变形菌视紫红质（proteorhodopsins, PRs）是一类能与视黄醛（retinal）共价结合的具有七个跨膜α螺旋（A-G）的吸光色素膜蛋白。其功能主要是作为光驱动的质子泵，即在光照的情况下，PR发生构象改变可以将质子从细胞内转移到细胞外，形成膜内外质子梯度差。PR在细胞膜两侧形成的质子梯度差可用于合成ATP，或供给细菌用于各项生命活动。因此，PR蛋白在生物体的光能利用和维持自然界生态系统稳定中具有重要的作用与意义。 PR与同为质子泵的嗜盐古菌视紫红质（bacteriorhodopsin, bR）在氨基酸序列上有30%的同源性，虽然大部分与视黄醛结合相关的氨基酸残基是保守的，但PR与bR蛋白间存在较大的差异。如bR中由第82位精氨酸（Arg82）、194位谷氨酸（Glu194）和204位谷氨酸（Glu204）残基组成的质子释放基团在PR中除了Arg94是保守的外，其余两个谷氨酸未能找到相对应的氨基酸，这说明PR和bR蛋白在质子释放的过程可能存在差异。另一个显著的差异就是，PR蛋白的质子受体，第97位的天冬氨酸（Asp97）羧基的pKa接近于7.0，而bR蛋白中相对应的质子受体，第85位天冬氨酸（Asp85）羧基的pKa值却小于3.0。 为了研究PR与bR蛋白中天冬氨酸pKa值有差异的原因，我们首先对不同微生物来源的视紫红质的氨基酸序列进行了比对，结果发现，PR蛋白的N末端存在一个富含酸性氨基酸残基（Asp22、Asp24和Asp27）的区域，而其它视紫红质不存在这个区域。为了明确这个区域的功能，我们接着对这个富含酸性氨基酸的区域进行了缺失突变，然后研究了这些缺失突变体的特性。结果表明，该酸性氨基酸区域的缺失不仅改变了突变体的吸光特性，还降低了Asp97的pKa值。通过构建和分析这个区域的酸性氨基酸残基的单位点突变体、双位点突变体及三位点突变体，我们发现其中Asp22是导致BPR蛋白Asp97的pKa值显著降低的关键氨基酸残基。进一步将Asp22突变成其他不同类型的氨基酸残基并对突变体进行分析后发现：无论将其突变成哪种类型的氨基酸残基，如丙氨酸（Ala）、谷氨酸（Glu）、苯丙氨酸（Phe）、天冬酰胺（Asn）、赖氨酸（Lys）、丝氨酸（Ser）和缬氨酸（Val），这些突变体的Asp97的pKa值都显著降低。随后，对我们实验室新近解析的BPR（Blue-light absorbing PR）的晶体结构进行分析，我们发现在这个结构中相邻的两个单体间除了疏水相互作用外，还存在三种亲水相互作用，即为Asp22与临近分子的第91、92和93位的主链氮原子通过氢键相互作用，但该作用与Asp97之间的相距很远；Trp34与临近的His75残基之间的氢键相互作用；以及位于胞质部分的一个非常复杂的亲水作用网络，涉及到多种氨基酸残基的相互作用。由于Asp22与Asp97之间的距离较远，我们推测其不太可能直接影响Asp97的pKa。为了明确Trp34与His75之间的相互作用的可能功能，我们也将Trp34分别突变为Ala、Asp、Glu、Phe、Lys、Leu、Gln、Arg和Tyr后，然后分别测定了这些突变体的Asp97的pKa值，结果显示Trp34突变也能够影响Asp97的pKa值。其中，Trp34突变成大部分氨基酸残基后都导致Asp97的pKa值下降，但当将其突变成酸性氨基酸残基（Asp和Glu）反而能增大Asp97的pKa值。然后又将His75分别突变成Ala、Glu、Met和Arg，然后测定了这些突变体的Asp97的pKa值，结果显示His75突变成Ala和Met后，Asp97的pKa值基本不变，而His75突变成Glu和Arg后能减小Asp97的pKa值。综合以上结果，考虑到His75与Asp97在位置上非常靠近，我们推测Asp22及Trp34调节Asp97的pKa值可能是通过His75起作用的，于是我们分别构建了PR中Asp22和His75的双突变体（Asp22突变成Asn，His75分别突变成Ala、Glu、Met和Arg）以及PR中Trp34和His75的双突变体（Trp34分别突变成Glu和Leu，而His75突变成Met），这些双突变体的Asp97的pKa值测定结果表明：所有双突变体Asp97 的pKa值与其相对应的His75单独突变体的Asp97的pKa值基本一致，因此我们认为Asp22和Trp34是通过His75来调节PR蛋白Asp97的pKa值的。 为了确定PR蛋白中质子释放基团及质子在PR蛋白中转运机理，我们首先结合BPR蛋白的三维晶体结构，对可能参与质子转运的氨基酸残基（包括Tyr76、Thr89、Thr91、Tyr95、Glu142、Tyr208、Asn221、Tyr224和Asn225）分别进行单独突变，然后对这些突变体的质子泵功能进行研究。结果表明，这些氨基酸的单独突变都能降低PR蛋白的质子泵功能，其中Glu142对质子释放起着至关重要的作用，E142L突变体基本丧失了质子转运功能。随后，我们对上述氨基酸进行了不同的组合突变。对这些突变体的质子泵功能测定的结果显示，在BPR中，三个氨基酸位点的突变BPR_Y95F_E142Q_Y224F导致BPR的质子泵功能完全丧失，而在GPR（Green-light absorbing PR）中，只有当六个氨基酸位点都突变了的突变体GPR_S89A_T91V_Y95F_E142Q_Y208F_Y223F才能完全丧失其质子泵功能。因此，结合已知的保守Arg94，我们认为：在BPR中，质子释放基团可能是由Arg94、Tyr95、Glu142和Tyr224这四个氨基酸残基组成，而在GPR中，质子释放基团可能是由Ser89、Thr91、Arg94、Tyr95、Glu142、Tyr208和Tyr223这七个氨基酸残基所组成。这个结果同时说明BPR和GPR之间的质子释放过程可能存在一定的差异。 为了探明淡水环境中PR蛋白在其宿主菌的生理功能，我们实验室前期从淡水水域中分离到一株含PR的细菌Exiguobacterium sp. JL-3（分离自南京市金牛湖水库），在此基础上，我们对PR蛋白（JL-3_PR）的生理功能进行了研究。我们首先分别测定了Exiguobacterium sp. JL-3菌体的质子转运功能，但未检测到，因此我们推测这可能是由以下两种原因导致的，第一个原因为JL-3菌株的厚细胞壁导致质子被俘获在胞质间隙，第二个可能原因为JL-3_PR没有质子转运功能。为了验证这些可能性，我们测定了JL-3的原生质体的质子转运功能，但仍未检测到质子泵活性。于是，我们又在大肠杆菌中克隆、表达了JL-3_PR，并对含有重组JL-3_PR的大肠杆菌质子泵活性进行了测定，结果显示：重组JL-3_PR有一定的质子泵活性，但其活性远低于其它PR（BPR或GPR）的质子泵活性。通过序列比对分析，我们发现微小杆菌属（Exiguobacterium）中PR蛋白的质子供体都是Lys，而不是典型的酸性氨基酸（Asp或者Glu）。因此，我们推测JL-3_PR的质子泵活性低可能与此有关。我们随后将JL-3_PR第96位的Lys分别突变成Ala、Asp、Glu和Arg，对这些突变体的质子泵活性进行检测发现，只有JL-3_PR_Lys96Asp突变体表现出与野生型的JL-3_PR相似的质子泵活性，而其余突变体几乎或者完全丧失了质子泵活性，这表明受体氨基酸残基Lys不是导致其活性低的唯一原因。运用SEFA-PCR的方法，我们还扩增并测定了JL-3_PR基因上下游的基因，并进行了序列分析。结果发现，在JL-3_PR基因的上游还存在一个孤立的编码传感器蛋白的基因。综合前人的研究结果及以上研究结果，我们推测，JL-3_PR在其宿主菌JL-3内可能是一个感光型的PR（sensory PR），而不是一种光驱动的质子泵。

Proteorhodopsins (PRs) are light-absorbing seven transmembrane α-helices (A-G) proteins covalently attached chromophore all-trans retinal. PRs are light-driven proton pumps that transport protons from the cytoplasmic to the extracellular side of the cells and generate proton gradients across the cell membrane. The proton gradient could be used to synthesize ATP, or to transport solutes and to move flagella. Therefore, PR plays an important role and significance in the light energy utilizing and ecosystem maintainance in nature. Protein sequence alignment showed that PRs share only 30% sequence identity with the bacteriorhodopsin (bR, also a proton pump) from Halobacterium salinarum, but most proton retinal binding related amino acid residues are conserved between PRs and bR. However, further investigation demonstrated that there are some big differences between these two proton pumps. For example, the two Glutamic acid residues in Proton Release Group (PRG) composed by Arg82, Glu194 and Glu204 in bR are not found in PRs except the conserved Arg94. This indicates that the proton translocation in PRs may be different from that of in bR. Another distinct difference between PRs and bR is that the pKa of the Schiff-base proton acceptor, the pKa of Asp97 in PRs is approximately equal to 7.0, while it is less than 3.0 in bR. To find out the reasons for the pKa difference of the counterion Asp97 between PRs and bR, the sequence comparison of PRs with other microbial rhodopsins was performed. The alignment results showed that the PRs harbor a conserved acidic amino acid-rich (Asp22, Asp24 and Asp27) N terminal in their matured polypeptide which is not present in other microbial rhodopsins. To test the function of this region, we deleted this acidic amino acid-rich region of PR and measured the pKa of the mutant, and found that the deletion of the acidic amino acid residues not only changed the maximum absorption, but also decreased the pKa of Asp97. Further studies on this acidic amino acid rich region showed that only the mutation of Asp22 could dramatically decrease the counterion pKa of PRs. To further study the Asp22, we substituted Asp22 by other type amino acids as follows: Asp22 to Ala, Glu, Phe, Asn, Lys, Ser and Val respectively, and then measured the pKa of Asp97 of these mutated PRs. The results showed that the pKa of Asp97 decrease in all mutants. We analyzed the 3D crystal structure of BPR and found that besides hydrophobic interaction, three kinds of hydrophilic interaction existed between two consecutive BPR monomers. The three hydrophilic interaction included Asp22, which is far way from Asp97, formed hydrogen bonds with backbone nitrogen atom of the first three amino acids of helix C; the second is the hydrogen bond between Trp34 and His75 of the consecutive BPR; the third is a very complex hydrophilic interaction network in the cytoplasmic side of BPR, containing various amino acid residues. Due to the far distance between Asp22 and Asp97, we thought that the influence of Asp22 to the pKa of Asp97 was not direct. In order to test the possible function of the interaction between His75 and Trp34, we mutated Trp34 to Ala, Asp, Glu, Phe, Lys, Leu, Gln, Arg and Tyr respectively in PRs, and then measured the pKa of these mutants. The measurement results showed that the Trp34 could increase and decrease the pKa of Asp97, such as Trp34 mutated to Asp or Glu could increase the pKa of Asp97 and the other mutants could decrease the pKa of Asp97. In addition, researchers had found that conserved His75 could change the absorbing spectra in PRs. To test whether His75 also regulating the pKa of the counterion, we replaced His75 by Ala, Glu, Met and Arg respectively and then measured the pKa of the Schiff-base Asp97 of these mutants. We found that the mutants His75Ala and His75Met could not modulate the pKa of Asp97, and however, the other two mutants His75Glu and His75Arg could decrease the counterion pKa. Base on the above results and the approximate position of His75 to Asp97, we hypothesized the modulation of Asp22 and Trp34 on the pKa of Asp97 might be performed through His75. To reveal the possible regulatory mechanism of Asp22 and Trp34 on pKa of Asp97, we constructed double-mutated Asp22_His75 and Trp34_His75 and measured their pKa of the Schiff-base counterion. The results showed that the pKa of double-mutants Asp22_His75 and Trp34_His75 are consistent with that of the corresponding single-mutated His75, not with that of single-mutants Asp22 and Trp34. So, the regulation of Asp22 and Trp34 on the pKa of Asp97 should be carried out through the His75 residue. To elucidate the mechanism of PRG in PRs, the possible amino acid residues were also studied based on the 3D crystal structure of BPR. The residues, including Tyr76, Thr89, Thr91, Tyr95, Glu142, Tyr208, Asn221, Tyr224 and Asn225, were mutated to Phe, Val, Val, Phe, Gln, Phe, Leu, Phe and Leu respectively and the proton pumping activities of these mutants were measured. The results showed that the proton translocation activities of all mutants were much lower than that of the wild type; Glu142 was a critical amino acid residue for proton translocation and the mutant E142L almost lost proton pumping activity. In order to get the mutant without proton pumping activity, we further constructed different combination types of these amino acid residues mutated and measured the proton pumping activities of these mutants. We found that the triple-mutant BPR_Y95F_E142Q_Y224F in BPR exhibited no proton pumping activity, and the hexa-mutant GPR_S89A_T91V_Y95F_ E142Q_Y208F_Y223F also lost proton pumping activity completely in GPR. Therefore, with the former recognized conserved Arg94, PRG in BPR should be composed by Arg94, Tyr95, Glu142 and Tyr224, and that in GPR should be Ser89, Thr91, Arg94, Tyr95, Glu142, Tyr208 and Tyr223. This result also indicates that the process of proton translocation in BPR is different from that in GPR. To study the physiological functions of PRs in their host strains from fresh water, a PR-bearing bacterium Exiguobacterium sp. JL-3, isolated from Jiuniu Lake at Nanjing, was characterized. Proton pumping assays with Exiguobacterium sp. JL-3 strain and its protoplast were performed respectively to test the function of the JL-3_PR in its native host. However, no proton pumping activities were observed. This phenomenon could be caused by two possibilities: one may be due to the thick cell wall of the Gram-positive. E. sp. JL-3 arrested the proton transfer from cytoplasm to the outside of the cells, and the other one may be the proton flux capability of JL-3_PR was blank or extremely low in E. sp JL-3. We then measured the proton translocation activity with the protoplast of E. sp. JL-3, but still no activity was observed. We then cloned and expressed JL-3_PR gene in E. coli and measured the proton translocation activity in the E. coli. The recombinant JL-3_PR showed typical light-driven proton pumping activity, although much lower activity as compared with other proton pumping activities of PRs like GPR or BPR. Protein sequence alignment revealed a distinct difference of the putative Schiff-base proton donor between the PRs of Exiguobacterium spp. and that of other microbial rhodopsins. It is a Lysine in PR from Exiguobacterium spp. rather than a typical acidic residue (Asp or Glu) in other PRs. So we proposed that the high pKa of the Lysine (proton donor) may be the reason for the low proton translocation activity. We then mutated the Lysine to Ala, Asp, Glu and Arg and also measured the proton pumping activities of these mutants. Our data indicated that only the mutant JL-3_PR_K96D had a similar proton pumping activity to the wild type JL-3_PR while other mutants exhibited very low or no activity of proton translocation. This indicated that the Lysine is not the sole reason for the low proton translocation efficiency of JL-3_PR. Moreover, the flanking genes of JL-3_PR gene in genomic DNA were sequenced and then the resulting sequences were analyzed. We also found an orphan transducer is present upstream of JL-3_PR gene. With all these results, we believe that JL-3_PR may function as a sensory PR rather than a proton pumping PR in strain E. sp. JL-3.