagulant compounds. To avoid using survival as an endpoint, we developed a TVT model establishing blood loss and bleeding time as endpoints and full anesthesia during the entire experiment. Briefly, anesthetized mice are positioned with the tail submerged in temperate saline (37°C) and dosed with the test compound in the right lateral tail vein. After 5 min, the left lateral tail vein is transected using a template guide, the tail is returned to the saline, and all bleeding episodes are monitored and recorded for 40 min while collecting the blood. If no bleeding occurs at 10 min, 20 min, or 30 min post-injury, the clot is challenged gently by wiping the cut twice with a wet gauze swab. After 40 min, blood loss is quantified by the amount of hemoglobin bled into the saline. This fast and relatively simple procedure results in consistent and reproducible bleeds. Compared to the TVT survival model, it uses a more humane procedure without compromising sensitivity to pharmacological intervention. Furthermore, it is possible to use both genders, reducing the total number of animals that need to be bred, in adherence with the principles of 3R's. A potential limitation in bleeding models is the stochastic nature of hemostasis, which can reduce the reproducibility of the model. To counter this, manual clot disruption ensures that the clot is challenged during monitoring, preventing primary (platelet) hemostasis from stopping bleeding. This addition to the catalog of bleeding injury models provides an option to characterize procoagulant effects in a standardized and humane manner.

Introduction

Animal models are essential for understanding the pathogenesis of hemophilia and developing and testing treatment regimens and therapies. The Factor VIII knock-out mouse (F8-KO) is a widely used model for the study of hemophilia A^1,2. These mice recapitulate key features of the disease and have been widely used for development of treatments, such as recombinant FVIII products^3,4,5 and gene therapy strategies^6,7.

There are various bleeding injury models for evaluating the pharmacological effects of different hemostatic compounds in vivo. One of these coagulation models is the tail vein transection survival model in mice^{8,9,10,11,12,13,14}, measuring the ability of hemophilic mice to survive exsanguination after tail transection. This method was introduced more than four decades ago¹⁵ and is still used^9,16,17. However, the model utilizes survival as an endpoint and requires observation of the animals over a period of up to 24 h, during which the animals are conscious and hence can experience pain and distress.

Bleeding models of shorter duration and under full anesthesia have been described previously, such as the tail clip model (also known as the tail tip)^{8,18,19,20,21,22,23,24,25,26,27,28}. Nevertheless, for a complete normalization of blood loss after the bleeding challenge, these models require doses of procoagulant compounds (e.g., FVIII) far higher than those administered clinically²⁹. A different injury model under anesthesia, the vena saphena bleeding method, is sensitive to lower doses of procoagulant compounds³⁰ but requires a high level of experimenter intervention since the clots must be disrupted frequently (as opposed to 3 times in the presented model).

Standardization towards a common protocol to test new procoagulant compounds would greatly facilitate data comparison between laboratories^31,32,33. In TVT models, there is not yet a common agreement on studied endpoints (blood loss^7,26, bleeding time^9,34, and survival rate^35,36), and experimental length varies between studies¹³.

Our primary objective is to describe and characterize an optimized model with high reproducibility, the possibility to study on-demand as well as a prophylactic treatment, sensitivity to pharmacological intervention equivalent to the survival model, yet not using death or near-death as endpoints. In order to reduce pain and distress, the animals should not be conscious during bleeding and a more ethical endpoint needs to be implemented³⁷.

Tail clip models are generally conducted in one of two variants, either amputating the tip of the tail, e.g., amputation of 1-5 mm^{18,19,20,21,23,24} or, in a more severe variant, transected at a tail diameter around 1-3 mm^8,22,25. This causes a combined arteriovenous bleed, as the lateral and dorsal veins and ventral artery are usually severed, and in general, the larger the amputation, the lower the sensitivity to a procoagulant compound. Furthermore, since the tail tip is amputated, the arteriovenous injury is exposed without any opposing tissue; thus, at least in theory, it is dissimilar to the most common hemophilic bleeds.

As the name implies, only the vein is injured in tail vein transection models such as described in this paper, thus resulting in an exclusively venous bleed. Since the vessel is not fully severed, the injury is expected to be smaller than in the amputation models, and the tissue around the cut, which a clot may adhere to, is retained. In addition, there is lower blood pressure in the vein as opposed to the artery. These factors contribute to an increased sensitivity relative to amputation models, such that normalization of bleeding can be achieved with clinically relevant doses of replacement therapy, e.g., with rFVIII in hemophilia A, which is useful for evaluating the magnitude and durability of effects of procoagulant treatment^26,38,39.

Protocol

All procedures described in this protocol have been approved by the Animal Welfare Body at Novo Nordisk A/S, and the Danish Animal Experiments Inspectorate, The Danish Ministry of Food, Agriculture, and Fisheries. The optimized 40 minutes method includes anesthesia and dosing time in the design (Figure 1). Hemophilic mice of both genders between 10-16 weeks of age are required for this procedure.

1. Preparations before the study

1. Prepare the dosing solutions in the correct concentrations.
2. Start the water bath and heat to 37 °C. Fill the 15 mL centrifuge tubes for blood collection with saline (0.9% NaCl).
3. Place the 15 mL saline tubes in the holes in the warmed base plate at least 15 min prior to the start of the experiment.
4. Identify the mice and record their weight. Avoid handling mice more than necessary as this can cause stress and affect the study.
5. Prepare the workstation in the fume hood before proceeding so that everything is within reach: napkins, tail holder, gauze, syringes, scalpels, stopwatches, and blood flow notation paper.
6. Place the tail mark and cutting blocks on the heating plate - cold blocks will make the veins contract and thereby affect the bleeding.

2. Anesthesia

1. Conduct the isoflurane anesthetic procedure inside a fume hood.
2. Set the gas vaporizer to initially 5% isoflurane in 30% O₂/70% N₂O in the anesthesia chamber with 1 L/min flow. Allow sufficient time for the anesthesia chamber to fill (about 5 min depending on chamber volume and gas flow rate). Ensure rapid induction (less than one minute).
3. Place the mice in the anesthesia chamber until they lose consciousness.
   NOTE: This should occur within a minute or less if the chamber is sufficiently filled.
4. Ensure proper anesthetization by the absence of painful response to pedal reflex (firm toe pinch).
5. Place the mice on the heating plate, making sure that the nose is in the nose cone.
6. Reduce the anesthesia to a maintenance level of 2% isoflurane in 30% O₂/70% N₂O and place a plastic cover above the mice to reduce the loss of heat. Apply a suitable eye ointment to prevent dryness while under anesthesia.
7. Mark the tail at a diameter of 2.5 mm using the tail mark block. Do not force the tail into the slit in the block - it must fit snugly (Supplemental Figure 1)
8. Place the tail in the saline tube for at least 5 min to ensure a warm tail vein that is optimal for intravenous (i.v.) dosing.

3. Dosing of test solution

1. Place one mouse in the tail holder with the nose in an anesthesia mask.
2. Dose the animal with the compound of interest (in this case, rFVIII) and immediately start the stopwatch (t = 0).
3. Place the mouse back on the heating plate with the tail in the saline tube. Repeat the procedure with the other mice.

4. Performing tail vein transection

1. Perform the tail vein transection exactly 5 min after dosing. Place the tail in the cutting block and turn 90° to expose the vein (Supplemental Figure 2).
2. Perform the cut on the opposite side/vein from where the test solution was dosed.
3. Draw the #11 scalpel blade through the slit of the cutting block holding the tail to create bleeding. Reset the stopwatch and return the tail immediately to the saline.

5. Observation time and challenges

1. Observe the bleeding and annotate the start and stop of the bleeding throughout 40 min; annotate it on the blood flow notation paper.
   NOTE: This visual assessment of bleeding may vary slightly due to subjectivity.
2. The primary bleeding must stop within 3 min after the cut is made. If this is not the case, disqualify the mouse, euthanize, and replace (failure to stop the primary bleed can indicate a too severe injury or lacking primary hemostasis, as in vWF KO mice).
3. If there is no bleeding at 10 min, 20 min, and 30 min post-injury, challenge the tail cut as described in steps 4-5.
4. Use a gauze swab soaked in warm saline from a separate tube kept in the water bath. Lift the tail out of the saline and softly wipe twice with the wet gauze in a distal direction over the tail cut.
5. After each challenge, immediately re-submerge the tail into the saline tube again.
6. At t=40 min, stop the blood collection by removing the tail from the saline tube.

6. Blood sampling

1. After t = 40 min, obtain blood samples from the supraorbital vein.

7. Euthanasia

1. Euthanize the mice by cervical dislocation while still under full anesthesia.

8. Treatment of samples

1. Centrifuge the 15 mL blood collection tubes with saline at 4000 x g for 5 min at room temperature.
2. Discard the supernatant from the 15 mL tubes, resuspend the pellet in 2-14 mL of erythrocytes (RBC) lysing solution, and then dilute it until it reaches a light coffee color.
3. Note the total volume (volume of blood + volume of erythrocytes (RBC) lysing solution added using the graduation marks on the tube).
4. Transfer 2 mL of the dilution to a hemoglobin tube and refrigerate it until the hemoglobin analysis.
5. Determine the blood loss by measuring the hemoglobin concentration in the saline. Measure the absorbance at 550 nm on a microplate reader (Table of Materials).
6. Convert the absorbance to nmol hemoglobin using a standard curve prepared from human hemoglobin (Table of Materials) and correct for the dilution with RBC lysing solution.

9. Statistical analyses

1. Analyze the data using appropriate software. Here GraphPad Prism software was used. Over a range of studies, the following statistical methods were found to perform well.
   NOTE: To analyze blood loss, bleeding time, exposure, platelet counts, and hematocrit; Brown-Forsythe and Welch ANOVA test was used, (as the data were continuous but without variance homogeneity of the residuals) applying Dunnett's test to adjust for multiple comparisons. A Pearson's test was used to test for correlations between bleeding time, blood loss, and doses. To determine ED₅₀ values, a four-parameter inverse log (dose) response equation was fitted to bleeding- and blood loss data. To analyze the gender effect, a two-way ANOVA test was used, applying Bonferroni correction to adjust for multiple comparisons. The significance level was defined as P < 0.05. Data are displayed as means ± SEM.

Results

To assess the applicability of the optimized model, a study was performed in F8-KO (C57BL genetic background) mice administered with a commercially available recombinant factor VIII replacement therapy (rFVIII); four different doses were tested: 1 IU/kg, 5 IU/kg, 10 IU/kg, and 20 IU/kg. Furthermore, we tested the corresponding vehicle (negative) control in F8-KO mice and wild-type (WT) group using C57BL mice as a positive control group to assess the response range in the model.

Following the o...

Discussion

This optimized method of tail vein transection (TVT) has several advantages compared to the TVT survival method. The animals are fully anesthetized for the entire study duration, which makes mouse handling easier and increases animal wellbeing. Further, unlike the TVT survival model, overnight observation is not required, and this optimized model offers the possibility of measuring blood loss and observing the exact bleeding time over 40 min. Also, longer periods of bleeding in conscious animals can cause death by exsang...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
#11 Scalpel blade	Swann-Morton	503
15 mL centrifuge tubes	Greiner Bio-One, Austria	188271
30 G needles connected to 300 µL precision (insulin) syringes for dosing	BD Micro-Fine + U-100 insulin syringe	320830
Advate	Takeda, Japan		Recombinant factor VIII replacement therapy (rFVIII)
Alcohol pads 70% ethanol	Hartmann, Soft-Zellin	999 979
Centrifuge	Omnifuge 2.0 RS, Heraus Sepatech
Cutting template (Stainless steel)	Self produced, you are welcomed to contact the authors for the exact drawings		Supplementary Figure 2: Size specifications: 20 mm x 40 mm x 10 mm (L x B x H). Groove: 3 mm depth and 3 mm width; radius 1.5 mm
Erythrocytes (RBC) lysing solution	Lysebio, ABX Diagnostics	906012
Gauze
Haematological analyser	Sysmex	CT-2000iv
Heating lamp on stand	Phillips	IR250
Heating pad with thermostat	CMA	model 150
Hemoglobin standards and controls - 8.81 mmol / l batch dependent	HemoCue, Denmark	HemoCue calibrator, 707037	Standards and controls are made from 2 different glasses of HemoCue calibrator. The value is determined against the International Reference Method for Hemoglobin (ICSH).
Isofluorane anaesthesia system complete with tubes, masks and induction box	Sigma Delta Dameca
Isoflurane	Baxter	26675-46-7
Magnifier with lights	Eschenbach
Measuring template (Aluminum)	Self produced, you are welcomed to contact the authors for the exact drawings		Supplementary Figure 1: Size specifications: 20 mm x 40 mm x 10 mm (L x B x H). Groove: 2.5 mm depth and 2.5 mm width; radius 1.25 mm
Micropipettes + tips	Finnpipette
Photometer	Molecular Devices Corporation, CA, USA	SpectraMax 340 photometer
Prism Software	GraphPad, San Diego, CA, USA	Version 9.0.1
Saline 0.9% NaCl	Fresenius Kabi, Sweden	883264
Special tail marker block for TVT tail cut
Tail holder
Vacuum liquid suction	Vacusafe comfort, IBS
Waterbath and thermostat	TYP 3/8 Julabo