nucleic acid aptamers stabilize proteins against different types of stress conditions

RESEARCH ARTICLE – Pharmaceutical Biotechnology

Nucleic Acid Aptamers Stabilize Proteins Against Different Typesof Stress Conditions

HARDIK C. JETANI, ANKAN KUMAR BHADRA, NISHANT KUMAR JAIN, IPSITA ROY

Department of Biotechnology, National Institute of Pharmaceutical Education and Research (NIPER), Punjab 160 062, India

Received 18 September 2013; revised 18 October 2013; accepted 25 October 2013

Published online 20 November 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jps.23785

ABSTRACT: It has been observed that the same osmolyte cannot provide protection to a protein exposed to more than one stress condition.We wanted to study the effect of nucleic acid aptamers on the stabilization of proteins against a variety of stress conditions. Adjuvantedtetanus toxoid was exposed to thermal, freeze–thawing, and agitation stress. The stability and antigenicity of the toxoid were measured.Using nucleic acid aptamers selected against tetanus toxoid, we show that these specific RNA sequences were able to stabilize alumina-adsorbed tetanus toxoid against thermal-, agitation-, and freeze–thawing-induced stress. Binding affinity of the aptamer–protein complexdid not show any significant change at elevated temperature as compared with that at room temperature, indicating that the aptamerprotected the protein by remaining bound to it under stress conditions and did not allow either the protein to unfold or to promoteprotein–protein interaction. Thus, we show that by changing the stabilization strategy from a solvent-centric to a protein-centric approach,the same molecule can be employed as a stabilizer against more than one stress condition and thus probably reduce the cost of the productduring its formulation. C© 2013 Wiley Periodicals, Inc. and the American Pharmacists Association J Pharm Sci 103:100–106, 2014Keywords: adjuvant; alumina; aptamer; formulation; oligonucleotides; proteins; stability; stress; tetanus toxoid

INTRODUCTION

Therapeutic proteins comprise a significant division of theglobal pharmaceutical industry, and include some of the mostadvanced pharmaceutical products such as monoclonal anti-bodies, erythropoietin, hormones like growth hormone, insulin,and so on. Protein pharmaceuticals have higher target speci-ficities, pharmacological potencies, and frequently, lower sideeffects, than small molecule drugs. So, there is a renewed inter-est in the development of protein pharmaceuticals. One of thebiggest challenges in the development of protein formulations isthe issue of instability of proteins.1,2 This is of mainly two types:physical and chemical.3 Chemical instability involves processesthat make or break covalent bonds, generating new chemicalentities. Physical instability does not alter the chemical com-position, but results in a change in the physical state of theprotein. The secondary and tertiary structure elements, whichare prerequisites for function, are affected. For any pharma-ceutical protein, long-term stability and acceptable shelf life, inaddition to bioactivity, decide its therapeutic suitability. Thera-peutic proteins are routinely exposed to several destabilizingenvironments during their production, purification, storage,and delivery.4–7 Devising stabilization strategies to increase thestability of therapeutic proteins is a prime goal of protein for-mulation science.8

Several theories have been proposed to explain stabiliza-tion of proteins by small molecules. Classically, these are re-ferred to as preferential exclusion theory, water replacementtheory, and vitrification (immobilization) theory. According to

Correspondence to: Ipsita Roy (Telephone: +91-172-229-2061; Fax: +91-172-221-4692; E-mail: [email protected])

Nishant Kumar Jain’s present address is Macromolecule and Vaccine Stabi-lization Center, Department of Pharmaceutical Chemistry, University of Kansas,Lawrence, Kansas 66047.

Journal of Pharmaceutical Sciences, Vol. 103, 100–106 (2014)C© 2013 Wiley Periodicals, Inc. and the American Pharmacists Association

the first scenario, water (solvent) is excluded from the sur-face of the protein and is structured around the compatibleosmolyte (excipient).9,10 The excipient acts as a kosmotrope orwater structure maker. In the second case, which is used mainlyto explain the action of sugars and polyols, the osmolyte isthought to substitute for the hydrogen bonding capability ofwater. This maintains the native conformation of the protein ina low-water setting but restricts its mobility, thus minimizingintermolecular interactions.11,12 In the third situation, at a highconcentration of the osmolyte, a “cocoon” is formed around thebiomolecule, which again confines the protein and increasesits stability.13 The thrust of all these theories is the absenceof any interaction between the protein and the osmolyte. Arecent study has brought to the fore the drawbacks of employ-ing osmolytes as protein stabilizers. When a monoclonal anti-body was subjected to thermal stress (heating till 90◦C) in thepresence of trehalose and glycerol, its melting temperature in-creased as these polyols were shown to be excluded from theprotein surface to the greatest extent.14 When incubated at65◦C for 5 days, the monoclonal antibody showed greater lossof monomeric fraction in the absence than in the presence oftrehalose and glycerol. When the same monoclonal antibodywas subjected to mechanical stress (shaking at 200 rpm for 5days), the loss in fraction of monomer was the greatest in thepresence of trehalose, followed by sucrose and glycerol.14 Theloss was the least in buffer alone and ethylene glycol. A polyolhydrophobicity index (N) was proposed,15 according to which apolar polyol-like trehalose is expected to be maximally excludedfrom the protein surface, decreasing its flexibility and increas-ing its thermal stability. Nonpolar polyols-like ethylene glycol,on the contrary, interact with hydrophobic patches on the pro-tein surface and do not allow it to unfold during agitation, keep-ing it in solution.14 We have also observed differential resultswith using polyols to stabilize adjuvanted tetanus toxoid. Whenthe toxoid was exposed to freeze–thawing cycles in the presenceof osmolytes, glucose was able to retain the toxoid on alumina

100 Jetani et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:100–106, 2014

RESEARCH ARTICLE – Pharmaceutical Biotechnology 101

but was not able to protect its antigenicity.16 Trehalose, beinga cryoprotectant, was able to retain the antigenicity, althoughmore amount of protein was desorbed from the matrix in thiscase. When the same preparation was subjected to agitation,trehalose was not able to stabilize the protein,17 as seen in thecase earlier with the monoclonal antibody.14 The best resultsfor retention of antigenicity of the toxoid subjected to agita-tion were obtained with glucose and sorbitol. Depending on thestress the protein is exposed, to the requirement of the osmolytechanges.14,16,17 Thus, there is a need to search for novel stabiliz-ers that will be able to stabilize proteins against a wider varietyof stress conditions.

Aptamers are single-stranded nucleic acids (DNA or RNA),which bind to their targets with high affinity and specificity.18,19

They are referred to as chemical antibodies and compare favor-ably with conventional antibodies. Because of their syntheticorigin, they are free of cellular contaminants and are amenableto scale-up and/or modification. Aptamers are nontoxic and non-immunogenic, which increase their applications in the thera-peutics domain.20

Tetanus toxoid is a formaldehyde-inactivated, attenuatedform of tetanus toxin and is used as a prophylactic aid to combatall forms of tetanus.21 Like other proteins, tetanus toxoid is sus-ceptible to stress conditions that lead to its deterioration.22,23

Aggregation of tetanus toxoid leads to reduced bioavailabilityof the toxoid, undesirable immunogenicity, and failure of globalimmunization programs.24,25 Earlier, we selected specific RNAaptamers against tetanus toxoid, which bind to the protein withhigh affinity and stabilize the toxoid against moisture-inducedaggregation.26 We show here that the same nucleic acid ap-tamers are also able to stabilize adjuvanted tetanus toxoidagainst thermal, mechanical, and freeze–thawing stress.

EXPERIMENTAL

Materials

Tetanus toxoid was obtained as a gift from Shantha Biotech-nics Ltd. (Hyderabad, India). Ampicillin, deoxyribonucleotides,ribonucleotides, ribonuclease A, Luria–Bertani (LB) broth, alu-minium hydroxide gel (13 mg/mL, AlhydrogelTM, Cat. No.A8222), and 96-well plates (Costar) were purchased fromSigma–Aldrich (Bangalore, India). RNase-free DNase I, T7RNA polymerase, and yeast inorganic pyrophosphatase werepurchased from Fermentas Inc. (Hanover, Maryland). GoTaq R©Flexi DNA polymerase and GoTaq R© Flexi polymerase chainreaction (PCR) buffer were obtained from Promega Corpo-ration (Madison, Wisconsin). RNaseOUT was obtained fromInvitrogen Corporation (Carlsbad, California). Mouse antite-tanus toxoid monoclonal antibody was obtained from SantaCruz Biotechnology, Inc. (Dallas, Texas). Goat anti-mousehorseradish peroxidase-conjugated antibody and tetramethylbenzidine/hydrogen peroxide substrate were purchased fromBangalore Genei (Bangalore, India). All solutions were pre-pared in DEPC (diethylpyrocarbonate)-treated water.

Methodology

Adsorption of Tetanus Toxoid on Alumina

Dialyzed tetanus toxoid (3.2 mL, 6.2 mg/mL) was incubatedwith alumina (5 mL, 13 mg/mL) overnight with mild shakingat 4◦C, as described earlier.16,17 The amount of protein in the

Table 1. Description of RNA Aptamers Used in This Work (26).

Name Length of Sequence (bp) [Mg2+] During Selection (mM)

TT-13 97 3TT-17 97 3TT-20 97 3TT-23 97 4TT-24 97 4

supernatant was estimated using dye binding method27 afterthe matrix settled down.

Exposure of Tetanus Toxoid to Denaturation Conditions

Thermal Stress. Adjuvanted tetanus toxoid was incubated at50◦C with mild shaking (50 rpm) in a water bath. Aliquotswere withdrawn after 50 and 150 min and centrifuged at 500gfor 3 min. The amount of protein desorbed in the supernatantwas estimated by the dye binding method,27 whereas the pelletwas resuspended in phosphate buffer (10 mM, pH 7.4) andenzyme-linked immunosorbent assay (ELISA) was performedto determine antigenicity of protein.28

Agitation. Adjuvanted tetanus toxoid was agitated (300 rpm)at 37◦C for 2 h. Samples were centrifuged at 500g for 3 min.The amount of protein desorbed in the supernatant was esti-mated by the dye binding method,27 whereas the pellet wasresuspended in phosphate buffer (10 mM, pH 7.4) and ELISAwas performed to determine antigenicity of protein.28

Freeze–Thawing. Adjuvanted tetanus toxoid was subjected tofreeze–thawing by incubating the suspensions at −20◦C for12 h, and then at 37◦C in a water bath for 2 h. This was termedas the first cycle of freeze–thawing.16 The suspensions werereincubated at −20◦C for 12 h to start the next cycle. This wasrepeated for a total of five cycles. The samples were centrifugedat 500g for 3 min. The amount of protein desorbed in the super-natant was estimated by the dye binding method,27 whereasthe pellet was resuspended in phosphate buffer (10 mM, pH7.4) and ELISA was performed to determine antigenicity ofprotein.28 In all cases, the amount of protein desorbed from thematrix was determined by estimating the amount of proteinin the supernatant and was denoted as a fraction of the to-tal amount of protein adsorbed on the matrix, which has beenassigned a value of 100%.

Synthesis and Purification of RNA Aptamers

The RNA sequences showing high affinity for tetanus toxoidwere selected earlier by an iterative selection process26 and arelisted in Table 1. Glycerol stocks of five different clones (labeledas TT-13, TT-17, TT-20, TT-23, and TT-24) were taken andinoculated in 10 mL of LB media containing ampicillin (100:g/mL), incubated at 37◦C with shaking at 200 rpm overnight.Grown cells were harvested by centrifugation. Plasmid DNAwas extracted by alkaline lysis method.29 The isolated plas-mids were subjected to PCR using the primers and conditionsdescribed earlier.26 In vitro transcription was carried out withthe amplified product and the transcribed product was puri-fied by 8% urea denaturing polyacrylamide gel electrophoresis.All experiments were repeated in the presence of these puri-fied RNA sequences (3 and 6 :g). The desired amount of RNA

DOI 10.1002/jps.23785 Jetani et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:100–106, 2014

102 RESEARCH ARTICLE – Pharmaceutical Biotechnology

Figure 1. Estimation of integrity of adsorbed tetanus toxoid upon exposure to 50◦C for different time intervals. Values are presented aspercentage change as compared with originally adsorbed tetanus toxoid whose antigenicity is assumed to be 100%. 100% desorbed proteinassumes that all the originally adsorbed protein has been desorbed. Values represent mean ± SEM of three independent experiments. Emptybars indicate antigenicity, whereas filled bars show desorbed protein. ***p < 0.001, **p < 0.01, *p < 0.05 against adsorbed toxoid incubated in theabsence of RNA.

aptamer was added to the adsorbed protein (16.5 :g) in a totalvolume of 200 :L, incubated at 25◦C for 2 h and subjected tothe same conditions of stress (thermal, mechanical, or freeze–thawing) as the adsorbed protein alone (0 :g RNA). Controlrefers to alumina-adsorbed tetanus toxoid incubated at 4◦C inthe absence of RNA for the same period as the stress condition.

Determination of Binding Affinity

Dot-blot assay was carried out to determine the binding affin-ity (measured as dissociation constant) of the RNA aptamerfor the protein (tetanus toxoid) at room temperature and at50◦C. A constant concentration of fluorescein-labeled RNA wasincubated in a reaction volume of 40 :L with increasing con-centrations of tetanus toxoid in the respective binding buffers(50 mM phosphate buffer, pH 7.4 containing 150 mM NaCl, anddifferent concentrations of MgCl2, viz., 3 and 4 mM). After in-cubation for 30 min at room temperature and at 50◦C, sampleswere filtered through prewetted PVDF (polyvinylidene fluoride)membrane (0.45 :m) using a 96-well vacuum filtration mani-fold (Whatman-Biometra, Goettingen, Germany). The retainedprotein–RNA complex was washed with 1000 :L of the respec-tive wash buffer (50 mM phosphate buffer, pH 7.4 containing150 mM NaCl, and different concentrations of MgCl2, viz., 3and 4 mM) added in five equal aliquots. Membranes were driedbetween folds of filter papers and the fluorescence intensity ofthe retained protein–RNA complex was measured on an im-age scanner (Typhoon Trio; GE Healthcare, Uppsala, Sweden)

in the fluorescence mode using blue laser (excitation 488 nmand 526 SP emission filter). The amount of fluorescein-labeledRNA retained on the membrane was quantified by ImageQuantsoftware (GE Healthcare). The dissociation constants were de-termined by fitting the data into the Boltzmann’s equation:

y = yi + mxi + yf + mxf

1 + ex−x0J

where y is the fluorescence intensity. Initial and final asymp-totes are described by yi + mxi and yf + mxf, respectively. x0

denotes the concentration for 50% of maximal fluorescence.30

Dissociation constant (Kd) is calculated from the midpoint ofthe linear portion of the exponential curve.26

RESULTS

Exposure of Tetanus Toxoid to Thermal Stress

Adjuvanted tetanus toxoid was exposed to elevated tempera-ture (50◦C) and antigenicity was determined by ELISA.28 In theabsence of RNA aptamers, the residual antigenicity of tetanustoxoid reduced to 74.9 ± 1.8% and 63.4 ± 1.3% after 50 and150 min of incubation, respectively. This was accompanied bydesorption of the protein from the matrix (Fig. 1). After 50 min,23.0 ± 1.6% of the protein was desorbed, which increased upto 37.1 ± 1.4% after 150 min of incubation at 50◦C. In the

Jetani et al., JOURNAL OF PHARMACEUTICAL SCIENCES 103:100–106, 2014 DOI 10.1002/jps.23785


Figure 2. Estimation of dissociation constant of tetanus toxoid–aptamer complex by dot-blot assay at room temperature (filled bars)and at 50◦C (empty bars). The fluorescence intensity of the retainedfluorescein-labeled complex was measured using Typhoon trio (GEHealthcare). Densitometric analysis of the dots was carried out byImageQuant software (GE Healthcare) and data were fitted into theBoltzman equation. In all cases, dissociation constant (Kd) of tetanustoxoid–aptamer complex incubated at room temperature has been as-sumed to be 100%. N.S. denotes not significant, *p < 0.05 against Kdof toxoid-TT-23 complex at room temperature.

presence of aptamers, thermal stability of the adsorbed toxoidwas found to increase significantly. When incubated with 3 :gof TT-24, for example, the residual antigenicity of the toxoidwas 92.1 ± 2.2% after 150 min of thermal stress (Fig. 1). In-creasing the amount of RNA bound to the adjuvanted toxoid to6 :g did not increase the residual antigenicity of the adsorbedtoxoid further to any significant extent. This was seen to be thecase with most of the RNA aptamers except for TT-13 where re-duction in residual antigenicity was seen to occur on increasingthe amount of RNA (Fig. 1). Preservation of antigenicity in thepresence of aptamers was accompanied by increased retentionof protein on the matrix. After exposure to 50◦C in the presenceof aptamers, the amount of protein retained on the matrix wasfound to be around 90% for almost all aptamers (Fig. 1).

Next, we wanted to determine whether incubation at 50◦Caffects the binding between the protein and the aptamer. Forthis, binding affinities of RNA–protein complexes were deter-mined after incubating the protein–RNA complexes at 50◦Cand comparing them with those obtained at room temperature.Dissociation constants at room temperature were found to besimilar to what was reported earlier.26 When incubated at 50◦C,the dissociation constants were found to be similar to those atroom temperature except for TT-23 where marginal decreasein binding affinity for the toxoid was seen at 50◦C (Fig. 2).Thus, exposure to higher temperature did not release the RNAaptamer from the protein, and the increased retention of anti-genicity observed after heat stress was because of the presenceof RNA aptamers. This was also confirmed by the fact that noRNA could be detected in the supernatant obtained after cen-trifugation of the adsorbed samples after incubation at 50◦C(Fig. 3). Thus, association of the toxoid with specific RNA ap-tamers did not allow the protein to unfold and desorb from thematrix and stabilized it against thermal stress.

Exposure of Tetanus Toxoid to Agitation

Adjuvanted tetanus toxoid was exposed to agitation (300 rpmat 37◦C) for 2 h and antigenicity was determined by ELISA.28

Figure 3. Determination of the amount of RNA desorbed fromaptamer–tetanus toxoid complex following exposure to thermal stress.Supernatants obtained after centrifugation of incubated samples weresubjected to RT-PCR and the products were monitored by agarose gel(1.8% cross-linking) electrophoresis after 19 cycles of amplification.Conditions of amplification are described in Ref. 26. The intensitiesof bands (lanes 6–10) were compared with those obtained with RT-PCR of known amounts of RNA (lanes 1–5). Lane M: DNA ladder;lane 1: standard cDNA (20× diluted); lane 2: standard cDNA (16× di-luted); lane 3: standard cDNA (12× diluted); lane 4: standard cDNA(8× diluted); lane 5: standard cDNA (4× diluted); lane 6: product ob-tained after RT-PCR with supernatant of TT-24–protein complex; lane7: product obtained after RT-PCR with supernatant of TT-23–proteincomplex; lane 8: product obtained after RT-PCR with supernatant ofTT-13–protein complex; lane 9: product obtained after RT-PCR withsupernatant of TT-17–protein complex; lane 10: product obtained afterRT-PCR with supernatant of TT-20–protein complex.

Figure 4. Estimation of integrity of aptamer–tetanus toxoid com-plexes subjected to mechanical stress (agitation). Values are presentedas percentage change as compared with originally adsorbed tetanustoxoid (control, not subjected to agitation) whose antigenicity is as-sumed to be 100%. 100% desorbed protein assumes that all the orig-inally adsorbed protein has been desorbed. Values represent mean ±SEM of three independent experiments. Empty bars indicate antigenic-ity, whereas filled bars indicate desorbed protein. *p < 0.05 againstantigenicity of alum-adsorbed tetanus toxoid agitated in the absence ofaptamer.

In the absence of RNA aptamers, the residual antigenicity oftetanus toxoid reduced to 50.7 ± 2.1% following agitation. Ag-itation also led to significant desorption of the protein fromthe matrix. After 2 h, 28.5 ± 4.9% of the protein was desorbed(Fig. 4). These values are similar to what has been reportedearlier.17 In the presence of aptamers, the stability of the ad-sorbed toxoid against agitation increased significantly. For ex-ample, the residual antigenicity of the adsorbed toxoid was70.3 ± 2.2% after being subjected to agitation-induced stress in



the presence of 3 :g of TT-24 (Fig. 4). Increasing the amountof RNA complexed with the toxoid to 6 :g did not increase theamount of antigenically active toxoid to a significant extent.This was seen to be the case with all aptamers studied (Fig. 4).Retention of antigenicity in the presence of aptamers was ac-companied by increased amount of protein being retained onthe matrix. The average amount of toxoid desorbed from thematrix was 15% when complexed with aptamers (Fig. 4). Wehave reported earlier that when subjected to agitation in thepresence of RNA aptamers, a higher fraction of insulin existedin the monomeric state than in its absence.31 Agitation createsan air–water interface that favors the unfolding of the proteinand exposure of its hydrophobic core.6,32 This results in aggre-gation of the agitated protein. In the presence of RNA aptamers,the adsorbed toxoid cannot unfold when exposed to agitation.This leads to its stabilization and retention of the antigenicallyactive protein when subjected to agitation-induced stress.

Exposure of Tetanus Toxoid to Freeze–Thawing

In tropical countries, transportation of thermally labile bio-pharmaceuticals poses a challenge as their bioactivity needs tobe retained under harsh conditions. The use of cold chain hasbeen proposed for this purpose. However, the cold chain breaksdown frequently, exposing the vaccine to suboptimal tempera-tures. In a few cases, freezing was found to occur at almost allstages of transportation.33,34 Hence, efforts are on to increasethermal stability of protein drugs, especially vaccines, so thatthe use of cold chain can be eliminated.23

Adjuvanted tetanus toxoid was exposed to five cycles offreeze–thawing and antigenicity was determined by ELISA.28

In the absence of RNA aptamers, the residual antigenicityof tetanus toxoid was 24.5 ± 3.1% following freeze–thawing(Fig. 5). This is similar to what has been reported earlier.16 Fol-lowing freeze–thawing, significant amount of protein (18.3 ±0.7%) was desorbed from the matrix. The presence of RNA ap-tamers stabilized the adsorbed toxoid against freeze–thawing-induced stress. In the presence of 3 :g of TT-20, the residualantigenicity of the adsorbed toxoid was 43.9 ± 6.8% after being

Figure 5. Estimation of integrity of aptamer–tetanus toxoid com-plexes subjected to freeze–thawing. Values are presented as percentagechange as compared with originally adsorbed tetanus toxoid (control,not subjected to freeze–thawing) whose antigenicity is assumed to be100%. 100% desorbed protein assumes that all the originally adsorbedprotein has been desorbed. Values represent mean ± SEM of threeindependent experiments. Empty bars indicate antigenicity, whereasfilled bars indicate desorbed protein. **p < 0.01, *p < 0.05 against anti-genicity of alum-adsorbed tetanus toxoid agitated in the absence ofaptamer.

subjected to freeze–thawing (Fig. 5). Similar to what was seenwith other stress conditions, increasing the amount of RNA (6:g) did not improve the antigenicity of the retained toxoid sig-nificantly. Increase in residual antigenicity was observed forall aptamers, with retention of protein on the matrix (Fig. 5).Retention of antigenicity in the presence of aptamers was ac-companied by increased amount of protein remaining adsorbedon the matrix. The amount of protein desorbed from the matrixwas less than half of what was observed when freeze–thawingwas carried out in the absence of aptamers (Fig. 5). We havereported earlier that freezing leads to agglomeration of thematrix,16 which presumably results in the already unfolded(and hence less tightly bound) protein to be desorbed from it.In the presence of RNA aptamers, the adsorbed toxoid does notunfold and hence is partially retained on the matrix. This leadsto its stabilization and retention of antigenically active proteinwhen subjected to freeze–thawing.

DISCUSSION

After immobilization, the first step in the loss of activity ofany protein results from its unfolding.16,35,36 A study with threemodel proteins (cytochrome c, "-chymotrypsinogen A, and oval-bumin) immobilized on aluminum hydroxide particles showedthat adsorption has no significant effect on the thermal stabil-ity of these proteins.37 At a lower concentration of the protein,the rate of aggregation was found to be retarded for the ad-sorbed protein as a new intermediate state emerged at highertemperature, which was not seen for the free protein in solu-tion. This slowed down the transition kinetics from native to a$-sheet-rich aggregated state, increasing the thermal stabilityof the adsorbed protein.37 RNA aptamers bind to tetanus tox-oid via shape complementarity. As reported in this work, whenexposed to harsh conditions, the aptamers do not allow the tox-oid to unfold readily. This allows the protein to be retained onthe matrix. The RNA-bound protein complexes are negativelycharged,26 which reduces their density on the surface of thematrix because of steric repulsion. Reduced density further di-minishes the probability of aggregation because of crowdingand thus stabilizes the adsorbed toxoid against thermal stress.It may be noted that mere desorption of the protein from thematrix does not necessarily result in a deleterious effect. Wehave shown earlier that in the presence of glucose, a loweramount of tetanus toxoid is desorbed from alumina followingfreeze–thawing but this protein is unfolded (confirmed by a redshift in the fluorescence spectrum) and antigenically inactive.16

On the contrary, trehalose is not able to inhibit desorption asmuch as glucose, but the protein in the formulation (eitheradsorbed or desorbed from the matrix) is conformationally sta-ble (no change in 8max in the emission spectrum) and hence isantigenically active. Thus, the retention of antigenicity in thepresent case indicates that the aptamer-bound protein remainsin an antigenically active, folded conformation.

Compatible solutes or osmolytes such as sugars, polyols,amino acids, and so on have conventionally been used as pro-tein stabilizers. The mechanism of stabilization is through in-teraction with the solvent molecule and not directly with theprotein.9–13,38,39 Any alteration in the environment affects theinteraction between the stabilizer (osmolyte) and the solvent(Fig. 6). Thus, these stabilizers are condition specific, that is,they are able to stabilize the protein against one kind of stress



Figure 6. Scheme showing probable mechanism of stabilization of tetanus toxoid by small molecules and RNA aptamers.

condition but may fail to stabilize it against another stress.This is because a stress condition such as heat may affect thesolvent and thus the osmolyte–solvent interaction, in a differ-ent manner than another kind of stress, such as freezing oragitation.14,16,17 Osmolytes may prove to be better stabilizersagainst a particular set of stress condition. For example, expo-sure of adsorbed tetanus toxoid to freeze–thawing in the pres-ence of 0.15 M trehalose led to 71% retention of antigenicity.16

However, the disaccharide was unable to protect the adjuvantedtoxoid against agitation-induced stress.17 It is clear that anystabilizer that is able to protect the protein molecule againsta variety of stress conditions should have a protein-centric,rather than a solvent-centric, mode of action. On the contrary,such molecules should not differentially recognize hydrophilicand hydrophobic fractions of the protein molecule, as denat-urants such as urea tend to do and consequently favor theshifting of the protein-folding equilibrium toward the unfoldedstate. Aptamers fulfill both these criteria. They recognize theirtargets by shape complementarity,40 and hence interact di-rectly with the target protein and not with the solvent. Thefirst step in protein aggregation is either intermolecular pro-tein interaction or protein unfolding (to acquire the “nonna-tive” monomer conformation). We have shown earlier that thepresence of negatively charged RNA molecules creates stericrepulsion between tetanus toxoid molecules and keeps them insolution.26 This lowers the probability of intermolecular inter-actions. Similarly, specific RNA aptamers were able to partiallyinhibit agitation-induced aggregation of insulin.31 The presenceof aptamers had no effect on the in vitro cell signaling (L6 skele-tal muscle cell line) or in vivo (STZ-induced diabetic rat model)activities of the protein hormone. By binding to their targetswith high affinity, aptamers are also able to inhibit/delay theinitial unfolding of protein molecules. Because of their protein-centric effect, aptamers remain bound to the protein with highaffinity even under harsh conditions (Fig. 6). Osmolytes aremolecules that accumulate in cells under stress conditions suchas changes in temperature, pressure, and salinity. The conceptof osmolytes is intricately linked with the solvent.41,42 Their in-creased levels maintain osmolarity and restore cell turgor. We

have not come across any evidence that short single-strandedRNA sequences, that is, aptamers, behave in a similar manner.The mode of stabilization offered by aptamers is not stress-specific (Fig. 6), and they are able to stabilize the same pro-tein against different types of stress conditions. To that extent,their mode of action is different from that of a conventionalosomolyte, which interacts with the solvent (solvent centric),modulates it, and hence affects the conformation/stability ofthe protein.

CONCLUSIONS

Protein pharmaceuticals are exposed to a number of stress con-ditions during their life cycle, from production to formulationto transport, storage, and administration. Depending on thechallenge, a mixture of stabilizers (excipients) may be neededto retain the activity of the protein drug. In this work, we haveproposed a novel route of protein stabilization, in which theadditive (aptamer) is able to stabilize the protein (adjuvantedtetanus toxoid) under different stress conditions that a pro-tein drug is likely to face. Because of its unusual mechanismof action, that is, interaction with the protein and not the sol-vent, the same additive is able to stabilize the protein againsta variety of stress conditions. Because of the generality of theapproach, we believe that aptamers selected against specificproteins may emerge as “universal” stabilizers of those pro-teins. As aptamers are nontoxic and nonimmunogenic,20 theirin vivo use should not pose any regulatory concern.

ACKNOWLEDGMENTS

Financial support for this work was received from Indian Coun-cil of Medical Research. The authors are grateful to Rajeev Ku-mar Chaudhary for his help in carrying out a part of this work.A.K.B. acknowledges the award of junior research fellowshipfrom DST-INSPIRE program. N.K.J. acknowledges the awardof senior research fellowship from Council for Scientific andIndustrial Research.



REFERENCES

1. Jiskoot W, Randolph TW, Volkin DB, Middaugh CR, Schoneich C,Winter G, Friess W, Crommelin DJ, Carpenter JF. 2012. Proteininstability and immunogenicity: Roadblocks to clinical application ofinjectable protein delivery systems for sustained release. J Pharm Sci101:946–954.2. Hamrang Z, Rattray NJ, Pluen A. 2013. Proteins behaving badly:Emerging technologies in profiling biopharmaceutical aggregation.Trends Biotechnol 31:4484–458.3. Manning MC, Chou DK, Murphy BM, Payne RW, Katayama DS.2010. Stability of protein pharmaceuticals: An update. Pharm Res27:544–575.4. Kueltzo LA, Wang W, Randolph TW, Carpenter JF. 2008. Effects ofsolution conditions, processing parameters, and container materials onaggregation of a monoclonal antibody during freeze–thawing. J PharmSci 97:1801–1812.5. Kishore RS, Kiese S, Fischer S, Pappenberger A, Grauschopf U,Mahler HC. 2011. The degradation of polysorbates 20 and 80 and its po-tential impact on the stability of biotherapeutics. Pharm Res 28:1194–1210.6. Malik R, Roy I. 2011. Probing the mechanism of insulin aggregationduring agitation. Int J Pharm 413:73–80.7. Gomez N, Subramanian J, Ouyang J, Nguyen MD, Hutchinson M,Sharma VK, Lin AA, Yuk IH. 2012. Culture temperature modulatesaggregation of recombinant antibody in CHO cells. Biotechnol Bioeng109:125–136.8. Singh SK, Cousens LP, Alvarez D, Mahajan PB. 2012. Determinantsof immunogenic response to protein therapeutics. Biologicals 40:364–368.9. Lee JC, Timasheff SN. 1981. The stabilization of proteins by sucrose.J Biol Chem 256:7193–7201.10. Kaushik JK, Bhat R. 2003. Why is trehalose an exceptional pro-tein stabilizer? An analysis of the thermal stability of proteins in thepresence of the compatible osmolyte trehalose. J Biol Chem 278:26458–26465.11. Cho CH, Singh S, Robinson GW. 1997. Understanding all of water’sanomalies with a nonlocal potential. J Chem Phys 107:7979–7988.12. Grasmeijer N, Stankovic M, de Waard H, Frijlink HW, HinrichsWL. 2013. Unraveling protein stabilization mechanisms: Vitrificationand water replacement in a glass transition temperature controlledsystem. Biochim Biophys Acta 1834:763–769.13. Chang L, Shepherd D, Sun J, Ouellette D, Grant KL, Tang XC, PikalMJ. 2005. Mechanism of protein stabilization by sugars during freeze-drying and storage: Native structure preservation, specific interaction,and/or immobilization in a glassy matrix? J Pharm Sci 94:1427–1444.14. Abbas SA, Sharma VK, Patapoff TW, Kalonia DS. 2012. Oppositeeffects of polyols on antibody aggregation: Thermal versus mechanicalstresses. Pharm Res 29:683–694.15. Abbas SA, Sharma VK, Patapoff TW, Kalonia DS. 2011. Solubil-ities and transfer free energies of hydrophobic amino acids in polyolsolutions: Importance of the hydrophobicity of polyols. J Pharm Sci100:3096–3104.16. Solanki VA, Jain NK, Roy I. 2011. Stabilization of tetanus toxoidformulation containing aluminium hydroxide adjuvant against freeze–thawing. Int J Pharm 414:140–147.17. Solanki VA, Jain NK, Roy I. 2012. Stabilization of tetanus toxoidformulation containing aluminium hydroxide adjuvant against agita-tion. Int J Pharm 423:297–302.18. Kaur G, Roy I. 2008. Therapeutic applications of aptamers. ExpertOpin Investig Drugs 17:43–60.19. Krishnan Y, Bathe M. 2012. Designer nucleic acids to probe andprogram the cell. Trends Cell Biol 22:624–633.20. Keefe AD, Pai S, Ellington A. 2010. Aptamers as therapeutics. NatRev Drug Discov 9:537–550.

21. Khan AA, Zahidie A, Rabbani F. 2013. Interventions to reduceneonatal mortality from neonatal tetanus in low and middle incomecountries—A systematic review. BMC Public Health 13:322.22. Jain NK, Roy I. 2011. Accelerated stability studies for moisture-induced aggregation of tetanus toxoid. Pharm Res 28:626–639.23. Lee BY, Cakouros BE, Assi TM, Connor DL, Welling J, Kone S, DjiboA, Wateska AR, Pierre L, Brown ST. 2012. The impact of making vac-cines thermostable in Niger’s vaccine supply chain. Vaccine 30:5637–5643.24. Kristensen D, Chen D. 2010. Stabilization of vaccines: Lessonslearned. Hum Vaccin 6:229–231.25. Brown DW, Burton A, Gacic-Dobo M, Karimov RI, Vandelaer J,Okwo-Bele JM. 2011. A mid-term assessment of progress towards theimmunization coverage goal of the Global Immunization Vision andStrategy (GIVS). BMC Public Health 11:806.26. Jain NK, Jetani HC, Roy I. 2013. Nucleic acid aptamers as stabi-lizers of proteins: The stability of tetanus toxoid. Pharm Res 30:1871–1882.27. Bradford MM. 1976. A rapid and sensitive method for the quan-titation of microgram quantities of protein utilizing the principle ofprotein-dye binding. Anal Biochem 72:248–254.28. Determan AS, Wilson JH, Kipper MJ, Wannemuehler MJ,Narasimhan B. 2006. Protein stability in the presence of polymer degra-dation products: Consequences for controlled release formulations.Biomaterials 27:3312–3320.29. Sambrook J, Russell D. 2001. Molecular cloning: A laboratorymanual. Third ed. New York: Cold Spring Harbor Laboratory Press,pp 1.31–1.38.30. Uversky VN, Li J, Fink AL. 2001. Evidence for a partially folded in-termediate in alpha-synuclein fibril formation. J Biol Chem 276:10737–10744.31. Malik R, Roy I. 2013. Stabilization of bovine insulin againstagitation-induced aggregation using RNA aptamers. Int J Pharm452:257–265.32. Liu L, Qi W, Schwartz DK, Randolph TW, Carpenter JF. 2013.The effects of excipients on protein aggregation during agitation: Aninterfacial shear rheology study. J Pharm Sci 102:2460–2470.33. Nelson CM, Wibisono H, Purwanto H, Mansyur I, Moniaga V,Widjaya A. 2004. Hepatitis B vaccine freezing in the Indonesian coldchain: Evidence and solutions. Bull World Health Organ 82:99–105.34. Matthias DM, Robertson J, Garrison MM, Newland S, Nelson C.2007. Freezing temperatures in the vaccine cold chain: A systematicliterature review. Vaccine 25:3980–3986.35. Roy I, Sharma S, Gupta MN. 2004. Smart biocatalysts: Design andapplications. Adv Biochem Eng Biotechnol 86:159–89.36. Klein MP, Nunes MR, Rodrigues RC, Benvenutti EV, Costa TM,Hertz PF, Ninow JL. 2012. Effect of the support size on the proper-ties of $-galactosidase immobilized on chitosan: Advantages and dis-advantages of macro and nanoparticles. Biomacromolecules 13:2456–2464.37. Bai S, Dong A. 2009. Effects of immobilization onto aluminum hy-droxide particles on the thermally induced conformational behavior ofthree model proteins. Int J Biol Macromol 45:80–85.38. Timasheff SN. 2002. Protein-solvent preferential interactions, pro-tein hydration and the modulation of biochemical reactions by solventcomponents. Proc Natl Acad Sci USA 99:9721–9726.39. Jain NK, Roy I. 2009. Effect of trehalose on protein structure.Protein Sci 18:24–36.40. Hermann T, Patel DJ. 2000. Adaptive recognition by nucleic acidaptamers. Science 287:820–825.41. Tanford C. 1964. Isothermal unfolding of globular proteins in aque-ous urea solutions. J Am Chem Soc 86:2050–2059.42. Auton M, Rosgen J, Sinev M, Holthauzen LM, Bolen DW. 2011.Osmolyte effects on protein stability and solubility: A balancing actbetween backbone and side-chains. Biophys Chem 159:90–99.


nucleic acid aptamers stabilize proteins against different types of stress conditions

Documents