- Journal List
- Int J Biochem Mol Biol
- v.10(3); 2019
- PMC6737386
As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice
Int J Biochem Mol Biol. 2019; 10(3): 23–31.
Published online 2019 Aug 15.
PMCID: PMC6737386
PMID: 31523478
Fatmah M Hani,* Ashley E Cole,* and Elliot Altman
Author information Article notes Copyright and License information PMC Disclaimer
Abstract
Numerous studies have been conducted on the ability of salts to stabilize proteins in vitro using purified proteins demonstrating the fact that the ability of salts to stabilize proteins correlates with the Hofmeister series of ions. Using the well characterized bacterial aqueous cytosolic β-galactosidase and catechol 2,3-dioxygenase enzymes, we demonstrated that salts can stabilize proteins in vivo or intracellularly as well and that the ability of salts to stabilize these two proteins intracellularly also correlates with the Hofmeister series of ions. Na2SO4 and Na2HPO4 were very effective at stabilizing both proteins, followed by NaCl, NH4Cl and (NH4)2HPO4, while NH4CH3CO2, (NH4)2SO4 and NaCH3CO2 did not stabilize either of the proteins. We also investigated the ability of salts to rescue a collection of well characterized nonfunctional β-galactosidase and catechol 2,3-dioxygenase missense mutants that our laboratory has created. 73.33% of the β-galactosidase missense mutants could be rescued by salt, while only 33.33% of the catechol 2,3 dioxygenase missense mutants could be rescued by salt. This observation was explained by the differences in densities for the two proteins. Catechol 2,3 dioxygenase is almost twice as dense or compact as β-galactosidase and thus it is far easier for salts to penetrate and rescue inactive β-galactosidase proteins. 68.42% of the missense mutants that were rescuable by salt contained mutations that affected amino acids on the surface of the protein and is consistent with the likelihood that salt is able to rescue missense mutants that affect amino acids located on the surface of the protein much more readily than salt can rescue missense mutants that affect amino acids buried in the protein.
Keywords: Hofmeister ions, protein stability, salt suppression
Introduction
From the seminal studies of Hofmeister [1] regarding the ability of different salts to precipitate egg white protein, the lyotropic or Hofmesier series of ions was developed in which ions are ordered by their ability to precipitate proteins. For salts with the same cation, SO42- > HPO42- > F- > CH3CO2- > Cl- > Br- > NO3- > I- > ClO4- > SCN- and for salts with the same anion, (CH3)4N+ > Rb+ > K+ > Na+ > Li+ > Mg2+ > Ca2+. For general reviews see Jungwirth and Cremer [2] and Salis and Ninham [3]. Given the great significance of the Hofmeister series of ions, the 1888 Hofmeister article, which was written in German, has been translated in its entirety into English [4].
In a series of studies with collagen, gelatin and ribonuclease, Von Hippel and Wong [5-7] showed that the ability of the Hofmeister series of ions to precipitate proteins also correlated with their ability to stabilize proteins. Amongst the salts studied in these experiments, K2HPO4 and (NH4)2SO4 had the greatest stabilizing effects, followed by KCl and NaCl. Subsequent studies by Nandi and Robinson [8,9] using short peptides confirmed the fact that proteins were stabilized with respect to the Hofmeister series of ions. In these studies, Na2SO4 had the greatest stabilizing effect, followed by NaCl and NaBr. More recently Pelgram [10] also completed a thorough study on the ability of Hofmeister salts to stabilize the DNA binding domain (DBD) of lac repressor. In these studies, Na2SO4 and KF had the greatest stabilizing effects, followed by KCl and NaCl.
The ability of salts to stabilize proteins intracellularly has also been demonstrated in a number of intracellular studies using nonfunctional missense mutants. Hawthorne and Friis [11] were the first to identify mutants whose functionality could be partially restored by the addition of salt as they demonstrated that 36 out of 231 Saccharomyces cerevisiae auxotrophic mutants could be rescued by the addition of KCl at 0.5 M or 1.0 M and proposed that salt correctable mutants were most likely to be missense mutants. Other researchers showed subsequently that this phenomenon occurred in bacteria as well [12-15]. Good and Pattee [12] isolated 15 temperature sensitive Staphylococcus aureus mutants that could be corrected by 1 M NaCl. Russell [13] demonstrated that 5 out of 14 temperature sensitive Escherichia coli mutants could be rescued by 1% (0.171 M) NaCl. Bilsky and Armstong [14] found that 32 out of 40 temperature sensitive E. coli mutants could be rescued by 0.5% (0.086 M) NaCl. Kohno and Roth [15] showed that 56 temperature sensitive and 47 cold sensitive Salmonella enterica serovar Typhimurium histidine auxotrophs could all by rescued by 0.2 M NaCl.
In this study, we tested the ability of Hofmeister salts to stabilize wild-type proteins intracellularly. The well characterized bacterial aqueous cytosolic β-galactosidase and catechol 2,3-dioxygenase enzymes were chosen for this analysis. The structures for both of these proteins has been determined [16,17] and their activity can be quantified using robust easy to use colorimetric enzyme assays [18,19]. We also examined the ability of salts to rescue a collection of inactive missense mutants that our laboratory has generated in these two proteins [20].
Materials and methods
Media and bacterial strains
Lysogeny broth (LB) [21] was used as the rich media to maintain the strains used in this study. While most studies that have been conducted in E. coli and S. enterica on the ability of salt to rescue nonfunctional missense mutants have used nutrient broth (NB), the basal salt concentration in nutrient broth is higher than desired. The concentration of sodium and chlorine in NB are 0.007 M and 0.003 M, respectively [22]. For this reason, we utilized yeast glucose broth (YGB), which consisted of 5.0% yeast extract supplemented with 0.2% glucose, as the rich medium in which all the strains were grown to determine the effects of the different salts that were tested. The concentration of sodium and chlorine in YGB are 0.003 M and 0.001 M, respectively [22] and thus the basal salt concentration in YGB is considerably lower than NB. Strains that were grown in YGB yielded saturated cultures with similar OD550 values to strains that were grown in NB or LB. To induce the expression of either the β-galactosidase enzyme coded by lacZ or the catechol 2,3-dioxygenase enzyme coded by xylE, isopropyl β-D-thiogalactopyranoside (IPTG) was added at a final concentration of 1 mM using the S. enterica strains TT18519, hisC10081::MudF(lac+) and ALS1442, proB::xylE(cat), respectively [19]. Salt solutions to be tested were prepared at 1M concentrations, adjusted to pH 7 if necessary, and added to 2X YGB to achieve the final concentration desired in 1X YGB.
β-galactosidase and catechol 2,3-dioxygenase enzyme assays
β-galactosidase and catechol 2,3 dioxygenase assays assays were performed as described [18,19] with two important modifications to eliminate the salt that is present in the buffers used in the assays. Z buffer was replaced with 10 mM Tris; pH 7 and both the ortho-nitrophenyl-β-galactoside (ONPG) and the catechol substrate solutions were prepared in 10 mM Tris; pH 7 instead of phosphate buffer.
Three dimensional analysis of the β-galactosidase and catechol 2,3-dioxygenase proteins
The PyMOL Molecular Graphics System (Schrödinger, LLC) was used to determine the volume of the β-galactosidase and catechol 2,3-dioxygenase proteins. The Research Collaboratory for Structural Bioinformatics Protein Data Bank (RCSB PDB) was the source of the three dimensional crystal structure data. The 1DP0 file was used for the β-galactosidase crystal structure data [17] and the 1MPY file was used for the catechol 2,3-dioxygenase crystal structure data [16].
Results
Determining the ability of different salts to stabilize the β-galactosidase and catechol 2,3-dioxygenase enzymes intracellularly
We selected eight salts, NH4CH3CO2 (ammonium acetate), NH4Cl (ammonium chloride), (NH4)2HPO4 (ammonium phosphate dibasic), (NH4)2SO4 (ammonium sulfate), NaCH3CO2 (sodium acetate), NaCl (sodium chloride), Na2HPO4 (sodium phosphate dibasic) and Na2SO4 (sodium sulfate), for these studies based on three criteria. First, they contained both very strong and moderate anions and cations for stabilizing proteins according to the Hofmeister series of ions. Second, these were generally the best salts for stabilizing proteins based on previous studies [5-10]. Third, of the anions and cations that were selected, all of the possible combinations were tested.
The wild-type strains that produced β-galactosidase and catechol 2,3-dioxygenase were grown in YGB plus 1 mM IPTG that contained increasing salt in 0.05 M increments. After 16 hours of growth β-galactosidase or catechol 2,3-dioxygenase enzyme assays were performed and the results are shown in Table 1. Na2SO4 was the best salt for stabilizing β-galactosidase, followed by Na2HPO4, NaCl, NH4Cl and (NH4)2HPO4. NH4CH3CO2, (NH4)2SO4 and NaCH3CO2 did not stabilize β-galactosidase. Na2SO4 was the best salt for stabilizing catechol 2,3-dioxygenase, followed by Na2HPO4, (NH4)2HPO4, NH4Cl and NaCl. NH4CH3CO2, (NH4)2SO4 and NaCH3CO2 did not stabilize catechol 2,3-dioxygenase. Thus there was very good agreement in the ability of the various salts to stabilize both proteins. The molar concentrations at which the different salts provided their maximal stabilization varied widely and ranged from 0.05 M to 0.35 M for both enzymes.
Table 1
Impact of different concentrations of Hofmeister salts on the activity of β-galactosidase and catechol 2,3-dioxygenase
| Salts | 0 M | 0.05 M | 0.1 M | 0.15 M | 0.2 M | 0.25 M | 0.3 M | 0.35 M | 0.4 M |
|---|---|---|---|---|---|---|---|---|---|
| β-galactosidase | |||||||||
| NH4CH3CO2 | 152.11 | 33.53 | 25.08 | 19.59 | 9.39 | ND | ND | ND | ND |
| NH4Cl | 152.11 | 179.73 | 174.20 | 150.20 | 140.81 | 141.21 | 140.26 | 139.84 | 136.25 |
| (NH4)2HPO4 | 152.11 | 171.10 | 105.94 | 97.84 | 84.16 | 81.42 | 32.19 | 30.82 | 20.84 |
| (NH4)2SO4 | 152.11 | 24.41 | 43.91 | 65.34 | 110.49 | 105.71 | 103.46 | 101.93 | 98.44 |
| NaCH3CO2 | 152.11 | 88.42 | 80.72 | 95.50 | 110.18 | 83.64 | 80.11 | 76.62 | 74.15 |
| NaCl | 152.11 | 233.40 | 225.28 | 232.12 | 252.71 | 248.30 | 255.90 | 277.20 | 311.02 |
| Na2HPO4 | 152.11 | 303.21 | 366.04 | 384.11 | 392.20 | 395.90 | 409.15 | 392.54 | 387.37 |
| Na2SO4 | 152.11 | 323.43 | 390.31 | 473.60 | 604.90 | 659.46 | 770.16 | 777.96 | 667.50 |
| Catechol 2,3-dioxygenase | |||||||||
| NH4CH3CO2 | 344.79 | 32.21 | 25.39 | 21.63 | 22.61 | ND | ND | ND | ND |
| NH4Cl | 344.79 | 368.45 | 413.99 | 464.59 | 464.05 | 442.13 | 399.63 | 428.02 | 483.88 |
| (NH4)2HPO4 | 344.79 | 658.53 | 643.23 | 635.43 | 613.40 | 484.03 | 301.61 | 247.72 | 117.15 |
| (NH4)2SO4 | 344.79 | 67.80 | 333.54 | 303.95 | 258.16 | 152.99 | 72.86 | 68.49 | 70.19 |
| NaCH3CO2 | 344.79 | 34.86 | 21.40 | 17.99 | 18.88 | 18.33 | 10.38 | 2.69 | 3.17 |
| NaCl | 344.79 | 307.45 | 393.44 | 385.58 | 397.93 | 387.37 | 389.36 | 375.36 | 367.17 |
| Na2HPO4 | 344.79 | 739.92 | 847.58 | 852.98 | 852.33 | 884.97 | 858.91 | 897.50 | 952.78 |
| Na2SO4 | 344.79 | 322.28 | 415.87 | 442.38 | 551.09 | 881.66 | 922.25 | 977.02 | 142.65 |
Open in a separate window
β-galactosidase activities are reported in Miller Units and catechol 2,3-dioxygenase activities are reported in XylE Units. All assays were repeated in triplicate and the standard deviation was less than 10%. ND denotes that these values could not be determined. NH4CH3CO2 completely inhibited cell growth at concentrations of 0.25 M or above.
Determining the ability of salt to suppress β-galactosidase and catechol 2,3-dioxygenase missense mutants
Numerous researchers have shown that nonfunctional missense mutants can be rescued by salt in vivo. We wanted to extend these studies by determining whether a collection of 42 unique well-characterized nonfunctional missense mutants, 15 in β-galactosidase and 27 in catechol 2,3-dioxygenase, could be rescued by salt. Because NaCl has been the salt of choice in the previous salt correctable missense mutant studies we used NaCl as well. Both 0.2 M and 0.3 M final concentrations of NaCl were tested to see if salt could rescue the missense mutants. Table 2 shows the results of the enzyme assays for missense mutants that had at least a 25% or 1.25 fold increase in enzyme activity upon the addition of salt. Interestingly, NaCl had a negative effect on the missense mutants that were not rescuable by the addition of NaCl. 11 out of the 15, or 73.33% of the nonfunctional β-galactosidase missense mutants could be rescued by the addition of NaCl, but only 9 out of the 27, or 33.33% of the nonfunctional catechol 2,3-dioxygenase missense mutants could be rescued by the addition of NaCl. For the β-galactosidase missense mutants, the β-galactosidase activity was increased by an average of 5.53 fold with the addition of 0.2 M NaCl and 6.04 fold with the addition of 0.3 M NaCl. For the catechol 2,3-dioxygenase missense mutants, the catechol 2,3-dioxygenase activity was increased by an average of 1.64 fold with the addition of 0.2 M NaCl and 1.83 fold by the addition of 0.3 M NaCl. Thus the higher concentration of NaCl was slightly more effective at rescuing the nonfunctional missense mutants that could be rescued by salt. lacZ39 was by far the most suppressible missense mutant as its β-galactosidase activity was increased 19.68 fold with the addition of 0.2 M NaCl and 26.54 fold with the addition of 0.3 M NaCl.
Table 2
Enzymatic activity of the salt correctible lacZ and xylE missense mutants in the presence of 0.2 M or 0.3 M NaCl
| Mutant | 0.0 M NaCl | 0.2 M NaCl | 0.2 M NaCl Fold Increase | 0.3 M NaCl | 0.3 M NaCl Fold Increase |
|---|---|---|---|---|---|
| β-galactosidase | |||||
| lacZ39 | 1.29 | 25.39 | 19.68 | 34.24 | 26.54 |
| lacZ337 | 0.65 | 1.03 | 1.58 | 0.90 | 1.38 |
| lacZ364 | 0.48 | 1.04 | 2.17 | 0.86 | 1.79 |
| lacZ2234 | 0.70 | 0.89 | 1.27 | 0.91 | 1.30 |
| lacZ2343 | 14.28 | 77.96 | 5.46 | 87.54 | 6.13 |
| lacZ2381 | 0.78 | 2.37 | 3.04 | 2.67 | 3.42 |
| lacZ2396 | 0.24 | 0.43 | 1.79 | 0.31 | 1.29 |
| lacZ2449 | 1.26 | 14.27 | 11.33 | 12.41 | 9.85 |
| lacZ2456 | 1.20 | 10.07 | 8.39 | 10.18 | 8.48 |
| lacZ2530 | 0.79 | 2.42 | 3.06 | 2.61 | 3.30 |
| lacZ2540 | 0.75 | 2.26 | 3.01 | 2.24 | 2.99 |
| Catechol 2,3-dioxygenase | |||||
| xylE1568 | 7.46 | 10.40 | 1.39 | 14.17 | 1.90 |
| xylE1569 | 20.59 | 25.69 | 1.25 | 30.58 | 1.49 |
| xylE1583 | 8.34 | 11.24 | 1.35 | 11.82 | 1.42 |
| xylE1585 | 8.17 | 17.67 | 2.16 | 15.00 | 1.84 |
| xylE1591 | 14.16 | 20.34 | 1.44 | 26.75 | 1.89 |
| xylE1683 | 6.76 | 16.82 | 2.49 | 11.77 | 1.74 |
| xylE1698 | 7.42 | 10.41 | 1.40 | 15.56 | 2.10 |
| xylE1702 | 9.60 | 14.14 | 1.47 | 21.09 | 2.20 |
| xylE1710 | 6.18 | 10.92 | 1.77 | 11.61 | 1.88 |
Open in a separate window
β-galactosidase activities are reported in Miller Units and catechol 2,3-dioxygenase activities are reported in XylE Units. All assays were repeated in triplicate and the standard deviation was less than 10%.
Using data from Cole et al. [20], Tables 3 and and44 list the β-galactosidase and catechol 2,3-dioxygenase missense mutants, respectively, and gives the amino acid change, what secondary structure is affected by the missense mutant and what type of impact the mutation is expected to have on the secondary structure, whether the mutated amino acid is located on the surface of the protein, partially located on the surface of the protein or buried, the change in hydropathy caused by the mutation, and whether the missense mutant can be rescued by salt. Based on our knowledge of protein folding and structure, one would expect that missense mutants which affect amino acids located on the surface of the protein would be far more likely to be rescuable by salt than missense mutants which affect buried amino acids of the protein. 13 out of the 19, or 68.42% of the missense mutants that were rescuable by salt contained mutations on the surface of the proteins. Amongst the β-galactosidase missense mutants, where the large majority were rescuable by salt, the four missense mutants that could not be rescued by salt all contained mutations which affected amino acids that were buried in the protein.
Table 3
β-galactosidase missense mutants
| lacZ mutant | Amino acid change1 | Structural perturbation2 | Mutation favorability3 | Protein Location4 | Change in hydropathy5 | Salt Suppressible |
|---|---|---|---|---|---|---|
| 39 | 1018 Leu to Gln | Beta Sheet | - | Surface (Partial) | +* | + |
| 337 | 905 Glu to Lys | Coil | + | Surface (Partial) | + | + |
| 361 | 900 Gly to Asp | Beta Sheet | - | Buried | +* | - |
| 364 | 208 Gly to Asp | Beta Sheet | - | Buried | +* | + |
| 2234 | 565 Gly to Asp | Beta Sheet | - | Buried | +* | + |
| 2343 | 943 Arg to His | Beta Sheet | + | Surface | -* | + |
| 2381 | 764 Gly to Asp | Coil | - | Buried | +* | + |
| 2382 | 566 Gly to Asp | Beta Sheet | - | Buried | +* | - |
| 2396 | 584 Gly to Asp | Coil | - | Surface | +* | + |
| 2449 | 202 Asp to Asn | Coil | + | Surface (Partial) | - | + |
| 2454 | 354 Gly to Asp | Beta Sheet | - | Buried | +* | - |
| 2456 | 5 Thr to Met | Alpha Helix | +* | Surface | -* | + |
| 2530 | 419 His to Tyr | Coil | -* | Surface | - | + |
| 2540 | 4 Ile to Asn | Alpha Helix | -* | Surface | +* | + |
| 2608 | 460 Gly to Arg | Coil | -* | Buried | +* | - |
Open in a separate window
1The lacZ gene codes for the 1,024 amino acid β-galactosidase protein. The resulting amino acid changes are given based on the coded protein predicted by the DNA sequence and not the Protein Data Bank file.
2The secondary structure affected by the mutation was determined using PyMol. Data from Jacobson et al. [23] had to be used to determine the secondary structures affected by lacZ2456 and lacZ2540, since these mutations were at the extreme amino terminus.
3The likelihood of a mutation to affect the α-helical, β-sheet or random coil structure was determined using Pα, Pβ or Pc values from the averaged propensity scale in Cole et al. [20]. (+) indicates a favorable change and (-) indicates an unfavorable change. An asterisk indicates a mutation that changes the original amino acid from or to one of the preferred amino acids that are found in α-helices, β-sheets or random coils.
4The location of the mutated amino acids was determined using PyMOL.
5The change in hydropathy was determined using the averaged hydrophobicity scale from Cole et al. [20]. (+) indicates a more hydrophilic change, (-) indicates a more hydrophobic change. An asterisk indicates a mutation that changes the hydropathy of the original amino acid significantly and results in a shift of at least 5 amino acids.
Table 4
Catechol 2,3-dioxygenase missense mutants
| xylE mutant | Amino acid change1 | Structural perturbation2 | Mutation favorability3 | Protein Location4 | Change in hydropathy5 | Salt Suppressible |
|---|---|---|---|---|---|---|
| 1568 | 106 Arg to Cys | Beta Sheet | +* | Surface (Partial) | -* | + |
| 1569 | 270 Gly to Gln | Coil | -* | Surface (Partial) | +* | + |
| 1571 | 127 Gly to Gln | Coil | -* | Surface (Partial) | +* | - |
| 1574 | 233 Ser to Phe | Alpha Helix | + | Surface | -* | - |
| 1577 | 21 Ala to Val | Alpha Helix | -* | Buried | -* | - |
| 1581 | 30 Gly to Ser | Coil | - | Surface (Partial) | + | - |
| 1582 | 64 Gly to Asp | Beta Sheet | - | Surface | +* | - |
| 1583 | 158 Gly to Asp | Beta Sheet | - | Buried | +* | + |
| 1584 | 243 Pro to Ser | Coil | - | Surface (Partial) | + | - |
| 1585 | 9 Gly to Ser | Coil | - | Buried | + | + |
| 1587 | 178 Gln to Lys | Beta Sheet | + | Surface | + | - |
| 1588 | 247 Gly to Asp | Coil | - | Surface (Partial) | +* | - |
| 1589 | 282 Trp to Ser | Beta Sheet | -* | Surface (Partial) | +* | - |
| 1591 | 8 Pro to Ser | Beta Sheet | + | Buried | + | + |
| 1592 | 193 Ser to Asn | Beta Sheet | - | Buried | + | - |
| 1593 | 245 Arg to Cys | Coil | + | Surface (Partial) | -* | - |
| 1665 | 242 Gly to Asp | Coil | - | Surface (Partial) | +* | - |
| 1666 | 251 Gly to Asp | Coil | - | Surface (Partial) | +* | - |
| 1670 | 202 Ala to Thr | Beta Sheet | +* | Buried | + | - |
| 1677 | 213 His to Tyr | Beta Sheet | +* | Buried | - | - |
| 1680 | 216 Ser to Phe | Beta Sheet | +* | Buried | -* | - |
| 1681 | 200 Asp to Asn | Coil | + | Buried | - | - |
| 1683 | 115 His to Tyr | Coil | -* | Surface (Partial) | - | + |
| 1684 | 137 Glu to Lys | Coil | + | Surface | + | - |
| 1698 | 152 Asp to Asn | Coil | + | Buried | - | + |
| 1702 | 24 His to Tyr | Alpha Helix | + | Surface (Partial) | - | + |
| 1710 | 290 Ala to Thr | Alpha Helix | -* | Surface (Partial) | + | + |
Open in a separate window
1The xylE gene codes for the 307 amino acid catechol 2,3-dioxygenase protein. The resulting amino acid changes are given based on the coded protein predicted by the DNA sequence and not the Protein Data Bank file.
2The secondary structure affected by the mutation was determined using PyMol.
3The likelihood of a mutation to affect the α-helical, β-sheet or random coil structure was determined using Pα, Pβ or Pc values from the averaged propensity scale in Cole et al. [20]. (+) indicates a favorable change and (-) indicates an unfavorable change. An asterisk indicates a mutation that changes the original amino acid from or to one of the preferred amino acids that are found in α-helices, β-sheets or random coils.
4The location of the mutated amino acids was determined using PyMOL.
5The change in hydropathy was determined using the averaged hydrophobicity scale in Cole et al. [20]. (+) indicates a more hydrophilic change, (-) indicates a more hydrophobic change. An asterisk indicates a mutation that changes the hydropathy of the original amino acid significantly and results in a shift of at least 5 amino acids.
Determining the ability of different salts to stabilize the β-galactosidase enzyme produced by the lacZ39 missense mutant
Because the lacZ39 missense mutant was the most rescuable by NaCl, we tested the ability of the eight salts we had used previously with wild-type β-galactosidase and catechol 2,3-dioxygenase to see how they affected the inactive β-galactosidase enzyme produced by lacZ39. Table 5 shows the result of this study. Na2SO4 was the best salt for stabilizing β-galactosidase from lacZ39, followed by NaCl, NH4Cl, (NH4)2SO4, NaCH3CO2, Na2HPO4 and (NH4)2HPO4. Only NH4CH3CO2 did not stabilize the β-galactosidase from lacZ39. The results clearly demonstrate that the salts affect wild-type β-galactosidase enzyme differently than an inactive mutant β-galactosidase enzyme. While Na2SO4 was the best salt at stabilizing both wild-type β-galactosidase and the β-galactosidase produced from lacZ39, NaCl was almost as effective at stabilizing the β-galactosidase produced from lacZ39. Neither NH4CH3CO2, (NH4)2SO4 or NaCH3CO2 could stabilize wild-type β-galactosidase, but only NH4CH3CO2 was ineffective at stabilizing the β-galactosidase produced from lacZ39.
Table 5
Impact of different concentrations of Hofmeister salts on the activity of β-galactosidase for the lacZ39 missense mutant
| Salts | 0 M | 0.05 M | 0.1 M | 0.15 M | 0.2 M | 0.25 M | 0.3 M | 0.35 M | 0.4 M |
|---|---|---|---|---|---|---|---|---|---|
| NH4CH3CO2 | 1.29 | .096 | 1.05 | 0.84 | 0.80 | ND | ND | ND | ND |
| NH4Cl | 1.29 | 3.11 | 8.97 | 10.08 | 13.65 | 16.88 | 19.94 | 21.44 | 18.47 |
| (NH4)2HPO4 | 1.29 | 2.28 | 1.91 | 1.38 | 1.14 | 1.08 | 0.60 | 0.52 | 0.43 |
| (NH4)2SO4 | 1.29 | 3.74 | 12.07 | 14.42 | 17.02 | 2.26 | 1.91 | 1.74 | 1.01 |
| NaCH3CO2 | 1.29 | 3.17 | 2.84 | 3.40 | 4.46 | 6.36 | 8.23 | 8.41 | 7.66 |
| NaCl | 1.29 | 7.06 | 10.34 | 16.06 | 25.39 | 31.38 | 34.24 | 35.31 | 40.25 |
| Na2HPO4 | 1.29 | 3.83 | 4.49 | 4.55 | 4.17 | 3.39 | 2.90 | 2.58 | 2.52 |
| Na2SO4 | 1.29 | 11.17 | 22.11 | 36.39 | 57.26 | 48.67 | 45.18 | 39.15 | 34.64 |
Open in a separate window
β-galactosidase activities are reported in Miller Units. All assays were repeated in triplicate and the standard deviation was less than 10%. ND denotes that these values could not be determined. NH4CH3CO2 completely inhibited cell growth at concentrations of 0.25 M or above.
Discussion
In this study, we have examined whether the best salts known to stabilize purified proteins in vitro based on previous studies [5-10] could also perform in vivo or intracellularly using the two very well characterized β-galactosidase and catechol 2,3 dioxygenase enzymes. To ensure that all the possible combinations of anions and cations were tested based on the salts that have proven the most effective at stabilizing proteins in in vitro studies, we tested the following eight salts, NH4CH3CO2 (ammonium acetate), NH4Cl (ammonium chloride), (NH4)2HPO4 (ammonium phosphate dibasic), (NH4)2SO4 (ammonium sulfate), NaCH3CO2 (sodium acetate), NaCl (sodium chloride), Na2HPO4 (sodium phosphate dibasic) and Na2SO4 (sodium sulfate). Generally, Na2SO4 and Na2HPO4 were very effective at stabilizing both proteins, followed by NaCl, NH4Cl and (NH4)2HPO4, while NH4CH3CO2, (NH4)2SO4 and NaCH3CO2 did not stabilize either of the proteins. However, Table 1 clearly shows some distinct differences in the ability of the different salts to stabilize the two proteins. Na2SO4 and Na2HPO4 are equally effective at stabilizing catechol 2,3 dioxygenase, but Na2SO4 is much more effective at stabilizing β-galactosidase than Na2HPO4. NaCl is much better at stabilizing β-galactosidase than NH4Cl and (NH4)2HPO4, while (NH4)2HPO4 is much better at stabilizing catechol 2,3 dioxygenase than NH4Cl or NaCl.
Numerous researchers have shown that salt can rescue inactive proteins intracellularly by testing the ability of salt to suppress nonfunctional missense mutants [11-15]. We have expanded these studies by testing the ability of salt to rescue a set of 42 well characterized unique nonfuctional missense mutants, 15 in β-galactosidase and 27 in catechol 2,3 dioxygenase. A total of 19 out of the 42, or 45.24% of the missense mutants were rescuable by salt (11 out of 15, or 73.33% for β-galactosidase and 9 out of 27, or 33.33% for catechol 2,3 dioxygenase). Thus clearly it was a lot easier to suppress nonfunctional missense mutants in β-galactosidase than it was to suppress nonfunctional missense mutants in catechol 2,3 dioxygenase. An examination of the structure of the two proteins provides an answer.
Both the β-galactosidase and catechol 2,3 dioxygenase proteins are tetrameric. The β-galactosidase monomer is 1,024 amino acids in size and the tetramer’s molecular weight is 465,912 g/mol with width, height and depth measurements of 174.10 Å, 136.00 Å and 86.75 Å, respectively. The catechol 2,3 dioxygenase monomer is 307 amino acids in size and the tetramer’s molecular weight is 140,616 g/mol with width, height and depth measurements of 94.95 Å, 65.70 and 51.60 Å, respectively. Thus the density of β-galactosidase is 0.377 g/cm3, while the density of catechol 2,3 dioxygenase is 0.725 g/cm3. Since β-galactosidase is a lot less dense or compacted than catechol 2,3 dioxygenase, it should be a lot easier for salts to penetrate and effect β-galactosidase than catechol 2,3 dioxygenase. Additionally, β-galactosidase has a unique structure that consists of a continuous system of channels running along the surface and within the tetramer. These channels appear to be accessible to bulk solvent and vary in width from 5-20 Å [17].
Which β-galactosidase and catechol 2,3 dioxygenase missense mutants were rescuable by salt can also be explained by considering the structures of the proteins. 13 out of the 19, or 68.42% of the missense mutants that were suppressible contained amino acid changes that were located on the surface of the proteins and thus would be expected to be more rescuable by salt. For β-galactosidase in which 11 out of 15, or 73.33% of the missense mutants were suppressible by salt, the only missense mutants that were not suppressible affected amino acids that were buried in the protein.
Several studies have demonstrated the propensity of Hofmeister salts to stabilize proteins via hydrophobic interactions [24-26] and that small cations and large anions facilitate this effect [27]. According to these findings NaCl, which has a smaller cation with an ionic radius of 0.102 nM and a larger anion with an ionic radius of 0.181 nM, would be expected to rescue inactive missense mutants where the mutated amino acid had a hydrophobic shift to a greater degree than missense mutants where the mutated amino acid had a hydrophilic shift. Of the 42 β-galactosidase and catechol 2,3 dioxygenase missense mutants that were characterized in this study, 14 of the mutants had an amino acid change that resulted in a hydrophobic shift and 28 of the missense mutants had an amino acid change that resulted in a hydrophilic shift. A total of 8 out of the 14, or 57.14% of the mutants that resulted in a hydrophobic shift were rescuable by NaCl, while only 12 out of 28, or 42.86% of the mutants that resulted in a hydrophilic shift were rescuable by NaCl. Thus there was a slight preference for NaCl to rescue mutants that resulted in a hydrophobic shift than to rescue mutants that resulted in a hydrophilic shift.
We further characterized the ability of the different salts to correct the lacZ39 missense mutant that was highly suppressible by salt. Na2SO4 was the most effective salt for stabilizing β-galactosidase from lacZ39, followed by NaCl, NH4Cl, (NH4)2SO4, NaCH3CO2, Na2HPO4 and (NH4)2HPO4. Only NH4CH3CO2 could not stabilize the β-galactosidase from lacZ39. Interestingly, unlike wild-type β-galactosidase where NaCl was not nearly as effective as Na2SO4, NaCl was almost as effective as Na2SO4 at stabilizing the β-galactosidase produced by lacZ39. This finding is consistent with NaCl being the choice of most researchers that have tested the ability of salts to rescue nonfunctional missense mutants.
Historically, most researchers have used NaCl to form the salt gradients in the anionic exchange or hydroxyapatite columns used in fast protein liquid chromatography (FPLC) to purify proteins and the resulting proteins are stored in buffers containing NaCl. The results of our intracellular studies and the numerous in vitro studies by other researchers that have investigated the ability of salts to stabilize proteins, suggest that Na2SO4 and Na2HPO4 would be better choices for stabilizing proteins.
Disclosure of conflict of interest
None.
References
1. Hofmeister F. About regularities in the protein precipitating effects of salts and the relation of these effects with the physiological behavior of salts. Arch Exp Pathol Pharmakol 1888; 24:247–260. [Google Scholar]
2. Jungwirth P, Cremer PS. Beyond Hofmeister. Nat Chem. 2014;6:261–263. [PubMed] [Google Scholar]
3. Salis A, Ninham BW. Models and mechanisms of Hofmeister effects in electrolyte solutions, and colloid and protein systems revisited. Chem Soc Rev. 2014;43:7358–7377. [PubMed] [Google Scholar]
4. Kunz W, Henle J, Ninham BW. ‘Zur lehre von der wirkung der salze’ (about the science of the effect of salts): Franz Hofmeister’s historical papers. Curr Opin Colloid Interface Sci. 2004;9:19–37. [Google Scholar]
5. Von Hippel PH, Wong KY. The effect of ions on the kinetics of formation and the stability of the collagenfold. Biochemistry. 1962;1:664–674. [PubMed] [Google Scholar]
6. Von Hippel PH, Wong KY. The collagen gelatin phase transition. I. Further studies of the effects of solvent environment and polypeptide chain composition. Biochemistry. 1963;2:1387–1398. [PubMed] [Google Scholar]
7. Von Hippel PH, Wong KY. Neutral salts: the generality of their effects on the stability of macromolecular conformations. Science. 1964;145:577–580. [PubMed] [Google Scholar]
8. Nandi PK, Robinson DR. The effects of salts on the free energy of the peptide group. J Am Chem Soc. 1972;94:1299–1308. [PubMed] [Google Scholar]
9. Nandi PK, Robinson DR. The effects of salts on the free energies of nonpolar groups in model peptides. J Am Chem Soc. 1972;94:1308–1315. [PubMed] [Google Scholar]
10. Pegram LM, Wendorff T, Erdmann R, Shkel I, Bellissmo D, Felitsky DJ, Record MT Jr. Why Hofmeister effects of many salts favor protein folding but not DNA helix formation. Proc Natl Acad Sci U S A. 2010;107:7716–7721. [PMC free article] [PubMed] [Google Scholar]
11. Hawthorne DC, Friis J. Osmotic-remedial mutants, a new classification for nutritional mutants in yeast. Genetics. 1964;50:829–839. [PMC free article] [PubMed] [Google Scholar]
12. Good CM, Pattee PA. Temperature-sensitive osmotically fragile mutants of staphylococcus aureus. J Bacteriol. 1970;104:1401–1403. [PMC free article] [PubMed] [Google Scholar]
13. Russell RR. Temperature-sensitive osmotic remedial mutants of Escherichia coli. J Bacteriol. 1972;112:661–665. [PMC free article] [PubMed] [Google Scholar]
14. Bilsky AZ, Armstrong JB. Osmotic reversal of temperature sensitivity in Escherichia coli. J Bacteriol. 1973;113:76–81. [PMC free article] [PubMed] [Google Scholar]
15. Kohno T, Roth J. Electrolyte effects on the activity of mutant enzymes in vivo and in vitro. Biochemistry. 1979;18:1386–1392. [PubMed] [Google Scholar]
16. Kita A, Kita S, Fujisawa I, Inaka K, Ishida T, Horiike K, Nozaki M, Miki K. An archetypical extradiol-cleaving catecholic dioxygenase: the crystal structure of catechol 2,3-dioxygenase (metapyrocatechase) from pseudomonas putida mt-2. Structure. 1999;7:25–34. [PubMed] [Google Scholar]
17. Juers DH, Jacobson RH, Wigley D, Zhang XJ, Huber RE, Tronrud DE, Matthews BW. High resolution refinement of β-galactosidase in a new crystal form reveals multiple metal-binding sites and provides a structural basis for α-complementation. Protein Sci. 2000;9:1685–1699. [PMC free article] [PubMed] [Google Scholar]
18. Miller JH. Experiments in molecular genetics. Cold Spring Harbor Laboratory. 1972 [Google Scholar]
19. Cole AE, Hani FM, Altman R, Meservy M, Roth JR, Altman E. The promiscuous sumA missense suppressor from salmonella enterica has an intriguing mechanism of action. Genetics. 2017;205:577–588. [PMC free article] [PubMed] [Google Scholar]
20. Cole AE, Hani FM, Allen BW, Kline PC, Altman E. Nonfunctional missense mutants in two well characterized cytosolic enzymes reveal important information about protein structure and function. Protein J. 2018;37:407–427. [PubMed] [Google Scholar]
21. Bertani G. Studies on lysogenesis. I. The mode of phage liberation by lysogenic Escherichia coli. J Bacteriol. 1951;62:293–300. [PMC free article] [PubMed] [Google Scholar]
22. Zimbro MJ, Power DA, Miller SM, Wilson GE, Johnson JA. Difco & BBL manual: manual of microbiological culture media, second edition. BD Diagnostics - Diagnostic Systems, Sparks, MD. 2009 [Google Scholar]
23. Jacobson RH, Zhang XJ, DuBose RF, Matthews BW. Three-dimensional structure of β-galactosidase from E. coli. Nature. 1994;369:761–766. [PubMed] [Google Scholar]
24. Thomas AS, Elco*ck AH. Molecular dynamics simulations of hydrophobic associations in aqueous salt solutions indicate a connection between water hydrogen bonding and the Hofmeister effect. J Am Chem Soc. 2007;129:14887–14898. [PubMed] [Google Scholar]
25. Geisler M, Pirzer T, Ackerschott C, Lud S, Garrido J, Scheibel T, Hugel T. Hydrophobic and Hofmeister effects on the adhesion of spider silk proteins onto solid substrates: an AFM-based single-molecule study. Langmuir. 2008;24:1350–1355. [PubMed] [Google Scholar]
26. Tadeo X, López-Méndez B, Castaño D, Trigueros T, Millet O. Protein stabilization and the Hofmeister effect: the role of hydrophobic solvation. Biophys J. 2009;97:2595–2603. [PMC free article] [PubMed] [Google Scholar]
27. Schwierz N, Horinek D, Netz RR. Anionic and cationic Hofmeister effects on hydrophobic and hydrophilic surfaces. Langmuir. 2013;29:2602–2614. [PubMed] [Google Scholar]
Articles from International Journal of Biochemistry and Molecular Biology are provided here courtesy of e-Century Publishing Corporation
