In this paper we provide details for the methods of production and quality control of an IL4-10 recombinant fusion protein. We also show how to test the effectiveness of this protein to resolve pain in a mouse model of inflammatory pain.
Chronic pain is difficult to treat and new approaches to resolve persistent pain are urgently needed. Anti-inflammatory cytokines are promising candidates for treating debilitating pain conditions due to their capacity to regulate aberrant neuro-immune interactions. However, physiologically they work in a network of various cytokines, and therefore their therapeutic effect may not be optimal when used as stand-alone drugs. To overcome this limitation, we developed a fusion protein of the anti-inflammatory cytokines IL4 and IL10. Here, we describe the methods for production and quality control of IL4-10 recombinant fusion protein and we test the effectiveness of the IL4-10 fusion protein to resolve pain in a mouse model of persistent inflammatory pain.
Chronic pain remains one of the most debilitating and under-treated medical problems of the 21st century, affecting >20% of the adult population1,2. However, treatments to provide relief from chronic pain are often ineffective or must be discontinued due to severe side effects3. Importantly, currently available drugs only provide symptomatic relief, but do not significantly modify or cure chronic pain. Although chronic pain appears to be a neurological disorder, evidence suggests involvement of the immune system in chronic pain development4,5. Moreover, immune-based approaches to treat pain are emerging. For example, anti-inflammatory cytokines inhibit pain in several models of chronic pain6,7,8. However, anti-inflammatory cytokines have a short half-life, reducing their potential pain-inhibiting effects. Moreover, anti-inflammatory cytokines work most optimally in concert with each other. To overcome these limitations, we recently fused the anti-inflammatory cytokines interleukin-4 (IL4) and interleukin-10 (IL10) into one molecule. The IL4-10 fusion protein shows superior efficacy in inhibiting chronic inflammatory and neuropathic pain compared to the individual cytokines9. Here we describe how such fusion protein is produced, purified, and how its quality is controlled.
IL4-10 fusion protein is produced in human cells by transient transfection of HEK293-F cells with a pUPE expression vector carrying the cDNA sequence coding the IL4-10 fusion protein. HEK293-F cells are chosen to allow for post-translational modification of the protein, something that does not occur in bacterial expression systems. To optimize glycan capping with sialic acid, cDNA coding beta-galactoside-2, 3-sialyl-transferase is incorporated in the vector as a second transgene. The fusion protein is purified using affinity protein purification of the culture supernatant because it is more powerful than purification by other methods e.g. size-exclusion or ion exchange chromatography10,11. To purify the IL4-10 fusion protein, we used in-house made monoclonal antibodies against IL4. Evaluation of the purity and bioactivity of purified IL4-10 fusion protein is performed as a part of quality control. The purity of produced batches is evaluated by Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDS-PAGE) and High Pressure Size Exclusion Chromatography (HP-SEC). The bioactivity of the IL4-10 fusion protein is evaluated by measuring its capacity to inhibit lipopolysaccharide (LPS)-induced tumor necrosis factor-alpha (TNFα) production in whole blood cultures, and comparing it to the combination of the individual cytokines.
Finally, in order to test the capacity of IL4-10 fusion proteins to inhibit chronic pain, we describe how the fusion protein can be tested as analgesic in widely used mouse models of persistent inflammatory pain12,13,14. Here we describe the methods of an inflammatory pain model. However, it is important to note that other pain models can be used (e.g., neuropathic pain models), depending on the research questions that need to be answered. To assess pain in these models, it is important to use a variety of behavioral measures that include evoked and non-evoked pain measures. Here, we described the methods of assessment for changes in mechanically and thermally evoked behavioral responses. Mechanical sensitivity to innocuous stimuli is assessed using the Von Frey test, whilst thermal sensitivity is assessed using the Hargreaves test. Importantly, non-evoked hyperalgesia/allodynia is measured using the Dynamic Weight Bearing test. These measures are widely accepted as pain measures and yield important information on pain thresholds and potential pain experienced by the animals15,16,17. Other measures to assess non-evoked pain (e.g., stimulus independent), such as the conditioned place preference test, may be valuable18. To assess the potential of the drug to inhibit pain, we performed intrathecal administration of the fusion protein, as with this route of administration less protein dose is needed to reach pain-related areas and avoid systemic (side-) effects19,20.
All animal experiments were performed in accordance with international guidelines and prior approval from the local experimental ethical committee. Whole blood was obtained from the Mini Donor Service (Mini Donor Dienst, MDD) at the University Medical Center Utrecht (UMCU) in the Netherlands. The MDD has received positive approval from the Medical Ethics Committee of the UMCU (Medisch Ethische Toetscommissie) for the protocol number 07-125/C.
1. Protein Production and Characterization
2. Mouse Model for Persistent Inflammatory Pain
NOTE: It is important to use both male and female mice (C57Bl/6) because this will enable the identification of sex-dependent effects. In general, the mice used are between 8 – 16 weeks old. To determine the number of animals required, power calculations should be performed. Web-based tools to perform such calculations are readily found on the internet (e.g., http://www.powerandsamplesize.com; http://www.sample-size.net).
3. Pain Measurements
Note: Ensure that researchers performing the behavioral experiments are blind to the mouse treatment. It is important to acclimatize the animals to the testing environment before performing any measurement. Preferably, the week before starting the experiments, animals are placed 1 – 2 times for 15 – 30 min in each of the different devices used for performing the tests. When handling males and females in the same experiment, evaluate males independently of females. Clean cages and surfaces extensively between the different groups to avoid potential unwanted effects on behavior of the different sexes. Measure baseline withdrawal latencies or mechanical thresholds two to three times to accurately determine baseline thresholds and identify whether these are stable before intraplantar injection of any compound.
4. Intrathecal Injection of IL4-10 fusion protein for analgesia
NOTE: Confirm that the ventilation to the room is open to ensure proper air flow. To test the efficacy of the IL4-10 fusion protein (1 µg/mouse; 5 µL total injection volume) it should be tested against vehicle injections and injections of the combination of the individual cytokines (IL4 + IL10, 0.5 µg + 0.5 µg/mouse, 5 µL total injection volume).
A representative picture of an SDS-PAGE gel containing different fractions obtained during affinity chromatography purification is shown in Figure 1A. In the load (L) and flow through (FT) fractions, all proteins present in the HEK293 supernatant are observed. No protein is observed in the wash (W) fraction. In the elution (E) fraction, two bands of 35 and 37 kDa are observed corresponding to two different glycoforms of IL4-10 fusion protein (arrows). In Figure 1B, a representative potency assay in a whole blood assay of two different IL4-10 fusion protein batches is shown. A dose-dependent inhibition of LPS-induced TNFα release is observed, showing comparable bioactivity for the two IL4-10 fusion protein batches.
Next, an example of test results of representative experiments in which the IL4-10 fusion protein is tested to inhibit persistent inflammatory pain is shown (Figure 2). Persistent inflammatory pain was induced by intraplantar injection of 2% carrageenan. Mechanical (Figure 2A) as well as thermal (Figure 2B) hyperalgesia was followed over time. At day 6 after induction of inflammatory pain, mice were injected intrathecally with the IL4-10 fusion protein or vehicle as a control. IL4-10 significantly resolved both mechanical (Figure 2A) and thermal (Figure 2B) hyperalgesia, and reached its maximal effect after 24 h after injection. Ideally, the efficacy of the fusion protein in pain inhibition should be also tested against equimolar concentrations of the individual cytokines to test whether its capacity to inhibit pain is better than the sum of the individual cytokines. These data are not displayed here, but we refer to a recent publication that displays this data9.
The effectiveness of the IL4-10 fusion protein was recently tested in the CFA-induced inflammatory pain model9. In this experiment, multiple injections of the IL4-10 fusion protein were administered intrathecally. The multiple injections completely resolved CFA-induced persistent inflammatory hyperalgesia, measured using von Frey and Hargreaves tests, demonstrating the potential of the IL4-10 fusion protein as an analgesic treatment9. Here we show that intraplantar CFA injection also induced a reduction in weight bearing of the affected paw (Figure 3A-B). Intrathecal injection of IL4-10 fusion protein at day 7 after intraplantar CFA injection attenuated the reduction in weight bearing of the affected paw compared to vehicle-injected mice 2 days after IL4-10 fusion protein administration (Figure 3B).
Figure 1: Purification and characterization of IL4-10 fusion protein. (A) Coomassie-stained SDS-PAGE analysis of the affinity chromatography fractions: L: load, FT: flow through, W: wash, E: elution. (B) Functional assay: IL4-10 fusion protein dose dependently inhibits TNFα production in LPS stimulated whole blood culture (expressed as percent inhibition compared to cultures stimulated with LPS alone). Data is represented as mean ± SD. Please click here to view a larger version of this figure.
Figure 2: Inhibition of inflammatory hyperalgesia by the IL4-10 fusion protein. Inflammatory pain was induced by an intraplantar injection of 20 µL of 2% carrageenan (CAR). Six days after intraplantar injection mice received an intrathecal injection (arrow) of 1 µg IL4-10 fusion protein or vehicle (PBS). (A) Mechanical hypersensitivity (Log 50% threshold (g)) was measured over time using Von Frey test and (B) thermal sensitivity (latency (s)) was measured using Hargreaves test. Data is represented as mean ± SEM. Asterisks represent significant differences analyzed by two-way ANOVA, between vehicle and IL4-10 injected animals. **, *** = p <0.01 and p <0.001, respectively. Please click here to view a larger version of this figure.
Figure 3: Inflammation-induced changes in weight bearing. Inflammatory pain was induced by an intraplantar injection of 20 µL of Complete Freud's Adjuvant (CFA). Weight bearing on each paw was measured using the dynamic weight bearing device. Weight bearing in grams of the affected paw (ipsilateral hind paw) and the total weight bearing of the non-affected paws (front paws and contralateral paw) is displayed, as well as the ratio of weight bearing of the ipsilateral paw compared to the other 3 paws. (A) Course of weight bearing during CFA-induced inflammation (n = 10). (B) At day 7 after intraplantar CFA, mice received an intrathecal injection of 1 µg IL4-10 fusion protein or vehicle (PBS). Weight bearing was measured at 0 and 7 days after intraplantar CFA, and 2 days after administration of IL4-10 fusion protein. Weight bearing of the ipsilateral paw compared to the other 3 paws is represented (n = 3 – 4). Data is represented as mean ± SEM. *** = p <0.001; ** = p <0.01; * = p <0.05 compared to baseline (day 0). # = p <0.05 compared to vehicle treated animals. Please click here to view a larger version of this figure.
This manuscript describes methods for the production and characterization of a recombinant IL4-10 fusion protein, and methods to test its efficacy in inhibiting inflammatory hyperalgesia in mouse models of persistent inflammatory pain. The production and purification of the IL4-10 fusion protein is performed on a small scale. HEK293 cells are selected as an expression system for protein production because they enable post-translational modifications that cannot be achieved in prokaryotic expression systems. Post-translational modifications are relevant for protein function, and the absence of such modifications potentially could affect the therapeutic efficacy of the IL4-10 fusion protein. The protein is purified from culture supernatant by affinity chromatography. We chose to use in-house made monoclonal antibody against IL4 for affinity purification and not to produce an IL4-10 fusion protein containing a tag to allow for purification with an affinity tag system. The reason for this was to avoid possibility of tag-related changes in protein folding and function. Nevertheless, if a highly selective monoclonal antibody is lacking, a protein with a tag needs to be developed, e.g. a his tag. In that case, it will be important to determine whether the C- or N-terminus of the protein is best suited for the tag to avoid affecting protein function.
From 1 L of HEK293 supernatant, approximately 3 mg of purified IL4-10 fusion protein is obtained, but this may vary from batch to batch. The column for affinity purification can be used for several purification cycles before discarding. The most sensitive step in the purification procedure is the elution of IL4-10 fusion protein from the column by low pH, because the low pH may induce irreversible protein denaturation and protein precipitation. Thus, the maintenance of native protein structure and the absence of protein aggregates is essential for good quality batches. To that end, evaluation of the bioactivity of the fusion protein needs to be tested in vitro before and after purification with different elution buffers. After purification of the IL4-10 fusion protein, the bioactivity was not reduced compared to non-purified protein. Moreover, aggregate formation was absent when the elution step was performed with 0.1 M glycine buffer (pH = 2.5), and the pH of the elution fractions was immediately neutralized with 1 M Tris buffer (pH = 9).
The concentration of purified IL4-10 fusion protein is estimated on the basis of human IL10 ELISA results. BCA protein assay is used to confirm the protein concentrations. Recovery of the protein after purification is evaluated based on concentrations measured in load, flow through, wash and elution fractions by IL10 ELISA. The BCA protein concentration determination of IL4-10 fusion protein can be evaluated only in dialyzed elution fractions. Non-dialyzed elution fractions contain imidazole, a compound that affects the BCA measurement. Non-purified fractions contain other proteins from the HEK293 culture medium.
The determination of the bioactivity of the IL4-10 fusion protein is performed in a whole blood assay. The bioactivity of IL4-10 fusion protein should be evaluated in the whole blood of several donors to determine whether functional activity is retained. Several donors are required because donor to donor variance exists in the capacity of IL10 and IL4 to inhibit LPS-induced TNFα release. Moreover, expression of IL4R and IL10R may differ between donors, affecting the capacity of IL4-10 fusion protein to inhibit LPS-induced TNFα production.
In the described methods, we detailed 2 mouse models of persistent inflammatory pain to evaluate the efficacy of IL4-10 fusion protein to resolve pain in vivo. Nevertheless, for full evaluation of the analgesic potential of recombinant fusion proteins, the molecules should be tested in other models of chronic pain as well. These models include, but are not limited to, models of nerve-injury and chemotherapy-induced neuropathic pain9,22. Although the von Frey test and Hargreaves test are commonly used for testing analgesic drugs, it has become clear that an additional test (e.g. dynamic weight bearing test) should be included to assess the effects on non-evoked pain behaviorsand also with the operant pain assay (e.g. conditioned place preference (CPP) that detect non-evoked pain behaviors17,18). To perform CPP, the onset of the pain-inhibiting effects of the fusion protein should be taken into account, because conditioning of the animals by pain relief requires quick acting analgesics (which exert effects in less than 15 min). Nevertheless, if the test compound has a slow analgesic onset, but is likely to induce long-lasting pain inhibition (more than 4 days), CPP can still be used to test whether the compound inhibited pain. In this case, compound-induced loss of conditioning with fast acting analgesics needs to be assessed.
Intrathecal injections are selected as a route of administration, as the compound will reach pain relevant areas such as the dorsal root ganglia and spinal cord. Notably, intrathecal delivery of analgesic drugs has become common practice as a treatment of patients with chronic pain, as this approach reduces the dose of analgesic drug needed and decreases toxicity and systemic side effects. Nevertheless, this administration route has some limitations; it requires well-trained personnel and, due to the small subarachnoid space, the volume of injection should be limited to 5 µL in mice, requiring highly concentrated solutions of the fusion protein. The technique of intrathecal injections requires training. Such training can be performed on non-recovering anesthetized animals with injection of a dye to verify correct intrathecal injection.
The authors have nothing to disclose.
Part of this work has been funded by a Utrecht University Life Sciences grant
FreeStyle 293-F cells | Invitrogen | R790-07 | Human embryonic kidney cells |
GIBCO FreeStyle 293 Expression Medium | Life technologies | 12338018 | Culture medium |
293fectin Reagent | Invitrogen | 12347019 | Transfection reagent |
GIBCO Opti-MEM + GlutaMAX | Life technologies | 51985026 | Reduced Serum Medium for use during cationic lipid transfections |
GIBCO RPMI Medium 1640 (1x) | Life technologies | 52400-025 | |
CNBr-Activated Sepharose 4B | GE Healthcare | 17-0430-01 | pre-activated media for coupling antibodies or other large proteins |
Hydrochloric acid fuming 37% | Merck | 1003171000 | |
Sodium chloride | Sigma | S7653-1kg | |
Sodium bicarbonate | Sigma | 31437 | |
Trizma hydrochloride | Sigma | T3253-500G | TRIS hydrochloride |
Acetic Acid 100% | Merck | 1.00063.1000 | |
Glycin-HCl | Sigma | G2879 | |
PBS | Pharmacie, UMCU | Phosphate-Buffered Saline | |
10X TGS | BIO-RAD | 161-0772 | Tris/Glycine/SDS Buffer for SDS electrophoresis |
Mini-PROTEAN TGX Gels | BIO-RAD | 456-1046 | 12% SDS precast gels |
Trans-Blot Turbo Transfer Pack | BIO-RAD | 170-4157 | Western blot transfer packs |
Yarra 3u SEC-2000 column | Phenomenex | ||
InstantBlue Protein Stain | Expedeon | ISB1L | ready to use Coomassie protein stain for polyacrylamide gels |
Human IL-10 DuoSet ELISA | R&D | DY217B | |
Human TNFα ELISA Set | Diaclone | 851570020 | |
BCA Pierce Protein Assay Kit | ThermoFisher Scientific | 23227 | |
Carrageenan | Sigma-Aldrich | 22049 | plant mucopolysaccharide |
CFA | Sigma-Aldrich | F5881 | vaccine adjuvant |
Hamilton syringe | Sigma-Aldrich | 20779 | glass syringe |
Animal Enclosure | IITC Life Science | 433 | Animal Enclosure |
Von Frey mesh stand | IITC Life Science | 410 | Mesh Stand |
von Frey hairs | Stoelting | 58011 | touch test sensory probes |
Plantar Test (Hargreaves Method) | IITC Life Science | 390G | plantar test with heated glass |
Dynamic Weight Bearing test | Bioseb | BIO-DWB-AUTO-M | postural deficit test |
Glass Econo-Column Columns, 1.5 × 30 cm | BIO-RAD | 7371532 | glass chromatography column |
SnakeSkin Dialysis Tubing | Thermo Scientific | 88242 | |
Minisart NML Syringe Filter | Sartorius | 16555-K | single use filter unit, 0.45 μM |
CASY Cell Counter and Analyzer | Roche |