Volume 292, number 1,2, 187-190 FEBS 10345 © 1991 Federation of European Biochemical Societies 00145793/91/$3.50 ADONIS 001457939101055D Refolding of recombinant porcine growth hormone in a environment limits in vitro aggregate formation November 1991 reducing Nirdosh K. Puri Bunge Scientific and Technical Services, 87-89 Flemington Road, JYorth Melbourne, Vic. 3051, Australia Received 2 August 1991 Recombinant porcine growth hormone (rPGH) solubilized from bacterial inclusion bodies (IBs) using a cationic surfactant was oxidized to form disulphide bonds in a simple buffer solution containing 2-mereaptoethanol within an empirically derived optimal molar ratio of 2-mercaptoetha- nol:protein. A final yieid of 55% monomeric rPGH was achieved at protein concentrations of up to 5 mg/ml without the need for removal of the 2-mercaptoethartol or the use of chaotrophic agents. In the absence of 2-mercaptoethanol only 15% rnonomeric rPGH was obtained, with the majority forming higher molecular weight aggregates. Using the procedure derived for porcine growth hormone, it may be possible to obtain high yields of native protein and overcome the need for using low protein concentrations and chaotrophie agents during in vitro refolding of other disulphide bonded recombinant proteins. Aggregation; Growth hormone; Mercaptoethanol; Recombinant; Refolding 1. INTRODUCTION Problems of aggregation and/or insolubility com- monly encountered due to aberrant (non-native) disul- phide bonding during the in vitro refolding procedures used to oxidize intra-disulphide bonded proteins have meant that yields of correctly refolded protein are ge- nerally low [I-4]. Various refolding protocols of differ- ing complexity have been attempted to overcome these difficulties [3,5-7]. However, there appear to be no simple generally applicable means of obtaining high yields of recombinant biologically active disulphide bonded proteins. The degree of aggregation of recom- binant proteins during in vitro refolding is generally controlled by using denaturants (3-5 M Urea or I-2 M GnHC1) and by reoxidizing at very low protein con- centrations, of the order of 1-100 gg/ml [3,5,8]. These constrains often pose serious problems, particularly during industrial scale production of relatively low value, high volume products such as animal growth hormones. The results reported here use 2-mercaptoe- thanol as a simple means of creating an optimal in vitro environment to form native disulphide bonds by air oxidation. Reoxidation of disulphide bonded recom- binant proteins within an empirically optimized, nett reducing environment, may provide a relatively simple means of controlling the significant problem of aggrega- tion during in vitro refolding. Correspondence address: N.K. Purl, Bunge Scientific and Technical Services, 87-89 Flemington Road, North Melbourne, Vic. 3051, Australia. Fax (61) (3) 3290190. 2. EXPERIMENTAL 2. I. Solubilization of rPGtt Methionyl rPGH derived from plasmid pMG93 was expressed in E. coli as described in UK Patent No. 8701848. The resultant IBs were solubilized using the cationic surfactant cetyltimethylammonium chlo- ride (CTAC; ICI Australia Pty). Briefly, IBs were isolated by cell disruption, harvested by differential centrifugation and washed with 0.1 M citric acid, 0.2 M disodium phosphate pH 4.0 (2 x) and distilled H20 (2 ×) prior to use. IBs were used immediatdy or stored at pH 5 to 6.0 in a nitrogen-purged atmosphere. Approximately 50 mg of IBs (117 mg/ml dry weight) was solubilised at 10 mg/ml in a solution of 0.1 M Tris-HCl, pH I0.0, containing 2% mercaptoethanol (v/v) and 5% (w/v) CTAC for 1 h at 55°C. Tl]e solubilised inclusion bodies were clarified by centrifugation (I0 000 xg, 5 rain) and the supernatant fraction immediately mixed with 8 bed volumes of Dowex 50W x 4 (100-150) mesh) ion exchange resin (Dow Chemical Corporation, USA) equilibrated in 0.1 M Glycine-HCl and 5 M urea, pH 10.0 (See Australian Patent Applications 11 412 and 15 010) to obtain surfac- rant free soluble rPGH. 2.2. Refolding of rPGH Solubilised surfactant-free rPGH (1.5 mg/ml, based on dry weight of IBs) was dialysed against 20 mM ethanolamlne-HCl, pH I0.0, for 24--36 h in order to initiate oxidation of disulphides (refolding). For refolding in 2-mercaptoethanol, rPGH was exchanged using G-25 sephadex gel filtration (PD-10, 'desalting' columns; Pharmacia-LKB) into a solution of 20 mM ethanolamine-HCl, pill0.0, containing final concentrations of 5, 45, 55, 75 and 100 mM 2-mercaptoethanol. Re- folding was for 24~,8 h with shaking in an aerated environment at 4oc. 2.3. HPLC Analysis RP-HPLC was performed using CI alkyl-bonded silica columns (TSK-TMS 250, Toyo Soda Manufacturing Co., Tokyo, Japan; ob- tained through Pharmacia - LKB (Australia) Pty. Ltd.). Elution was performed at a flow rate of 0.5 ml/min at room teml~rature with water/acetonitrile mixtures containing 0.1% (v/v) trifluoroacetic acid (TFA) as modifier. A stepwise linear gradient was constructed as follows: 100% buffer A (0.1% TFA in dH~O) to 40% buffer B (100% Published by Elsevier Science Publishers B.V. 187
`
`APOTEX EX1030
`
`Page 1
`
`

`
`Volume 292, number 1,2 FEBS LETTERS November 1991 (a) OXl01~EO (311 MONOMI~II (~6%) = kd ~ 42.7 31.15 21.5 14.4 ........ ..---- r~''~'l~ (b) ~tFrDUCED GH ),IONOMER (96~) m--SOLVENT FRONT 1 2 3 .... ~ • .~. . .-.. i] 'll ' i '1 i (c) GH AGGREGATE (60"~) OXIDISRO GH MO~OME;I |1~%)--*- 1 2 Fig. I. RP-HPLC and 15% SDS-PAGE analyses comparing various purified 'standard' rPGH preparations and rPGH refolded without 2-mercaptoethanol. (a) RP-HPLC and SDS-PAGE analysis era puri- fied monomeric rPGH 'standard'. The rPGH is 96% pure by HPLC and has an M~ of 21.5K by SDS-PAGE. (b) RP-HPLC and SDS- PAGE analysis of reduced rPGH 'standard' prepared by treating the sample in Ca) with 2% (v/v) 2-mercaptoethanol for 1 h in 3.0 M urea. (SDS-Gel: lane I, purified rPGH: lane 2, reduced rPGH standard; lane 3. 'mol.wt.' markers.) Note that the correctly disulphide bonded (mo- nomeric) and the reduced rPGH preparations are clearly resolvable by their respective retention times on RP-HPLC (typically retention times differ by I rain) and by mobility on SDS-PAGE. (c) Solubilized rPGH rcfolded in the absence of 2-mercaptoethanol and analysed by RP-HPLC and SDS-PAGE. The species corresponding to oxidized. momomeric rPGH can be clearly identified as a peak (I 5% of protein) eluting with the same retention time ~s the purified rPGH 'standard' (cf. Fig. In) and with the same mobility on SDS-PAGE (cf. Fig. In, lane I). The other major eluting protein species (80% of protein) with a peak retention time typically 1.5-2.0 rain - greater than monomer, corresponds to a polydisperse 'aggregated'. that is, inter-molecular disulphide bonded rPGH of varying molecular weight on SDS-PAGE (lane 1, refolded rPST', lane 2. purified 'standard' rPGH). acetonitrile, 0,1% TFA) in 10 rain: 40% buffer B to 70% buffer B in 15 rain; and to 100% buffer B in 5 rain. The 100% acetonitrile. 0.1% TFA eluent was maintained for tl further 0.5 rain before re-equilibra- lion, prior to the next injection. 3. RESULTS AND DISCUSSION 3.1. Re[biding of rPGH #2 aqueous solutio, rPGH refolded against 20 mM ethanolamine.HCl, pH 10.0, was ~malysed by RP-HPLC to quantitate the 188
`
`Page 2
`
`

`
`Volume 292, number 1,2 FEBS LETTERS November 1991 proportion of correctly disulphide bonded (i.e. mono- meric, 21.5K rPGH) as a percentage of the total mono- mer and 'aggregated' forms. From the results shown in Fig. Ic approximately 15% of the rPGH was judged as monomerie after refolding. Approximately 80% existed as a polydisperse 'aggregate' species as judged by RP- HPLC and SDS-PAGE. No reduced rPGH was de- tected. 3.2. Refolding of rPGH in the presence of 2-mercaptoet- hanoi Samples of rPGH refolded in various concentrations of 2-mercaptoethanol were analysed by RP-HPLC. The yields ofmonomeric rPGH as a percentage of total peak area were respectively: 12% in the presence of 5 mM- mercaptoethanol; 25% in the presence of 45 mM 2- mercaptoethanol; 28% in the presence of 75 mM 2- mercaptoethanol, and 24% in the presence of I00 mM 2-mercaptoethanol (results not shown). The other major forms of rPGH present after refolding in 45-65 mM 2-mercaptoethanol comprised a major reduced species and a lesser proportion of the 'aggregated' species. The presence of these forms of rPGH was con- firmed by SDS-PAGE. Representative results comprising RP-HPLC and SDS-PAGE analysis of rPGH refolded in the optimal 2-mercaptoethanol concentration, 55 raM, are shown in Fig. 2. In addition to 28% oxidized monomeric rPGH, note the significant proportion of reduced (33%) rPGH in contrast to the results shown in Fig. Ic where the majority (80%) of the rPGH existed as a polydisperse aggregate population. 3.3. Effect of prote& concentration on yieM of monomer dur#~g refolding in 2-meycaptoethanol rPST at protein concentrations of 1.5-7.5 mg/ml was refolded in 55 mM 2-mercaptoethanol and yield of mo- nomeric rPGH estimated by RP-HPLC. Yields of 28%, 38% and 27% respectively for rPGH refolded at 1.5, 3.5 and 7.5 mg/mi were obtained (results not shown). As observed previously, the residual non-monomeric rPGH existed mainly as a reduced and lesser 'aggregat- ed species'. Clearly, to maximise yield, both the respec- tive concentrations of protein and 2-mercaptoethanol are required to be controlled during refolding, although surprisingly, higher yields of monomer were obtained at elevated prorein concentrations. 3.4. Secomlary oxidationh'efolding of rPGH The results described above demonstrated the need for a critical ratio of [protein]:[2-mercaptoethanol] dur- ing refolding in order to increase yield of monomer at the expense of undesirable 'aggregated' forms. How- ever, the presence of significant residual reduced rPGH (33%, cf. Fig. 2) even after 48 h of refolding in 55 mM 2-mercaptoethanol (irrespective of protein concentra- tion) suggested that an additional oxidation step in the 0XIDISED GH ~ ~. MONOMER (=~.) ++, !l,t + p,///', ,iV / i + i i Fig. 2. Results of RP-HPLC and 15% SDS-PAGE analysis of rPGH at 1.5 mg/ml refolded in 55 mM 2-mercaptoethanol. The identity of the oxidized monomeric (28% of protein), reduced (33%) and 'aggre- gated' species of rPST was established from Fig. 1 and conlimled by SDS-PAGE (lane I, rPGH refolded in 55 mM 2-mercaptoethanol; lane 2, purified 'standard' rPST; lane 3, reduced rPGH; lane 4, mol.wt. markers). absence of 2-mercaptoethanol might increase the final yield of monomeric rPGH above 38%. Final yields of approximately 55% monomeric rPGH were obtained after secondary oxidation as judged by RP-HPLC (Fig. 3). The remaining rPGH (40%) was present as polydis- perse aggregates as confirmed by SDS-PAGE (Fig. 3, lane 1 ). 4. CONCLUSION The results reported in this study show yields of up to 55% monomeric rPGH at protein concentrations of 3-5 mg/ml in the absence of denaturants such as urea or GnHCI. These results are comparable to those claimed for rPGH refolded at low concentrations in urea [9] and represent, at least for rPGH, several signif- icant departures from the current general dogma's for refolding recombinant proteins [1,5]: (i) high yield re- covery of native recombinant growth hormones and most other recombinant proteins has almost invariably necessitated using appropriate concentrations of chao- trophic agents during refolding, and (ii) refolding of recombinant proteins at concentrations of 1 mg/ml or 189
`
`Page 3
`
`

`
`Volume 292, number 1,2 FEBS LETTERS November 1991 1 2 3 .... OXIDI~;leO GH MONOMER (SS%) 4 5 QFI AGGREGATE ~,.<----. (,,e,O%) i i i i : Fig. 3. Results of RP-HPLC and SDS-PAGE analysis of rPGH fol- lowing secondary oxidation (refolding). The rPGH at 3.5 mg/ml was refolded for 2411 to obtain a yield of 38% monomer and approximately 30% residual reduced species, and subsequently exchanged via dialysis into 20 mM ethanolamine-HCl containing 50 mM CuCI2 for 24 h with aeration. Final yield of rPGH monomer was 55% by HPLC. SDS- PAGE gel (lane 1, rPGH after secondary refolding; lane 2, rPGH after primary refolding in 55 mM 2-mercaptoethanol; lane 3, purified 'standard' rPGH; lane 4, reduced 'standard' rPGH and lane 5, tool. wt. markers). Note the absence of reduced rPGH in lane I. less (commonly I-I00 gg/ml) has been necessary to maximize yields of native protein. Surprisingly, using 2-mercaptoethanol, we observed an inverse relationship between yield of monomer and concentration of protein during refolding, up to a value of 5 mg/ml. Clearly, the maintenance of a critical ratio of 2-mer- captoethanol:protein during disulphide bond formation provides an optimized in vitro environment that signif- icantly lessens rPGH aggregation via otherwise undesir- able intermolecular disulphide bonding. Moreover, the formation of the 'correct' disulphide bonds via air oxi- dation, occured without the necessity of removal of the 2-mercaptoethanol, where previously in the literature formation of protein disulphide bonds from the reduced state has involved substantially complete removal of reducing agent. It is interesting to consider whether the use of a nett reducing environment via a critical 2- mercaptoethanol:protein ratio during refolding will be generally applicable to the simplified and high yield recovery of the native forms of other disulphide bonded recombinant proteins. REFERENCES [1] Marston, F.A.O. (1986) Biochem. J. 240, 1-12. [2] Sharma, S.K. (1986) Sepcration Science and Technology 21,701- 726. [3] Schein, C.H. (1990) Biotech.nology 8, 308-317. [4] Spalding, B.J. (1990) Biotechnology 9. 229-233. [5] Jaenieke, R. and Rudolph. R. (1989) in: Protein Structure- A Practical Approach (Creighton T.E. ed.) pp. 191-223, IRL press, Oxford. [6] Cleland, J.L. and Wang, D.I.C. (1990) Biochemistry 29, 11072- ! 1078. [7] Cleland. J.L. and Wang, D.I.C. (1990) Biotechnology 8, 1274- 1278. [8] Schein, C.H. (1989) Biotechnology 7, 1141-1149. [9] Bentle, L.A., Mitchell, J.W. and Storrs, S.B. (1987) US Patent Number 4, 652, 630. 190
`
`Page 4