In Vivo Measurement of Redox-Regulated TG2 Activity

Abstract

Transglutaminase 2 (TG2) is a ubiquitous mammalian enzyme that is implicated in a variety of physiological processes and human diseases. Normally, extracellular TG2 is catalytically dormant due to formation of an allosteric disulfide bond between Cys370–371 of the enzyme. In this protocol, we describe a method to reduce this disulfide bond in living mice and to monitor the resulting in vivo TG2 activity. Briefly, exogenous thioredoxin-1 protein (TRX) is prepared and administered as a specific, physiologically relevant reductant of the Cys370–371 disulfide along with the small molecule 5-biotinamidopentylamine (5-BP) as a TG2 activity probe. Tissue cryosections then analyzed by immunohistochemistry to ascertain the extent of 5-BP incorporation, which serves as a record of the redox state of TG2 in vivo. This protocol focuses on the modulation and measurement of TG2 in the small intestine, but we encourage investigators to evaluate it in their organ(s) of interest.

1. Introduction

Transglutaminase 2 (TG2) is a ubiquitous mammalian enzyme that catalyzes the transamidation or deamidation of its substrates (1). In virtually all tissue types, TG2 is expressed intracellularly and is also exported to the extracellular matrix (ECM) by a non-canonical secretory mechanism (2, 3). While the enzymatic chemistry of TG2 has been extensively characterized, its bona fide physiological functions have largely remained elusive. Aberrant TG2 activity has been implicated in several pathophysiological conditions in humans, including celiac disease, cancers, and fibrotic disorders (4–6). Our lack of understanding of TG2 biology and its potential involvement in many diseases creates a demand for a reliable method of modulating and measuring TG2 activity in the tissues of intact, living animal models.

Under basal physiological conditions, both intracellular and extracellular TG2 are catalytically inactive due to tight regulation of its activity at the post-translational level (7). Enzymatic activity of TG2 requires the presence of Ca²⁺ and is inhibited upon binding of GTP or GDP (8). The relatively low abundance of Ca²⁺ and the presence of GTP/GDP inside cells explains the inactivity of the intracellular pool of TG2. However, despite relatively high concentrations of Ca²⁺ and scarcity of guanine nucleotides in the ECM, studies in vitro and in vivo indicate that extracellular TG2 is also catalytically inactive under normal conditions (7, 9, 10). During the past decade, our laboratory has uncovered an unusual, reversible redox switch that is now believed to be the principal mechanism for modulating the activity of TG2 in the extracellular environment (10–12). This redox control is mediated through an allosteric disulfide bond between Cys370 and Cys371 of TG2, which is located distal to the active site Cys residue (Cys277) (11). Formation of this disulfide bond completely abolishes catalytic activity of the enzyme. The presence of the Cys370–371 disulfide in the oxidizing environment of the ECM explains why extracellular TG2 is catalytically inactive. In vitro, a redox protein cofactor, thioredoxin-1 (TRX) is capable of reducing this regulatory disulfide bond with high specificity, rendering TG2 active (10). On the other hand, endoplasmic reticulum–resident protein 57 (ERp57) is capable of promoting TG2 oxidation and inactivation (12). The fact that two distinct extracellular partner proteins found modulate the activity of TG2 provides further evidence for the physiological relevance of extracellular redox regulation of TG2.

Recently, we sought to determine whether reduction of the Cys370–371 disulfide bond of TG2 is sufficient to activate the enzyme in the small intestine in vivo. To address this question, we needed a method to reduce TG2 and to monitor its enzymatic activity in the context of living mice (13). To this end, we developed a protocol for administration of recombinant TRX protein as a specific reductant of TG2, and adapted a procedure to measure in vivo TG2 activity via incorporation of the small molecule 5-biotinamidopentylamine (5-BP) into extracellular matrix proteins (7, 14). Intravenous administration of recombinant TRX in C57BL/6 mice leads to a rapid rise in intestinal transglutaminase activity in wild-type but not TG2-null mice in a manner that can be inhibited by a small molecule inhibitor of TG2. Here, we describe the preparation of the pharmacological tools needed to carry out this protocol, as well as our recommendations for their administration to mice and the subsequent data analysis. We note that we have focused on measurement of intestinal TG2 activity, but we encourage investigators to adapt and evaluate this protocol for use in their organ(s) of interest.

2. Materials

All solutions are prepared in ultrapure water (ddH₂O) obtained from a Milli-Q® filtration system. Materials for injection are prepared fresh, sterile filtered using 0.22 μM syringe-driven PVDF filters, and used within 1.5 hours of preparation.

2.1. Preparation of recombinant human thioredoxin-1

1. E. coli (Rosetta 2) transformed with an expression plasmid containing a T7 promoter and encoding His-tagged full-length human thioredoxin-1 (GenBank™ Accession NM_003329) (see Note 1).

2. LB media.

3. Centrifuge capable of light (5000 × g) and hard (30,000 × g) spins.

4. Lysis buffer: 50 mM phosphate, 10 mM imidazole, 150 mM NaCl, pH 7.6.

5. 250 mM isopropyl β-D-1-thiogalactopyranoside (IPTG).

6. Nickel-nitrilotriacetic acid (Ni-NTA) resin.

7. Glass column for immobilized metal affinity chromatography (IMAC).

8. Elution buffer: 50 mM phosphate, 250 mM imidazole, 150 mM NaCl, pH 7.6.

9. 1 M dithiothreitol (DTT) in ddH₂O (see Note 2).

10. Protein polyacrylamide gel.

11. 2x Laemmli sample buffer: 125 mM Tris HCl, 4% (v/v) sodium dodoceyl sulfate, 20% (v/v) glycerol, 10% (v/v) 2-mercaptoethanol, 0.004% (v/v) bromophenol blue, pH 6.8.

12. Fast protein liquid chromatography (FPLC) system.

13. 5 mL HiTrap Q HP anion exchange column (GE Healthcare).

14. FPLC Buffer A: 20 mM Tris, 1 mM EDTA, pH 7.2.

15. FPLC Buffer B: 20 mM Tris, 1 mM EDTA, 1 M NaCl, pH 7.2.

16. 3 kDa MWCO Amicon Ultra-15 centrifugal filter units.

17. Glycerol.

2.2. In vivo mouse studies

1. C57BL/6J mice (Charles River Laboratories, Boston, MA) and relevant knock-out/knock-in mice on C57BL/6J background (see Note 3).

2. 5-biotinamidopentylamine (5-BP) (see Note 4).

3. ERW1041E, transglutaminase 2 inhibitor (see Note 4).

4. ERW1041E injection vehicle: 2.5% (2-hydroxypropyl)-β-cyclodextrin, 2.0% Tween-80 in sterile PBS, pH 7.4 (see Note 5).

5. Recombinant human thioredoxin (TRX), made according to Section 2.1.

6. 3 kDa MWCO Amicon Ultra-15 centrifugal filter units (Millipore Sigma).

7. 1 M dithiothreitol (DTT).

8. PD-10 desalting column (GE Healthcare).

9. 0.22 μM polyvinylidenefluoride (PVDF) syringe-driven sterile filters.

10. Syringes and needles appropriate for intraperitoneal and intravenous injections in mice (e.g., 1 mL syringes with attached 27-gauge needles).

2.3. Harvesting and preparation of mouse tissues

1. Carbon dioxide euthanization chamber.

2. Instruments for dissection of mice: scissors, forceps, scalpels, and any other desired tools.

3. Optimal cutting temperature (OCT) compound and biopsy cryomolds (“OCT blocks”).

4. Dry ice, isopentane, and open-faced Dewar flask.

5. Large, −80 °C-compatible freezer bags.

2.4. Immunohistochemistry of mouse tissues

1. Cryotome.

2. SuperFrost Plus™ glass slides (ThermoFisher).

3. PBS, pH 7.4.

4. 4% (w/v) paraformaldehyde.

5. Plastic (opaque) microscope slide chambers.

6. Vectashield + DAPI mounting medium (Vectorlabs H-1200).

7. Glass cover slips.

8. Rabbit anti-TG2 antibody (e.g., Abcam ab109200) (see Note 6).

9. Secondary antibody to anti-TG2 (see Note 6).

10. Fluorophore-conjugated streptavidin (see Note 6).

11. Wash buffer: PBS, pH 7.4 supplemented with 0.1% Tween-20 (PBS-T).

12. Blocking buffer: 5% bovine serum albumin (BSA) in PBS-T.

2.5. Collection and analysis of microscopy data

1. Fluorescence microscope.

3. Methods

In this section, a workflow to activate and then probe the activity of extracellular TG2 in living mice is presented. Briefly, recombinant human thioredoxin-1 (TRX) is expressed in E.coli and purified. TRX is then injected intravenously to activate extracellular TG2. 5-BP, a chemical probe of TG2 activity, is also administered, and the specificity of probe incorporation is confirmed with the use of ERW1041E (an active-site directed TG2 inhibitor) (Fig 1).

Figure 1.

Molecular tools to investigate the biology of extracellular TG2. (A) 5-BP is an activity-based probe for TG2. Active TG2 catalyzes the transamidation of 5-BP and extracellular proteins, and the amount of 5-BP incorporation to the ECM can be used to quantify TG2 activity. (B) TRX reduces the allosteric disulfide redox switch (Cys370 and Cys371), activating TG2. ERW1041E inactivates TG2 by covalent attachment to its active site cysteine (Cys277).

After treatment with the respective control/experimental conditions, mice are sacrificed, and tissues of interest are harvested and frozen. The tissues are then sectioned, stained immunohistochemically, and analyzed for the incorporation of 5-BP as the readout for TG2 activity. It is important to note that although TG2 is a ubiquitously expressed protein, we have only conclusively shown that TRX-mediated TG2 activation occurs in the small intestine. This may be due to our limited pharmacokinetic understanding of TRX and/or 5-BP, or may be an indication of more complex regulation of TG2 in other organs (13). Nonetheless, we anticipate that this protocol will be a useful starting point for investigators who seek to modulate the redox state and probe the activity of TG2 in their organ(s) of interest.

3.1. Preparation of recombinant human thioredoxin-1

1. Prepare a seed culture of Rosetta 2 cells containing the TRX expression plasmid. Grow the seed culture in LB media supplemented with the appropriate amount of antibiotics for the vector used. Grow the seed culture overnight.

2. Prepare desired amount of LB media and supplement it with the appropriate amount of antibiotics. Inoculate the media with seed culture, and shake the cultures at 37 °C, 200 RPM (see Note 7).

3. Check the OD₆₀₀ at 2 hours post inoculation. After this point, check the OD frequently. When the OD₆₀₀ reaches approximately 0.5, change the temperature to 18 °C. When the OD₆₀₀ is at 0.6, induce the cultures by adding 1 mL of 250 mM IPTG per liter of culture (final concentration = 250 μM IPTG). Shake the cultures overnight at 18 °C, 180 RPM for 14–18 hours.

4. Pellet cells by centrifugation at 5000 × g for 10 min at 4 °C. The cell pellets from 1 L culture should be placed in a 50 mL conical vial and resuspended in 40 mL of buffer. Lyse cells via sonication (see Note 8).

5. Clear the cellular debris by centrifugation of the lysate at 30,000 × g for 60 min at 4 °C.

6. During this centrifugation step, prepare the Ni-NTA resin by equilibrating it with the lysis buffer (see Note 9). This can be done by centrifuging the resin slurry at 2000 × g for 1 minute, discarding the supernatant, and resuspending the resin in lysis buffer a total of three times.

7. Incubate the supernatant from the centrifugation with the equilibrated Ni-NTA resin for 60 min at 4 °C with continuous rotation or shaking.

8. Apply the lysis mixture to a column and discard the flow through. Wash the resin bed three times, with 2 column volumes of lysis buffer per wash.

9. Elute the bound protein off the column with four column volumes of elution buffer. Dilute the eluent with ddH₂O to 50 mL, add DTT to a final concentration of 10 mM, and let the solution stand for 10–20 minutes at 4 °C (see Note 10).

10. Purify the Ni-NTA eluate using a HiTrap Q HP anion exchange column with TRX FPLC Buffers A and B. A typical gradient is 0–30% FPLC B over 15 column volumes at a flow rate of 3 mL/min and 2 mL per fraction. TRX should elute at approximately 10% FPLC B. Take the elution fraction containing TRX (confirm with an SDS-PAGE gel), and concentrate the protein to the desired concentration using a 3 kDa MWCO Amicon centrifugal filter. (We typically concentrate TRX to 100 mg/mL for use in mouse studies.) This step should also be done at 4 °C.

11. TRX concentrations should be measured using ultraviolet absorbance at 280nm and the predicted extinction coefficient of 7570 M⁻¹cm⁻¹ (see Note 11). Add glycerol to a final concentration of 20% (v/v), measure the concentration of TRX one more time, and aliquot the glycerol-containing sample as desired. Flash freeze the aliquots with liquid nitrogen and store at −80 °C.

12. Prior to using the recombinant protein in the downstream mouse studies, test the activity of TRX in vitro (16).

3.2. In vivo mouse studies

1. Select relevant mice for the experiment to be conducted and begin by recording their weights. Use these weights to calculate the dosages of each agent necessary for each mouse (see Note 12). Three to six mice are typically sufficient per experimental condition.

2. Prepare 5-BP by dissolving it in PBS to a final concentration of 20 mg/mL. Filter the resulting solution through a 0.22 μm syringe-driven PVDF filter.

3. Dissolve ERW1041E in DMSO at a concentration of 100 mg/mL. Dilute this tenfold to 10 mg/mL in ERW1041E injection vehicle. Filter the resulting solution through a 0.22 μm syringe-driven PVDF filter.

4. Prepare the TRX to be injected by concentrating the expressed TRX using a 3 kDa MWCO Amicon centrifugal filter to 100 mg/mL (see Note 13). Add a tenfold molar excess of DTT by adding a concentrated stock dissolved in ddH₂O and let stand at room temperature for one hour. Following this, buffer exchange the reduced TRX into PBS pH 7.4 using a PD-10 column. Filter the resulting solution through a 0.22 μm syringe-driven PVDF filter and determine the final concentration of TRX. Record this concentration for determination of the precise volumes to inject. Use this reduced TRX solution within 1.5 hours of preparation, and keep on ice until use.

5. Begin the experiment. Injections should be staggered by about 5 minutes such that there is ample time between mice, as this will be important downstream during the harvesting of the desired tissues. A possible scheme for probing the TRX-mediated activation and subsequent ERW1041E-mediated inhibition of TG2 in wildtype mice is shown in Table 1 (see Note 14).

Table 1.

An example of a mouse study involving the activation and inhibition of murine extracellular TG2. Due to the short half-lives of 5-BP and ERW1041E in vivo, we have found that multiple injections are needed to obtain clear results. The group of mice receiving 5-BP should display minimal incorporation of the probe, as TG2 activity should be virtually absent in the absence of an activator. The 5-BP + TRX group should display significantly enhanced TG2 incorporation, whereas the 5BP + TRX + ERW1041E should look similar to the 5-BP only group. The three groups listed in this table provide a minimal set of positive and negative control conditions for TG2 activity, but if desired, other groups, such as 5-BP + TRX administered to TG2−/− mice, can also be included.

Experimental condition	Mouse	5-BP (IP) Time (min)	ERW1041E (IP) Time (min)	TRX (IV) Time (min)	Sacrifice (CO2) Time (min)
5-BP only	1	0, 30, 60, 120	Vehicle	Vehicle	150
	2	5, 35, 65, 125	Vehicle	Vehicle	155
	3	10, 40, 70, 130	Vehicle	Vehicle	160
5-BP, TRX	4	15, 45, 75, 105	Vehicle	45	165
	5	20, 50, 80, 140	Vehicle	50	170
	6	25, 55, 85, 145	Vehicle	55	175
5-BP, TRX, ERW1041E	7	30, 60, 90, 150	30, 90, 150	60	180
	8	35, 65, 95, 155	35, 95, 155	65	185
	9	40, 70, 100, 160	40, 100, 160	70	190

3.3. Harvesting and preparation of mouse tissues

The following steps should be done quickly and efficiently. Work in a team of three or four. One person euthanizes the mice, one person dissects the mice, and one person preserves the organs in OCT blocks. It is best to pre-label all OCT blocks before starting the experiment.

1. Once a mouse has been euthanized via carbon dioxide, perform an appropriate secondary method of sacrifice.

2. Cut the mouse along the cranial-caudal axis with a vertical incision. Proceed to dissect the mouse according to which organs must be isolated.

3. Meanwhile, fill each OCT block to half capacity with OCT compound. It is important to avoid the introduction of bubbles into the OCT blocks, as the presence of bubbles will greatly affect the quality of the tissue sections.

4. Add one organ per OCT block and carefully fill the rest of the OCT block with OCT compound. After doing so, immerse the OCT block in a dry ice/isopentane bath to flash freeze the organ. At this point, the organs are ready to either be cut immediately, or to be stored at −80 °C in deep freezer-compatible bags.

3.4. Immunohistochemistry of mouse tissues

The organs are now ready to be analyzed by immunohistochemistry.

1. Equilibrate the cryotome to −20 °C and section the OCT blocks in the desired manner, taking great care to be gentle with the mounted blocks. Typically, 10 μm sections are sufficient for the downstream immunohistochemical workup. Arrange four cut sections per slide in a square pattern for four technical replicates per mouse per condition. Carefully transfer the cut sections onto Superfrost Plus™ microscope slides.

2. Fill a plastic microscope slide chamber with 4% (w/v) paraformaldehyde such that inserting the slides in the chamber results in total coverage with solution. Add slides to the slide chamber and incubate for 15 minutes at room temperature.

3. After fixation, wash the tissue sections by adding PBS to the slide chamber and carefully decanting. Repeat this four times for a total of five washes.

4. Add blocking buffer to a new opaque microscope slide chamber and place slides inside. Block slides overnight at 4 °C.

5. Add anti-TG2 antibody to blocking buffer at the desired concentration. Fill a fresh plastic microscope slide chamber with the antibody solution, place the slides inside, and incubate overnight at 4 °C.

6. Wash 3x with PBS.

7. Repeat Step 4, but with the secondary antibody to anti-TG2 (for TG2 localization) and fluorophore-conjugated streptavidin (for TG2 activity through 5-BP incorporation).

8. Wash with wash buffer. Repeat this three for a total of four washes. For the final wash, to remove the PBS-T, tilt the slide carefully and allow the capillary action of an absorbent surface to withdraw the PBS-T.

9. Mount cover slips to the slides using Vectashield + DAPI mounting medium.

10. To anchor the cover slip to the microscope slide, use clear nail polish to seal the sides of the cover slip. After drying, the slides can be stored in the dark at 4 °C until imaging.

3.5. Collection and analysis of microscopy data

1. Image the slides using a fluorescence microscope. Input the wavelengths to be measured according to the primary/secondary antibodies used during the immunohistochemical workup of the microscope slides, and adjust the gain settings so that the signal for each fluorophore is strong, but not saturating. It is essential that the data for all of the slides from a given experiment are collected under identical acquisition settings (e.g., laser power, detector gain, and pinhole size). Since the goal is to semi-quantitatively compare the relative amounts of 5-BP incorporation between different experimental conditions, any changes in the microscope settings will render the data analysis invalid.

2. For each slide to be analyzed, take 3–4 representative images.

3. Quantify levels of 5-BP incorporation by obtaining the mean signal intensity for 5-BP and normalize to the total tissue area (see Note 15). The total tissue area can be determined by a threshold algorithm.

4. Perform appropriate statistical analyses.

4. Notes

1. As described in (10), our lab has prepared this construct by amplifying the TRX gene from cDNA with primers 5′AAAAAACATATGAAAATCCATCACCATCACCATCACATGGTG AAGCAGATCGAGAG-3′ (forward primer) and 5′-TTTTTTCTCGAGTTAGACTAATTCATT AATGG-3′ (reverse primer), digesting the amplimer with NdeI and XhoI, and inserting the cut amplimer between the NdeI and XhoI sites of the vector pQE-T7 (Qiagen, Valencia, CA).

2. DTT solutions must be freshly prepared.

3. The mice to be used for the experiment should be age- and weight-matched as much as possible to reduce variability. For custom mouse lines, the control mice (e.g., TG2-null mice) are ideally of the same background.

4. We typically synthesize 5-BP and ERW1041E, as described in refs. 14 and 15, but they are also commercially available from ThermoFisher and Millipore-Sigma, respectively.

5. The appropriate injection vehicle will vary among the agents to be injected. For example, hydrophobic compounds may require a small amount of organic solvent for their solubilization, e.g., DMSO, and/or the use of a solubility enhancer such as β-cyclodextrin. In the case of the TG2 inhibitor ERW1041E, the preferred injection vehicle is 2.5% (2-hydroxypropyl)-β-cyclodextrin, 2.0% Tween-80 in sterile PBS, pH 7.4. Proteins such as TRX can be prepared in an aqueous, nontoxic buffer (e.g., PBS pH 7.4).

6. The specificity of the antibodies used for the detection of relevant proteins should be tested. This can be accomplished by standard protocols, such as comparing staining with isotype control antibodies and/or using tissues that are null for the protein of interest (e.g., use tissues collected from TG2-null mice to confirm the specificity of anti-TG2 antibodies). In our experience, many commercial antibodies against TG2 have poor reactivity with murine TG2, and the alternative is raising a custom antibody using recombinant murine TG2 as the immunogen, as we have described (13). For antibody detection, one can choose any appropriate combination of fluorophore-conjugated second antibodies. We have used Streptavidin Alexa Fluor 647 (Invitrogen S32357) for 5-BP detection and goat anti-rabbit Alexa Fluor 488 for detection of TG2 protein (ThermoFisher A11008).

7. The typical titer of TRX expressed and purified through this protocol is roughly 100 mg/liter of culture. As we use a dose of 500 mg/kg in mice, scale the total volume of expression culture depending on the scale of the mouse experiment.

8. Prior to sonication, it is helpful to break up the cells with a spatula and vortex the mixture. After the cells are completely lysed, the mixture will go from a cream, light brown color to a darker shade of brown.

9. Check the vendor for the amount of resin needed. A typical amount of resin to use is 1 mL resin for every 1 L expression culture.

10. The entirety of this step (centrifugation, IMAC column, and reduction) should be performed at 4 °C.

11. Ultraviolet absorbance should be used for measuring the concentration of TRX, as it is too small for the use of assays such as the Bradford assay. The given value for the extinction coefficient of TRX was calculated using the ExPASy ProtParam tool (14).

12. The maximum volume allowable for administration of injections must be taken into account when designing the experiment and preparing the reagents (17). We typically inject no more than 150 μL for intravenous injections of TRX and 100 μL for intraperitoneal injections for 5-BP and ERW1041E.

13. Concentrating the TRX sample to 100 mg/mL can be done during Step 10 in Section 3.1.

14. If appropriate activators and detection agents are available, this scheme may be adapted to measurement of enzymes other than TG2. The timing and dosage for each agent administered will be different depending on their pharmacokinetics. Also, the injection method (intraperitoneal, intravenous, etc.) must be considered depending on the nature of the detection reagent and the desired tissues that are to be harvested and investigated.

15. Other normalization schemes may be used for the measurement of TG2 activity via 5-BP incorporation. For example, the DAPI mean signal intensity may be used to normalize to cell number per image field, or TG2 localization mean signal intensity may be used to normalize to total TG2 per image field (18).