MPKaDB: A p K a Database for Exploring pH Dependence in Membrane Proteins

MPKaDB: A pKa Database for Exploring pH Dependence in Membrane Proteins | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results MPKaDB: A p K a Database for Exploring pH Dependence in Membrane Proteins Jiahao He , Yansheng Chen , Jinxi Wu , Zhitao Cai , Wenting Jia , Qiuchen Yue , View ORCID Profile Yandong Huang doi: https://doi.org/10.1101/2025.11.14.688495 Jiahao He a College of Computer Engineering, Jimei University , Xiamen 361021, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Yansheng Chen b College of Science, Jimei University , Xiamen 361021, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Jinxi Wu a College of Computer Engineering, Jimei University , Xiamen 361021, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Zhitao Cai a College of Computer Engineering, Jimei University , Xiamen 361021, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Wenting Jia a College of Computer Engineering, Jimei University , Xiamen 361021, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Qiuchen Yue a College of Computer Engineering, Jimei University , Xiamen 361021, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site Yandong Huang a College of Computer Engineering, Jimei University , Xiamen 361021, China Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Yandong Huang For correspondence: yandonghuang{at}jmu.edu.cn Abstract Full Text Info/History Metrics Supplementary material Data/Code Preview PDF Abstract The biological activities of many membrane proteins are pH-regulated, yet mapping their pH dependence experimentally is slow and expensive. In this work, we present MPKaDB ( http://computbiophys.com/DeepKa/mpkadb ), a comprehensive p K a database for membrane proteins that instantly decode the protonation states of ionizable residues under a specified pH. Leveraging MPKaDB, we performed pH-coupled electrostatic characterization of trans-membrane proteins. To facilitate use, a user-friendly search engine was developed to retrieve a protein of interest and returns its p K a values, isoelectric points of both cytoplasmic and extra-cytoplasmic faces, and an automated screening of active-site residues. In the end, two case studies were proposed to demonstrate how p K a ’s from MPKaDB could be applied to explore the pH-dependent relationship between membrane protein structure and function. Download figure Open in new tab 1. Introduction pH plays a central role in many biological processes, including transports of ions or small molecules across the cellular membrane by membrane-embedded proteins, such as sodium-proton antiporters [ 1 ], proton channels [ 2 ], glucose-proton symporters [ 3 ], and cholesterol transporters [ 4 ]. Normally, such a proton-coupled mass transport event occurs under a specific pH window, exhibiting high pH dependence. For instance, the secondary sodium-proton antiporter NhaA of Escherichia coli functions only when the pH is higher than 6.5 and the reactivity reaches the maximal around pH 8.5 [ 5 ]. Thus, to understand the underlying mechanism, it is essential to identify the proton carriers according to the p K a values of ionizable residues in the reaction core [ 6 ]. Above sodium transports are driven by the pH gradient maintained primarily by proton channels, activities of which are governed by pH too [ 2 ]. Beyond screening the candidates of proton carriers, an accurate description of the salt-bridge network lining the proton-conducting path is equally critical, and this also hinges on reliable p K a calculations of ionizable residues that involve in salt bridging [ 7 ]. More than ten thousand crystal structures of transmembrane proteins have been experimentally determined. Based on the average occurrence (Fig. S1A) and transmembrane-centre likelihood (Fig S1B) of each ionisable residue type determined from about one thousand non-redundant transmembrane proteins in the OPM database [ 8 ], ionizable residues are essential constituents of reaction cores typically situated in the middle of the membrane and their titration behaviors determined by p K a ’s should be considered explicitly, especially those carried out near physiological pH. In addition, there is a considerable high probability of encountering an acid and a base simultaneously (FIg. S1C), implying the importance of coupling between ionizable residues. Evidently, p K a values are indispensable for linking protein function to pH. However, experimental p K a values for membrane proteins remain scarce, probably due to pH-triggered conformational changes during the transport cycle. For example, it was found that the inward-facing gate of NhaA opens in response to the deprotonation of an essential Asp residue [ 6 ], providing a direct connection between side-chain protonation state and large-scale conformational switching. Thus, accurate p K a predictions in theory become urgent. Constant pH molecular dynamics (CpHMD) simulations have been demonstrated applicable to transmembrane proteins [ 6 ]. In particular, CpHMD simulations explicitly couple changes in protonation state to conformational rearrangements, enabling direct estimation of macroscopic p K a ’s within a lipid-bilayer environment. However, CpHMD simulations are time demanding and therefore impractical for the high-throughput p K a screens demanded by industrial work flows. Alternatively, DeepKa [ 9 ], a deep learning-based scheme, has been applied to probing the influence of conformational changes sampled by MD simulations on p K a ’s and the resulting deprotonation equilibria [ 10 , 11 ]. Notably, ionizable residues in the active sites are often deeply buried within the proteins and therefore hidden from lipids. As a result, the environment that an ionizable side chain senses would still be dominated by the protein, which to some extent explains the feasibility of DeepKa working on trans-membrane proteins. Recently, performances of several fast p K a predictors, including the artificial intelligence (AI)-based DeepKa [ 12 ], the empirical formula Propka [ 13 ] and Poisson-Boltzmann (PB) equation-based H++ [ 14 ], on the outer membrane porin F (OmpF) from Escherichia coli , a trimeric channel, was evaluated using CpHMD as benchmark [ 15 ]. It was found that DeepKa outperforms the widely used PropKa and H++, in consistence with the benchmark by Wei et al. on p K a ’s of ionizable residues that belong to water-soluble proteins [ 16 ]. In this work, the p K a database MPKaDB that contains p K a data of 3.3 million Asp, Glu, His and Lys residues in 12 k membrane proteins deposited in the protein data bank (PDB) was created by DeepKa [ 9 ]. The database is freely downloadable from the web page of MPKaDB ( http://computbiophys.com/DeepKa/mpkadb ) or GitLab ( https://gitlab.com/computational-biophysics/deepkaserver/-/tree/main/DeepKaDB ). Additionally, a user-friendly interface is available for searching membrane proteins of special interest via PDB codes. Notably, the MPKaDB web page has been integrated into the existing DeepKa web server [ 10 ]. In the result section, electrostatics of transmembrane proteins were characterized based on charges determined by p K a values and pH. Besides, two case studies were set out to demonstrate how the p K a database could be applied to preliminarily exploring the pH-governed mechanisms of transmembrane proteins. 2. Materials and Methods 2.1. Data Collection, Annotation and Access The present MPKaDB contains p K a values estimated by DeepKa for 3,306,058 Asp, Glu, His, and Lys residues in 12,602 membrane proteins from the PDB. Cd-hit [ 17 ] combined with BLAST+ [ 18 ] was utilized to screen non-homogeneous proteins from eight thousand transmembrane proteins in OPM. The sequence identity threshold of 30 percent was applied. As a result, a cluster of 968 non-homogeneous transmembrane proteins was obtained to characterize titration-associated electrostatics of transmembrane proteins. All p K a ’s of MPKaDB were saved in a CSV file that contains five columns where the first four together locate a residue and the last is the p K a value. The CSV file can be downloaded freely from the MPKaDB page of the DeepKa web server or our GitLab page. Apart from the one-click downloading of the entire database. The MPKaDB page also contains a search engine. To search a membrane protein of interest, users are supposed to enter a valid PDB code. If the membrane protein is not available, users are recommended to use the Prediction page where both PDB codes and files are acceptable inputs [ 10 ]. 2.2 Isoelectric Point (pI) Calculation pI is the pH value at which a molecule carries zero net charge. Charges of Asp/Glu/Cys/Tyr ( q a ) and His/Lys/Arg ( q b ) residues that vary as a function of pH were calculated by Eq. 1 and Eq. 2 , respectively, equivalent to the Henderson-Hasselbalch (HH) formula [ 19 ]. Here the reference p K a ’s measured in aqueous solvent, 8.55, 9.84 and 12.0, were utilized for Tyr, Cys and Arg, respectively, for pI calculations [ 20 ]. On the other hand, to compute the p K a shifting in response to protein environment, reference p K a values of 3.67, 4.25, 6.54 and 10.4 were applied to Asp, Glu, His and Lys, respectively [ 20 ]. In addition to pI, such pH-dependent charge calculations will be utilized to the membrane-associated statistics of electrostatics. 2.3. Molecular Dynamics (MD) Simulation All-atom MD simulations of a glucose-proton symporter (GlcP Se , PDB code: 4LDS) [ 3 ], a proton pump (NsXeR, PDB code 6EYU) [ 21 ] and an anion channel (iC++, PDB code 6CSM) [ 22 ] were set up to investigate the coupling between protonation state and electrostatics in pores. Details about the system preparation and simulation protocols can be found in the Supporting Information. Finally, protein structures were visualized by VMD [ 23 ] and analyzed by MDAnalysis [ 24 ]. 3. Results and Discussion 3.1. MPKaDB Web Server Description The web server provides two core functions, a one-click button to download the entire MPKaDB database as a ZIP-compressed archive, and a search engine that retrieves any entry via a valid PDB code ( Fig. 1A ). If the protein of interest is found, a clickable arrow directs the user to its Result page that contains two columns. As shown in Fig. 1B , the left column includes a link to download the PDB file. Below are the pI Cyto (z≤0) and pI Extr (z>0) that correspond to cytoplasmic (Cyto) and extra-cytoplasmic (Extr) domains, respectively, and the total charges varying as the pH are visualized. Download figure Open in new tab Figure 1: MPKaDB web page and downstream analyses. MPKaDB (A) front and (B) result page. (C) Region definitions across the membrane. The protein is displayed with the surface mode. Lipids are displayed with lines and P atoms at lipid heads are highlighted with orange spheres. The origin was marked by a open red circle. (D) Mean (red) and standard deviation (cyan) of net charge as a function of z coordinate. (E) Correlation plots between lateral surface areas ( Å 1 ) and mean net charges as well as regression fitting lines for Extr (gray and black) and Cyto (blue and orange), respectively. The correlation coefficients are 0.54 and −0.24 for Cyto and Extr, respectively. (F) Mean net charge as a function of pH at Extr (blue) and Cyto (orange), respectively. (G) pI distributions at Extr (blue) and Cyto (orange), respectively. (H) p K a shift distributions at five regions, Cyto (black), Cyto leaflet (blue), Center (red), Extr leaflet (green) and Extr (gray), respectively, across the membrane. The right column contains the other link to download the p K a data as a CSV le. As one can see, p K a values are enumerated on the web page as well. Residues highlighted in blue are screened by the active-site residue criteria that the p K a value ranges from 5 to 9 and simultaneously the z coordinate of the C α atom is within the central region of the membrane (Center) as shown in Fig. 1C . Specifically, for each transmembrane protein in the OPM database, the membrane thickness (T, around 30 Å) was used to define the outer limits of the Cyto and Extr leaflets, −T/2 and T/2, respectively. The distance from the outer boundary of either leaflet towards the Center was set to 10 Å. As a result, the scope of Center is T-20 Å. 3.2. Electrostatic Characterization of Transmembrane proteins The positive-inside rule, a fundamental phenotype of membrane asymmetry [ 25 ], for transmembrane proteins has long been an consensus [ 26 , 28 ] and is critical to membrane-protein topology [ 27 ]. Previous studies, however, simply assumed Asp/Glu and Lys/Arg to be negatively and positively charged, respectively, without estimating pH-associated charges. Now that p K a values in MPKaDB enable the charge of an ionizable residue to vary as a function of pH according to the HH equation, in this work electrostatic properties across the membrane were revisited based on the screened set of non-homogeneous transmembrane proteins. As shown in Fig. 1D , the peak positive net charge at the physiological pH seven is −24 Å, about 10 Å away from the count maximal of Arg or Lys nearby the Cyto leaflet [ 26 , 28 ]. Interestingly, negative net charge, though weak, at the distance of 15 Å from the Extr leaflet is observed, consistent with the statistics by Baker et al. [ 28 ]. The biological relevance of the positions for positive and negative net charges in average at two sides of the membrane are not clear. Given protein structures, the correlation between protein lateral surface area and the net charge in ranges of −T/2 ± 10 for Cyto and T/2 ± 10 for Extr. As illustrated in Fig. 1E , the absolute net charge scales linearly with the lateral surface area on both the Cyto and Extr, revealing, albeit modest, an interplay between molecular size and net charge. Furthermore, the mean net charges at different pH conditions were computed and shown in Fig. 1F , from which one can tell that the averaged pI values for Extr and Cyto are 6.0 and 8.5, respectively. Here we assume that charges at Cyto could be fully isolated from those at Extr by the low-dielectric membrane, which leads to two pI values, one for Cyto and the other for Extr. pI values for individual proteins were computed too. As illustrated in Fig. 1G , The pI distribution for Cyto is more concentrated when compared with that for Extr, exhibiting high certainty of the positive charge preference at the cytoplasmic side. p K a ’s are the basis of aforementioned analyses. To examine the p K a shifting in response to the membrane, distributions of p K a deviations from reference p K a ’s, namely p K a shifts, were computed in five regions over the membrane normal. From Fig. 1H , it is evident that large p K a shifts are more populated while approaching the Center region. For one thing, an ionizable residue prefers the neutral state when buried, which would explain large p K a shifts. For another, the likelihood (Fig. S1) of finding only Asp/Glu (23.2%) in the Center region is much higher than that for Lys/Arg (9.1%), which would further explain the bias of p K a shifting up for a buried Asp/Glu residue. Such a membrane-coupled p K a shifting highlights the necessity of taking titration behaviors into account while investigating electrostatic properties of transmembrane proteins. 3.3. Case Study I: Identification of Proton Carriers in Sodium-Proton Antiporters Sodium-proton antiporters are well known by pH-regulated translocation of ions. Structures of electrogenic NhaA in Escherichia coli and NapA from Thermus thermophilushave that share the same fold type, characterized by two anti-parallel discontinuous helices crossing over in core domains, have been extensively studied. As collected in Tab. 3.3, eight and three crystal structures are available for NhaA and NapA, respectively. Although the experimental pH scans from the inactive 4.0 to the optimum 8.5, all existing crystal structures of NhaA should correspond the resting inward-facing (IF) state, judging from radius of gyrations for both IF and outward-facing (OF) gates (Tab. S1). In contrast, either the IF or OF structure was obtained under the optimal pH of 8.0 for NapA. Either NhaA or NapA transports one sodium by exchanging two protons. However, the identity of the two residues that carry the protons remains disputed. To address this question, p K a ’s of ionizable residues in the active site for these structures were extracted from MPKaDB. Taking NhaA as an example, Asp163, Asp164, and Lys300 buried in the reaction center (Tab. S1) have been proposed as candidate proton carriers. Asp164 as one proton carrier has been a consensus, thus the controversy focuses on Asp163 and Lys300. In general, p K a ’s for Asp163 and Asp164 are higher than those reported from CpHMD simulations, most notably for Asp163, which likely reflects insufficient conformational sampling of the allosteric motions unique to NhaA [ 6 , 10 , 11 ]. Notably, we found that p K a of Asp164 is correlated to the solvent accessible surface area (SASA) of the carboxyl oxygens on its side chain (Fig. S2), which implies that the titration of Asp164 could be dominated by hydrophobic forces. As to Lys300, all p K a ’s (Tab. 3.3) cluster at or just above the reference value of 10.4, except for the 1ZCD structure determined at the inactive pH of 4.0, where the loss of the Asp163–Lys300 salt bridge (Fig. S3A) shifts the p K a downward (Fig. S3B). A more thorough discussion of this divergence can be found in two previous studies [ 6 , 10 ]. Regarding Asp133 also buried in the reaction center, we found that p K a values are all lower than 5.0, well below the functional pH optimum of 8.5 for NhaA (8.0 for NapA), in line with previous CpHMD data on NhaA [ 6 ], except for structure 7A0X, which was determined at pH 6.0 (Tab. 3.3). Asp133 side chain has been suggested to precisely neutralize the positive charge that accumulates between two helical termini (Fig. S4A) [ 29 ]. We note that Asp133 is also surrounded by three hydrophobic residues, namely Lue18, Phe136 and Phe339, which would induce p K a to shift up (Fig. S4B), allowing Asp133 to carry a proton, supporting Asp133 as an alternative proton-binding site proposed by Shen group [ 36 ]. Thus, p K a values curated in MPKaDB support Asp163, Asp164 and Lys300 in NhaA as viable proton-carrier candidates and Asp133 remains a plausible backup. p K a ’s of counter-residues in NapA, namely Glu333, Asp156, Asp157 and Lys305, that correspond to Asp133, Asp163, Asp164 and Lys300, respectively, are also provided. Notably, the p K a values of the three Asp residues are modestly higher in OF, in line with NhaA [ 11 ], whereas that of Lys300 is slightly lower, compared with IF. Finally, based on the mean p K a values (Tab. 3.3), the same ensemble of candidate proton carriers identified for NhaA could be applied to NapA as well. View this table: View inline View popup Download powerpoint Table 1: p K a ’s of Na + /H + antiporters NhaA and NapA deposited in PDB 3.4. Case Study II: Coupling between Protonation State and Salt Bridging in Transmembrane Proteins In this case study, p K a ’s (Tab. S2) combined with MD simulations were employed to probe how protonation states modulate the salt-bridge networks that line mass-transport pathways. Here three transmembrane proteins named GlcP Se [ 3 ], NsXeR [ 21 ] and iC++ [ 22 ] were selected based on the active-site residue criteria above. As illustrated in Fig. 2 , residues that satisfy the criteria above are Asp22 in GlcP Se , Asp76 in NsXeR and Glu68/Asp234 in iC++. Notably, the calculated p K a of Asp22 (Tab. S pka of three acids) coincides with the pH optimum 5.5, identifying this residue as the most probable H + -binding site of the proton pump GlcP Se [ 3 ]. Download figure Open in new tab Figure 2: Crystal structures and salt bridge networks of three transmembrane proteins. Crystal structures are visualized with the new cartoon model and chains A, B and C are colored gray, blue and red, respectively. Acids screened are displayed with green or red spheres. Residues involved in salt bridge networks are displayed with sticks where carbon, oxygen and nitrogen atoms are colored white, red and blue, respectively. Side and top views of crystal structures and the comparisons of salt bridge networks in pores between crystallized and MD simulated structures for proteins (A) GlcP Se , (B) NsXeR and (C) iC++, respectively. The orange arrow represents the membrane normal. Dashed red lines represent salt bridges. Symbols - and 0 in parentheses indicate deprotonated and protonated states of Asp or Glu, respectively. Judging from the crystal structures, Asp22 of GlcP Se ( Fig. 2A ) and Asp76 of NsXeR ( Fig. 2B ) are in contact with the positively charged Arg102 and Lys213, respectively, and therefore are likely to be charged or deprotonated (Tab. S3). In contrast, Glu68 of iC++ remains too far from Lys238 to form a salt bridge ( Fig. 2C ), and thus would take the neutral or protonated state (Tab. S3). Noting that these initial guesses were subsequently validated by fixed-charge MD simulations (Fig. S5-10), it is of interest to see what would happen if the protonation states adopted by crystal structures were changed. As illustrated in Fig. 2A , neutralization of Asp22 completely breaks the Asp22-Arg102 salt bridge (Fig. S5) and swings the Arg102 side chain toward the tunnel entrance. On the other hand, as shown in Fig. 2B , the Asp76-Lys213 salt bridge is disrupted when Asp76 is neutralized (Fig. S6). Unlike GlcP Se , Lys213 does not turn away and would still interact with Asp76. The salt-bridge networks above are simple when compared with that found in iC++. As illustrated in Fig. 2C , when Glu68 is charged, it forms a stable salt bridge with Lys238 (Fig. S7), which nevertheless does not affect the Lys238-Asp234 salt bridge (Fig. S8). Interestingly, the Arg94-Asp223 salt bridge nearby the entrance is disrupted (Fig. S9). Arg94 turns to forms a new salt bridge with Asp234 (Fig. S10), which affects the stability of the Lys238-Asp234 salt bridge (Fig. S11). Intriguingly, though Glu68 is distant from Asp223, the Asp223-Arg94 salt bridge is governed by the titration of Glu68. Such a long-distance coupling between ionizable residues was also observed in NhaA by Henderson et al. [ 36 ]. When Glu68 is charged, the net charge at the Center becomes negative, which might explain the attraction of Asp94 towards the Center to neutralize the charge and therefore preserve electrostatic equilibrium. 4. Conclusions and Perspectives Here we present MPKaDB, a p K a database dedicated exclusively to membrane proteins. The current release catalogs 3.3 million Asp, Glu, His and Lys residues from 12,006 membrane protein structures deposited in the PDB. The complete database is available for free download as a single compressed archive from the MPKaDB website. A user-friendly protein search engine has also been implemented where users can instantly retrieve any entry by entering its PDB code. Beyond p K a values, the result page delivers downstream annotations, including an initial screening of potential active-site residues and the isoelectric points calculated separately for the cytoplasmic and extra-cytoplasmic faces. Finally, MPKaDB will be updated in parallel with the PDB repository. Equipped with p K a data, we dissected the electrostatic properties of membrane proteins, including the well-known positive-inside rule, with pH-resolved detail. Two case studies further illustrate how the database could be used, in conjunction with pH, to probe membrane protein structure–function relationships. In the first case study, by integrating PDB structures with in silico screening for potential active-site residues, we evaluate the likelihood of individual residues serving as proton carriers in two sodium – proton antiporters. Whereas, in the second case study, MD simulations were used to probe how changing protonation states modulate the salt-bridge networks at protein centers. Notably, we revealed a long-distance coupling between two acidic residues, which is hardly captured by a static PDB structure. The current MPKaDB omits Cys, Tyr, and Arg because DeepKa intrinsically cannot predict their p K a values, asking for the development of a general-purpose model in the future. However experimental or theoretical p K a data for the three reidue types are extremely scarce [ 37 ], which poses a major challenge for model development in future. Besides, DeepKa was trained on water-soluble proteins. Thus, residues located nearby the water-lipid interface should be warranted, as DeepKa treats the lipid phase as water, which would underestimate p K a shifts. Finally, pI values at cytoplasmic and extra-cytoplasmic faces were computed separately, which asks for precedent spatial arrangements of membrane proteins with respect to the hydrocarbon core of the lipid bilayer from the third-part resources, like OPM and PDBTM [ 38 ]. Thus, the pI field on the Result page will remain empty if the protein is unavailable in either OPM or PDBTM. 5. Funding This work is supported by the Natural Science Foundation of Fujian Province, China (2023J01329). 6. CRediT authorship contribution statement Yandong Huang: Writing, Supervision, Project administration, Conceptualization, Validation, Methodology. Jiahao He, Yansheng Chen, Jinxi Wu: Software, Resources, Methodology. Zhitao Cai: Validation, Methodology. Wenting Jia and Qiuchen Yue: Software. 7. DATA AVAILABILITY The database is freely available as a web server at http://computbiophys.com/DeepKa/mpkadb or GitLab at https://gitlab.com/computational-biophysics/deepkaserver/-/tree/main/DeepKaDB . 8. DECLARATION OF COMPETING INTEREST The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Appendix A. Supplemental data Supplementary material to this article can be found online. Funder Information Declared Natural Science Foundation of Fujian Province, China , 2023J01329 Footnotes http://computbiophys.com/DeepKa/mpkadb https://gitlab.com/computational-biophysics/deepkaserver/-/tree/main/DeepKaDB References [1]. ↵ Yeo , H. , Mehta , V. , Gulati , A. , Drew , D. , ( 2023 ). Structure and electromechanical coupling of a voltage-gated Na + /H + exchanger . Nature 623 , 193 – 201 . OpenUrl PubMed [2]. ↵ Ramsey , I.S. , Moran , M.M. , Chong , J.A. , Clapham , D.E. , ( 2006 ). A voltage-gated proton-selective channel lacking the pore domain . Nature 440 , 1213 – 1216 . OpenUrl CrossRef PubMed Web of Science [3]. ↵ Iancu , C.V. , Zamoon , J. , Woo , S.B. , Aleshin , A. , Choe , J.Y. , ( 2013 ). Crystal structure of a glucose/H + symporter and its mechanism of action . Proc. Natl. Acad. Sci. U.S.A . 110 , 17862 – 17867 . OpenUrl Abstract / FREE Full Text [4]. ↵ Qian , H. , Wu , X. , Du , X. , Yao , X. , Zhao , X. , Lee , J. , Yang , H. , Yan , N. , ( 2020 ). Structural Basis of Low-pH-Dependent Lysosomal Cholesterol Egress by NPC1 and NPC2 . Cell 182 , 98 – 111 . OpenUrl CrossRef PubMed [5]. ↵ Taglicht , D. , Padan , E. , Schuldiner , S. , ( 1991 ). Overproduction and puri cation of a functional Na + /H + antiporter coded by nhaA (ant) from Escherichia coli . J. Biol. Chem . 266 , 11289 – 11294 . OpenUrl Abstract / FREE Full Text [6]. ↵ Huang , Y. , Chen , W. , Dotson , D.L. , Beckstein , O. , Shen , J. , ( 2016 ). Mechanism of pH-dependent activation of the sodium-proton antiporter NhaA . Nat. Commun . 7 , 12940 . OpenUrl CrossRef PubMed [7]. ↵ Shen , M. , Huang , Y. , Cai , Z. , Cherny , V.V. , Decoursey , T.E. , Shen , J. , ( 2024 ). Interior pH-sensing residue of human voltage-gated proton channel Hv1 is histidine 168 . Biophys. J . 123 , 4211 – 4220 . OpenUrl PubMed [8]. ↵ Lomize , M.A. , Lomize , A.L. , Pogozheva , I.D. , Mosberg , H.I. , ( 2006 ). OPM: Orientations of proteins in membranes database . Bioinformatics 22 , 623 – 625 . OpenUrl CrossRef PubMed Web of Science [9]. ↵ Cai , Z. , Liu , T. , Lin , Q. , He , J. , Lei , X. , Luo , F. , Huang , Y. , ( 2023 ). Basis for Accurate Protein pK a Prediction with Machine Learning . J. Chem. Inf. Model . 63 , 2936 – 2947 . OpenUrl CrossRef PubMed [10]. ↵ Cai , Z. , Peng , H. , Sun , S. , He , J. , Luo , F. , Huang , Y. , ( 2024 ). DeepKa Web Server: High-Throughput Protein pK a Prediction . J. Chem. Inf. Model . 64 , 2933 – 2940 . OpenUrl CrossRef PubMed [11]. ↵ Xie , T. , He , J. , Sun , S. , Chen , Y. , Huang , J. , Huang , Y. , ( 2025 ). Molecular mechanism of Na + /H + antiporting in NhaA.J . Chem. Theory Comput . 21 , 9107 – 9119 . OpenUrl [12]. ↵ Cai , Z. , Luo , F. , Wang , Y. , Li , E. , Huang , Y. , ( 2021 ). Protein pK a Prediction with Machine Learning . ACS Omega 6 , 34823 – 34831 . OpenUrl PubMed [13]. ↵ Olsson , M.H.M. , Søndergaard , C.R. , Rostkowski , M. , Jensen , J.H. , ( 2011 ). PROPKA3: Consistent Treatment of Internal and Surface Residues in Empirical pK a Predictions . J. Chem. Theory Comput . 7 , 525 – 537 . OpenUrl CrossRef PubMed [14]. ↵ Anandakrishnan , R. , Aguilar , B. , Onufriev , A.V. , ( 2012 ). H++ 3.0: automating pK a prediction and the preparation of biomolecular structures for atomistic molecular modeling and simulations . Nucleic Acids Res . 40 , W537 – W541 . OpenUrl CrossRef PubMed Web of Science [15]. ↵ Tavares-Neto , E. , Aguilella-Arzo , M. , Aguilella , V.M. , ( 2025 ). Predicting residue ionization of OmpF channel using Constant pH Molecular Dynamics as benchmark . PLoS Comput. Biol . 21 , e1013628 . OpenUrl PubMed [16]. ↵ Wei , W. , Hogues , H. , Sulea , T. , ( 2023 ). Comparative Performance of High-Throughput Methods for Protein pK a Predictions . J. Chem. Inf. Model . 63 , 5169 – 5181 . OpenUrl CrossRef PubMed [17]. ↵ Li , W. , Godzik , A. , ( 2006 ). Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences . Bioinformatics 22 , 1658 – 1659 . OpenUrl CrossRef PubMed Web of Science [18]. ↵ Camacho , C. , Coulouris , G. , Avagyan , V. , Ma , N. , Papadopoulos , J. , Bealer , K. , Madden , T.L. , ( 2009 ). BLAST+: architecture and applications . BMC Bioinformatics 10 , 421 . OpenUrl CrossRef PubMed [19]. ↵ Lee , M.S. , Salsbury , F.R. Jr . , Brooks , C.L. III . ., ( 2004 ). Constant-pH molecular dynamics using continuous titration coordinates . Proteins 56 , 738 – 752 . OpenUrl CrossRef PubMed Web of Science [20]. ↵ Thurlkill , R.L. , Grimsley , G.R. , Scholtz , J.M. , Pace , C.N. , ( 2006 ). pK a values of the ionizable groups of proteins . Protein Sci . 15 , 1214 – 1218 . OpenUrl CrossRef PubMed Web of Science [21]. ↵ Shevchenko , V. , Mager , T. , Kovalev , K. , Polovinkin , V. , Alekseev , A. , Juettner , J. , Chizhov , I. , Bamann , C. , et al. , ( 2017 ). Inward H + pump xenorhodopsin: Mechanism and alternative optogenetic approach . Sci. Adv . 3 , e1603187 . OpenUrl FREE Full Text [22]. ↵ Kato , H.E. , Kim , Y.S. , Paggi , J.M. , Evans , K.E. , Allen , W.E. , Richardson , C. , Inoue , K. , Ito , S. , et al. , ( 2018 ). Structural mechanisms of selectivity and gating in anion channelrhodopsins . Nature 561 , 349 – 354 . OpenUrl CrossRef PubMed [23]. ↵ Humphrey , W. , Dalke , A. , Schulten , K. , ( 1996 ). VMD: Visual molecular dynamics . J. Mol. Graph . 14 , 33 – 38 . OpenUrl CrossRef PubMed Web of Science [24]. ↵ Michaud-Agrawal , N. , Denning , E.J. , Woolf , T.B. , Beckstein , O. , ( 2011 ). MDAnalysis: A toolkit for the analysis of molecular dynamics simulations . J. Comput. Chem . 32 , 2319 – 2327 . OpenUrl CrossRef PubMed [25]. ↵ Pabst , G. , Keller , S. , ( 2024 ). Exploring membrane asymmetry and its e ects on membrane proteins . Trends Biochem. Sci . 49 , 333 – 345 . OpenUrl CrossRef PubMed [26]. ↵ Juretic , D. , Zoranic , L. , Zucic , D. , ( 2002 ). Basic charge clusters and predictions of membrane protein topology . J. Chem. Inf. Comput. Sci . 42 , 620 – 632 . OpenUrl CrossRef PubMed Web of Science [27]. ↵ Elazar , A. , Weinstein , J.J. , Prilusky , J. , Fleishman , S.J. , ( 2016 ). Interplay between hydrophobicity and the positive-inside rule in determining membrane-protein topology . Proc. Natl. Acad. Sci. U.S.A . 113 , 10340 – 10345 . OpenUrl Abstract / FREE Full Text [28]. ↵ Baker , J.A. , Wong , W.C. , Eisenhaber , B. , Warwicker , J. , Eisenhaber , F. , ( 2017 ). Charged residues next to transmembrane regions revisited: “Positive-inside rule” is complemented by the “negative inside depletion/outside enrichment rule” . BMC Biol . 15 , 66 . OpenUrl CrossRef PubMed [29]. ↵ Hunte , C. , Screpanti , E. , Venturi , M. , Rimon , A. , Padan , E. , Michel , H. , ( 2005 ). Structure of a Na + /H + antiporter and insights into mechanism of action and regulation by pH . Nature 435 , 1197 – 1202 . OpenUrl CrossRef PubMed Web of Science [30]. Lee , C. , Yashiro , S. , Dotson , D.L. , Uzdavinys , P. , Iwata , S. , Sansom , M.S.P. , von Ballmoos , C. , Beckstein , O. , et al. , ( 2014 ). Crystal structure of the sodium-proton antiporter NhaA dimer and new mechanistic insights . J. Gen. Physiol . 144 , 529 – 544 . OpenUrl Abstract / FREE Full Text [31]. Fippel , A. , Mir , S.H. , Wirth , C. , Gross , N.M. , Lentes , C.J. , Bialas , E. , Frank , F. , Ulbrich , M.H. , et al. , (to be published). Molecular determinants for substrate uptake in electrogenic sodium/proton antiporters . [32]. Winkelmann , I. , Uzdavinys , P. , Kenney , I.M. , Brock , J. , Meier , P.F. , Wagner , L.M. , Gabriel , F. , Jung , S. , et al. , ( 2022 ). Crystal structure of the Na + /H + antiporter NhaA at active pH reveals the mechanistic basis for pH sensing . Nat. Commun . 13 , 6383 . OpenUrl CrossRef PubMed [33]. Kokane , S. , Gulati , A. , Meier , P.F. , Matsuoka , R. , Pipatpolkai , T. , Albano , G. , Ho , T.M. , Delemotte , L. , et al. , ( 2025 ). PIP 2 -mediated oligomerization of the endosomal sodium/proton exchanger NHE9 . Nat. Commun . 16 , 3055 . OpenUrl PubMed [34]. Lee , C. , Kang , H.J. , von Ballmoos , C. , Newstead , S. , Uzdavinys , P. , Dotson , D.L. , Iwata , S. , Beckstein , O. , et al. , ( 2013 ). A two-domain elevator mechanism for sodium/proton antiport . Nature 501 , 573 – 577 . OpenUrl CrossRef PubMed [35]. Coincon , M. , Uzdavinys , P. , Nji , E. , Dotson , D.L. , Winkelmann , I. , Abdul-Hussein , S. , Cameron , A.D. , Beckstein , O. , et al. , ( 2016 ). Crystal structures reveal the molecular basis of ion translocation in sodium/proton antiporters . Nat. Struct. Mol. Biol . 23 , 248 – 255 . OpenUrl CrossRef PubMed [36]. ↵ Henderson , J.A. , Huang , Y. , Beckstein , O. , Shen , J. , ( 2020 ). Alternative proton-binding site and long-distance coupling in Escherichia coli sodium-proton antiporter NhaA . Proc. Natl. Acad. Sci. U.S.A . 117 , 25517 – 25522 . OpenUrl Abstract / FREE Full Text [37]. ↵ Chen , A.Y. , Panday , S.K. , Ri , K. , Alexov , E. , Brooks , B.R. , Damjanovic , A. , ( 2025 ). PKAD-R: Curated, redesigned and expanded database of experimental pK a values in proteins . J. Comput. Biophys. Chem . 24 , 1189 – 1203 . OpenUrl PubMed [38]. ↵ Kozma , D. , Simon , I. , Tusnády , G.E. , ( 2013 ). PDBTM: Protein Data Bank of transmembrane proteins after 8 years . Nucleic Acids Res . 41 , D524 – D529 . OpenUrl CrossRef PubMed Web of Science View the discussion thread. Back to top Previous Next Posted November 17, 2025. Download PDF Supplementary Material Data/Code Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following MPKaDB: A pKa Database for Exploring pH Dependence in Membrane Proteins Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share MPKaDB: A p K a Database for Exploring pH Dependence in Membrane Proteins Jiahao He , Yansheng Chen , Jinxi Wu , Zhitao Cai , Wenting Jia , Qiuchen Yue , Yandong Huang bioRxiv 2025.11.14.688495; doi: https://doi.org/10.1101/2025.11.14.688495 Share This Article: Copy Citation Tools MPKaDB: A p K a Database for Exploring pH Dependence in Membrane Proteins Jiahao He , Yansheng Chen , Jinxi Wu , Zhitao Cai , Wenting Jia , Qiuchen Yue , Yandong Huang bioRxiv 2025.11.14.688495; doi: https://doi.org/10.1101/2025.11.14.688495 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Molecular Biology Subject Areas All Articles Animal Behavior and Cognition (7618) Biochemistry (17633) Bioengineering (13856) Bioinformatics (41841) Biophysics (21399) Cancer Biology (18529) Cell Biology (25422) Clinical Trials (138) Developmental Biology (13352) Ecology (19860) Epidemiology (2067) Evolutionary Biology (24282) Genetics (15582) Genomics (22462) Immunology (17700) Microbiology (40295) Molecular Biology (17140) Neuroscience (88419) Paleontology (666) Pathology (2823) Pharmacology and Toxicology (4813) Physiology (7632) Plant Biology (15107) Scientific Communication and Education (2042) Synthetic Biology (4284) Systems Biology (9808) Zoology (2267)