Investigating Acid Production by Streptococcus mutans with a Surface-Displayed pH-Sensitive Green Fluorescent Protein

Acidogenicity and aciduricity are the main virulence factors of the cavity-causing bacterium Streptococcus mutans. Monitoring at the individual cell level the temporal and spatial distribution of acid produced by this important oral pathogen is central for our understanding of these key virulence factors especially when S. mutans resides in multi-species microbial communities. In this study, we explored the application of pH-sensitive green fluorescent proteins (pHluorins) to investigate these important features. Ecliptic pHluorin was functionally displayed on the cell surface of S. mutans as a fusion protein with SpaP. The resulting strain (O87) was used to monitor temporal and spatial pH changes in the microenvironment of S. mutans cells under both planktonic and biofilm conditions. Using strain O87, we revealed a rapid pH drop in the microenviroment of S. mutans microcolonies prior to the decrease in the macro-environment pH following sucrose fermentation. Meanwhile, a non-uniform pH distribution was observed within S. mutans biofilms, reflecting differences in microbial metabolic activity. Furthermore, strain O87 was successfully used to monitor the S. mutans acid production profiles within dual- and multispecies oral biofilms. Based on these findings, the ecliptic pHluorin allows us to investigate in vivo and in situ acid production and distribution by the cariogenic species S. mutans.


Introduction
Streptococcus mutans is generally considered to be the principal causative agent for dental caries (tooth decay) [1,2]. This bacterium can generate acids from fermentable sugars (acidogenicity), which is the main virulence factor in the etiology of dental caries development [3]. This acidogenicity in combination with its ability to survive in the acidic environment it produces (aciduricity), further provides a competitive edge for S. mutans over other commensal species [4]. While these important virulence features have been studied in mono-species experiments [3,4], little is known about the spatio-temporal aspects of acid production and how it is influenced by the presence of other species in the complex multi-species microbial communities that inhabit the oral cavity.
Several techniques have been developed to analyze pH changes within biofilms, including in-dwelling or micro-touch electrodes, fluorescent dyes, and NMR microscopy. In-dwelling electrodes were used to measure the pH at the base of biofilms developed on teeth [5], but their size limited the ability to accurately detect spatial pH changes in the microenvironment of oral biofilms. Micro-touch electrodes that have been employed for several biological systems [6,7] exhibited similar limitations due to their tip size [8]. Furthermore, both devices are difficult to position and prone to disturbing the very microenvironment that they are supposed to monitor. More recently, pH-sensitive fluorescent dyes as well as various modern imaging systems including laser-induced fluorescence lifetime imaging, two-photon microscopy and NMR microscopy have been used to overcome some of these problems and several studies have demonstrated their utility in measuring pH profiles in microbial biofilms [9][10][11][12][13][14][15].
An interesting alternative to above approaches are the pHsensitive green fluorescent protein variants (pHluorins), developed by Miesenbőck et al. [16]. Chimeras consisting of a pHluorin and the proteins of interest have been used as in vivo pH sensors in a variety of biological systems such as neurons and yeast [17][18][19][20][21][22] to signal not only the anatomical positions of tagged cells or proteins but also their physiological states. In particular two classes of pHluorin have been engineered, which are termed 'ratiometric' and 'ecliptic' pHluorins. Ratiometric pHluorin produces two excitation peaks -one that increases in intensity with rising pH and one that decreases with rising pH, whereas ecliptic pHluorin gradually loses fluorescence with decreasing pH [23,24]. Despite the wide application of ecliptic pHluorin as in vivo pH sensor in various eukaryotic systems, it was rarely used in bacteria [25]. In fact, to the best of our knowledge, it has never been used as in vivo pH sensor for microbial physiological studies under biofilm conditions.
In this study, we explored the application of ecliptic pHluorin to monitor acid production by S. mutans in a biofilm environment. An ecliptic pHluorin was fused to the S. mutans SpaP protein for in vivo expression on the cell surface and the resulting strain was used to simultaneously monitor the localization of S. mutans and track the local extracellular pH in an in situ, real time manner.

Bacterial Strains and Growth Conditions
E. coli strain DH5a was used for cloning and plasmid amplification. The bacteria were grown aerobically at 37uC in Luria-Bertani (LB) medium supplemented with 250 mg/mL spectinomycin or 100 mg/mL ampicillin when needed for plasmid selection. S. mutans, S. sanguinis and S. gordonii were routinely cultured in Todd-Hewitt (TH) media (Difco) at 37uC in the presence of 5% CO 2 . The S. mutans strain expressing the pHluorin-SpaP fusion was cultured in the same medium supplemented with 800 mg/mL spectinomycin.

Strain Construction
The backbone vector for constructing gene fusions was pBluescript II SK (-), and pFW5 was used as suicide vector for integration of the recombinant gfp-spaP fusion into the S. mutans genome. To construct a cell surface displayed ecliptic pHluorin derivative, the 0.7 kb ecliptic pHluorin encoding ORF [16] was amplified by PCR from vector pGM87 (kindly provided by Dr. Gero Miesenbőeck) using primers phluorin-01 and phluorin-02 ( Table 1). The pHluorin coding sequence with a linker sequence (accggtcccgccgcttccgccgct) at its 39 end was then in-frame inserted via overlapping PCR between the second and third amino acids after the identified signal-peptide cleavage site of SpaP [26], a surface protein antigen-encoding gene of S. mutans. Three initial PCR reactions were performed to generate overlapping gene segments (leading sequence-gfp-linker-spaP). Internal primers ovpcr-b, -c, -d and ovpcr-e (Table 1) (Table 1). PCR was initiated by a hot-start procedure and reaction mixture containing all other reagents except Pfu polymerase (Stratagene) was heated at 94uC for 10 sec, then brought to an 80uC holding temperature prior to addition of 5.0 U Pfu polymerase. Then cycling proceeded as follows: 94uC 10 sec, 68uC 4 min for 15 cycles; then 94uC 10 sec, 68uC, 4 min plus 15 sec extension period per cycle for further 15 cycles; followed by a 10 min extension at 72uC. A 1.1 kb DNA sequence upstream of the ldh start codon containing the ldh promoter [27] was PCR amplified from genomic DNA of S. mutans UA 140 using the primers ldh-p01 and ldh-p02 ( Table 1). The resulting chimeric gene product was then ligated downstream of the lactate dehydrogenase gene (ldh) promoter (Fig. 1A, and a schematic drawing was provided as supporting information).
The recombinant construct was confirmed by restriction analysis, PCR, DNA sequencing and testing of the GFP-SpaP reporter protein function. pFW5W(ldh p -leading sequence-gfplinker-spaP) were transformed into S. mutans strain UA140 to generate pHluorin-expressing strain O87. The integration of the construct into the chromosome of S. mutans via single crossover homologous recombination was confirmed by PCR and its functional expression was confirmed by fluorescent microscopy analysis (Nikon ECLIPSE E400 microscope).
Strains, plasmids and primers used in this study are shown in Tables 1 and 2.

Western Blots
Cell wall/membrane preparations from S. mutans were processed as described previously [32]. Briefly, cells were grown overnight at 37uC, collected by centrifugation and resuspended in 50 mM TrisNHCl (pH 8.0) containing 1 mM PMSF. After being transferred to a chilled 2 ml microcentrifuge tube containing 425-600 mm diameter glass beads (Sigma), cells were lysed using a Mini-Bead Beater homogenizer (Biospec Products) at full speed for 10 min. The glass beads were removed by decantation and the undisrupted cells were pelleted by centrifugation at 2,0006g for 10 min. Half of the supernatant was stored as whole-cell lysate. The remaining supernatant was further centrifuged at 150,0006g for 2 h, and the supernatant was collected as cytoplasmic fraction of the cells. The cell wall/membrane fraction containing pellet was washed once with distilled water and resuspended in 50 mM TrisNHCl (pH 8.0) supplemented with 1 mM EDTA and treated with RNase (10 mg L 2l ) and DNase (10 mg L 2l ) at 37uC for 2 h.
Western blotting of whole-cell lysates, cytoplasmic fractions, as well as the cell wall/membrane fractions was performed using anti-GFP antibody (Thermo) [33]. The signal was developed using the Pico chemiluminescence kit (Pierce).

General Phenotypic Characterization Assays
Growth kinetics were measured for S. mutans UA140 and the strain O87 containing the cell surface displayed pHluorin. Biofilm formation, glucan production and glycolytic rate were also performed as previously described [34][35][36].

Growth of in vitro Biofilms
Overnight cultures (OD 600 nm = 1) of S. mutans wild type or its derivative were diluted 1:100 in fresh TH medium containing 0.5% (wt/vol) sucrose. Five hundred ml of diluted culture was inoculated in each well of a sterile 8-well Lab-Tek TM Chambered Coverglass (Nalge Nunc International, Naperville, IL), followed by 16 h static incubation at 37uC in 5% CO 2 atmosphere to allow mono-species biofilm formation.
To establish dual-species biofilms, 1.25610 6 CFU/ml of S. mutans was mixed with the base-producing bacteria, S. sanguinis ATCC 10556 or S. gordonii DL1 [37,38] at a ratio of 1:10 and followed by 16 h incubation under same condition as described above for mono-species biofilm formation. The saliva-derived biofilms spiked with S. mutans cells were established as follows: Unstimulated saliva samples were collected from 6 subjects 6 h post-cleaning. Individual saliva samples were diluted 1:4 in TH medium. The suspension was subjected to a low speed centrifugation for 10 min at 6006g to remove eukaryotic cells and large debris [39]. Five hundred ml of the supernatant was inoculated into each well of a sterile 8-well Lab-Tek TM Chambered Coverglass, together with about 1.25610 6 CFU/ml S. mutans cells expressing surface associated ecliptic pHluorin (O87). The ratio of salivary bacteria to S. mutans is about 10:1. Subsequently, 16 h dual-species and saliva-derived biofilms spiked with S. mutans were challenged with either 2% sucrose or 50 mM arginine-HCl (pH 7.6).

Fluorescence Measurement
Fluorescence emission of pHluorin was measured by a Cary Eclipse fluorescence spectrophotometer (Varian, Mulgrave, Victoria, Australia) using a bandpass filter with a center wavelength of 40065 nm. The fluorescence signal intensity of planktonic cells of S. mutans strain O87 with surface-expressed pHluorin was monitored at different pH. The buffer solutions (pH 5.5, 6.0, 6.5, 7.0, 7.5) were prepared by mixing potassium phosphate monobasic anhydrous and sodium phosphate dibasic heptahydrate in different ratios to obtain the desired pH. The pH of the buffer solutions was determined by the AB15 pH meter (Fisher).

Imaging
All biofilm images were collected with a Zeiss LSM 5 PASCAL confocal laser scanning microscope (CLSM) using LSM 5 PASCAL software (Zeiss, Jena, Germany).
For visualization, the dual-species biofilms and saliva-derived biofilms spiked with S. mutans O87 were stained with 5 mM CellTracker Orange CMRA (Invitrogen) according to manufacturer's instructions. S. mutans within biofilms was specifically labeled with an anti-S. mutans monoclonal antibody SWLA1-IgG2a as described previously [40]. Alexo Fluor 633 conjugated goat anti-mouse IgG (Sigma, St Louis, MO) was used as secondary antibody.
For evaluating the respiratory activity of microcolonies in S. mutans biofilm and semiquantitative estimate of metabolic vs dormant cells, 5 mM 5-cyano-2, 3-ditolyl tetrazolium chloride (CTC) (BacLight RedoxSensor CTC Vitality Kit, Invitrogen) [41] was added at the beginning of biofilm formation, and CTC fluorescence was monitored when the biofilms were 16-h old.
Excitation at 488 nm with an argon laser in combination with a 505-530 nm bandpass emission filter was used for pHluorin fluorescence imaging and a 560 nm longpass emission filter for CTC fluorescence imaging. An excitation at 543 nm and 633 nm with helium-neon laser and a 560 nm and 650 nm longpass emission filter was utilized respectively to reveal CellTracker Orange-stained and Alexa Fluor 633-labeled cells within the biofilms. The scanning module of the system was mounted onto an inverted microscope (Axiovert 200 M). The 406(NA/1.3) or 636(NA/1.4) numerical aperture oil-immersion objectives were used for imaging. To visualize pHluorin expression with CLSM, cells were left for 1 h at room temperature to aerate the cells for proper folding of the pHluorin fluorophore [42]. The samples were placed on a commercially available thermoelectric heating stage insert (Brook Industries, Lake Villa, IL) (13069561 mm) to keep the biofilms at 37uC during imaging. Image stacks of three randomly chosen spots were collected for each biofilm for data analysis. In order to record pH changes in biofilms following a sucrose spike, an initial series of images was taken as a baseline, followed by addition of sucrose to a final concentration of 2%. The same field was imaged every 15 min over a 1 h time period. Time point t = 0 was measured just before sucrose addition.

Image Analysis
Image stacks (1024 by 1024-pixel tagged image file format) were quantified using the image analysis software COMSTAT with MATLAB 7.0 (The MathWorks, USA). In order to control for photobleaching, image stacks in a buffered neutral environment were taken. In fact, the decrease of fluorescence intensity of pHluorin due to photobleaching was less than 10% over 15 min. DeconvolutionJ was used to remove or reverse the blurring present in microscopic images and improve the contrast and resolution of digital images.

Construction of a S. mutans Strain Expressing pHluorin on its Cell Surface
In an effort to display ecliptic pHluorin on the cell surface of S. mutans for effective monitoring of local extracellular pH at the individual cell level in a real time, in situ manner, we constructed strain O87. In this strain ecliptic pHluorin was fused to the S. mutans surface antigen protein SpaP to express a cell wallassociated chimeric protein (Fig. 1A). Western blot analysis with a GFP-specific antibody confirmed that the ecliptic pHluorin was present in the whole cell and cell wall/membrane fractions but not in the cytoplasmic fraction of strain O87 (Fig. 1B). Furthermore, when using the lipophilic fluorescent dye FM 4-64 to reveal the cell membrane, we observed that the pHluorin fluorescence signal of strain O87 colocalized with the membrane fluorescence signal (Fig. 1C).
The integration of pHluorin reporter system into the chromosome of S. mutans did not affect cell viability, and no significant difference in biofilm formation, glucan production and glycolytic

pH-responsivity of S. mutans Cells Carrying the Surface Displayed pHluorin
To examine the in vivo pH-responsivity of the ecliptic pHluorin when expressed on the cell surface of S. mutans as a pHluorin-SpaP fusion protein, planktonic cells of strain O87 were suspended in buffer with different pH values and their fluorescence signals were monitored. A reduction in fluorescence signal intensity corresponding to the decrease in pH was observed ( Fig. 2A), confirming the pH-responsivity of the surface displayed ecliptic pHlourin in S. mutans strain O87 and its potential for detecting pH changes in real time.
The pH-responsivity of the cell surface displayed pHluorin-SpaP fusion protein in strain O87 was further tested under biofilm growth conditions. Similar to their planktonic counterparts, the biofilm cells responded to the shift in external pH from 7.5 to 5.5 with a drastic reduction (,70%) in pHluorin fluorescence intensity within 10 min. The signal intensity was almost fully restored (.90%) after 10 min exposure to a readjusted external pH of 7.5 (Fig. 2B).

pHluorin Enables Temporal and Spatial Evaluation of the in situ pH Profiles within S. mutans Biofilms
The dynamic changes of pHluorin-SpaP fusion protein fluorescence intensity within S. mutans biofilms were monitored after addition of 2% sucrose under phosphate buffered (pH 7.5) and unbuffered condition (Fig. 3). Results showed that, under buffered condition, the medium pH remained stable at 7.5 and the biofilms displayed bright fluorescence throughout the 60 min incubation period even in the presence of 2% sucrose. In unbuffered medium in contrast, the addition of sucrose induced a pH drop from 7.5 to below 6.0 within 60 min accompanied by a striking decrease in the fluorescent signals within S. mutans biofilms. Interestingly, after 30 min of incubation, the pHluorin fluorescence intensity decreased by nearly 50%, although the medium pH only dropped to about 7.0. According to the change in fluorescent intensity observed in response to pH reduction in Figure 2A, the microenvironment pH of microcolonies decreases to almost pH 5.5. This suggests that the pH in the microenvironment decreases prior to the pH drop in the medium following sucrose fermentation.

pHluorin Disclosing Non-uniform pH Profiles Associated with Differences in Microbial Metabolic Activity
One intriguing discovery in this study is the pH heterogeneity of different microcolonies within the same S. mutans biofilm in the presence of sucrose ( Figure 4A). Within biofilms of S. mutans O87 some microcolonies displayed a much more intense green fluorescence signal than others (indicated by arrows) after addition of sucrose, suggesting the pH in the microenvironment of microcolonies is not homogeneous throughout S. mutans biofilms following sucrose fermentation. To investigate the possible correlation between microenvironment pH heterogeneity and the metabolic status of microcolonies, CTC (5-cyano-2,3-ditolyl tetrazolium chloride) was added to the experimental setup for evaluating the respiratory activity of microcolonies within S. mutans biofilm in the presence of sucrose and a correlation between metabolic status and pH was observed ( Fig. 4Ba and 4Bb). Microcolonies with less CTC signal typically exhibited brighter pHluorin fluorescence, while strong CTC signals were often associated with reduced pHluorin fluorescence (Fig. 4Bc and 4Bd).

Surface-displayed pHluorin as a Tool for Monitoring in situ Acid Production by S. mutans within Oral Biofilms
Next we explored the application of the surface-displayed (O87) pHlourins as ''pH-meters'' to study acid production by S. mutans under biofilm conditions in the presence of other microorganisms. For pHlourin-independent identification of O87 within dualspecies biofilms with S. sanguinis or S. gordonii as well as a multispecies environment, we employed a previously described S. mutans-specific monoclonal antibody (MAb) together with a corresponding red-fluorescent labeled secondary antibody. This approach allowed simultaneous monitoring of the spatial localization of S. mutans and its acid production in situ in dual-species (Fig. 5A) and multi-species (Fig. 5B) biofilms. In the presence of sucrose, acids generated by glycolysis lowered the pH in the microcolonies microenvironment of S. mutans as indicated by the corresponding reduction in pHluorin fluorescence intensity. For the dual-species biofilms, the pHluorin fluorescence intensity decreased more significantly after addition of sucrose for 30 min in comparison with addition of arginine (20.069.33% vs 6.14612.86% in fluorescence reduction). Similar reduction rates in fluorescence intensity were observed for the saliva-derived multi-species biofilms (16.1467.16% vs 5.5266.23% in fluorescence reduction).

Discussion
In this study, we demonstrated, for the first time, the use of a pH-sensitive variant of GFP the ecliptic pHluorin, [16] as an in vivo, in situ pH meter to monitor the acid production by cariogenic S. mutans at the individual cell level.
SpaP was chosen as a fusion partner to orientate ecliptic pHluorin onto the cell surface of S. mutans for sensing extracellular microenviromental pH within biofilms. The surface-anchored chimeric fusion protein displayed similar in vivo pH responsiveness characteristics to the original ecliptic pHluorin, indicating that the in-frame fusion of ecliptic pHluorin to SpaP of S. mutans retained its spectral properties, and the resulting S. mutans O87 strain can be used to monitor extracellular microenviromental pH within biofilms.
Our study demonstrated that, by changing its fluorescence intensity in a pH-dependent manner, S. mutans O87 was able to monitor cellular acid production within mono-species biofilm in the presence of fermentable carbohydrates. Intriguingly, when exposed to 2% sucrose for a short period of time, although the solution pH stayed around 7.0, a more drastic reduction in pH (5.5) was monitored within S mutans microcolonies indicating the microenvironments within microcolonies experience a much faster pH drop. The exposure of acidogenic bacteria, such as S. mutans, to an excess amount of fermentable carbohydrates can result in the rapid accumulation of glycolytic intermediates and acidification of the extracellular microenvironments. However, due to the diffusion barrier of the matrix and buffering capacity of the medium, the pH monitored in the macro-enviroment often does not reflect the acidity within the microbial colonies. For example, in the human oral cavity the pH of saliva collected directly adjacent to plaque remains near neutral, while localized pH within plaque can drop below 5.5 following sugar consumption, and trigger demineralization process [43]. Previous studies also showed that the pH within a biofilm may differ substantially from that encountered in the surrounding solution [11,20,44,45]. Within biofilms, the pH of the local microenvironment is more important and clinically relevant than the overall pH since the intra-plaque pH determines the dental plaque structure and physiological process, and ultimately affects enamel and dentine demineraliza- tion [46]. It has been proposed that pH fluctuations in microenvironments could cause a heterogeneous distribution of mineral phases within natural biofilms [47,48]. Sissons et al. pointed out that the pH at any point within dental plaque biofilm resulted from an interaction of cells and cell buffers with bacterial metabolism of a diffusion-limited supply and clearance of substrates, metabolic products and mobile buffers [49]. The presence of extracellular polysaccharides (EPS) within and surrounding microcolonies could restrict the diffusion of a variety of molecules, thus affect the microenvironment pH within the biofilms [12,15,[50][51][52]. Meanwhile, it is also likely that the S. mutans cell surface might retain protons generated from its own acid production to prepare for acid-stress ahead of the rest of the bacteria.
Our study further revealed that pH reduction among S. mutans microcolonies was not uniform when biofilms were exposed to sucrose. pH heterogeneity of different microcolonies exists within S. mutans biofilm in presence of sucrose (Fig. 4A). While the   majority of microcolonies lost their fluorescence due to fermentation-induced pH drop after addition of sucrose for 60 min, there were still microcolonies with fairly strong fluorescence signals (Fig. 4B). The heterogeneity of pH profiles suggested that, within the same mono-species biofilm different cell populations with differential physiological status could coexist. Haagensen et al., have shown that the development of Pseudomonas aeruginosa biofilms involved an early differentiation of the attached cells into at least two isogenic but clearly distinguishable subpopulations [53]. It has also been demonstrated that S. mutans cells could segregate into two subpopulations induced by CSP, one becoming competent and the other lysed, resulting in intrapopulation diversity [54]. The factors that contribute to physiological heterogeneity in biofilms include micro-scale chemical gradients, adaptation to local environmental conditions, stochastic gene expression and genotypic variation that occurs through mutation and selection [55]. Our CTC staining further revealed that the differential acid production of S. mutans microcolonies was correlated with their different metabolic activities.
Using the pHlourin expressing strains, we found that pH values on the S. mutans surfaces changed when incubated in the presence of oral commensal organisms such as S. sanguinis or S. gordonii. In the presence of arginine, S. sanguinis (Fig. 5A) or S. gordonii (data not shown) in the dual species biofilm, or other arginine-utilizing bacteria (Fig. 5B) within saliva-derived biofilm can consequently produce alkali, and the elevated microenvironment pH of microcolonies as a result of alkali production increases the pHluorin fluorescence intensity. While in the presence of sucrose, S. sanguinis or S. gordonii in the dual species biofilm, or other acidogenic bacteria such as lactobacilli within saliva-derived biofilm also metabolize carbohydrate to acid besides S. mutans, and will reduce the microenvironment pH of microcolonies as a result of acid production decreases the pHluorin fluorescence intensity. The results suggest that the surface-anchored pHlourin in S. mutans cannot only be used to analyze acid production by S. mutans, but also demonstrated a promising application in monitoring local microenvironment pH changes within multispecies biofilms. The latter might reflect the metabolic interactions between S. mutans and its neighboring bacterial species within oral communities.
The data presented in this paper demonstrated that the ecliptic pHluorin is a valuable tool for in vivo/in situ monitoring of acid production by the cariogenic S. mutans in a multispecies environment. The surface-anchored ecliptic pHluorin in S. mutans has a number of unique properties: (1) it is genetically encoded and biosynthesized in situ; (2) it is localized onto the cell surface of S. mutans and allows fluorescence imaging of microenvironment pH with high spatial and temporal resolution in vivo; (3) in combination with fluorescently labeled anti-S. mutans monoclonal antibody, we can effectively study the spatial distribution of S. mutans in correlation with its pH profile without worrying about the diffusion and penetration problem encountered when using the pH sensitive dyes; and (4) it does not interfere with normal structure and physiological processes of biofilm. In addition to its value for basic research in the acid-associated pathogenic processes involved in oral caries disease, the S. mutans strain could also be practical tool for evaluating the efficacy of dental products in eliminating S. mutans or reducing its acidogenicity.
In summary, we have developed a novel technique for optical imaging of pH in oral biofilms. Using this tool, we were able to demonstrate the faster pH drop observed within microcolonies than that of the surrounding medium following sucrose fermentation. Furthermore, we revealed the heterogeneity of pH profiles within S. mutans biofilms which reflected the differences in microbial metabolic activity. These results are consistent with similar findings by other group [12]. Our studies strongly support the notion that the ecliptic pHluorin is a valuable tool for in vivo/in situ monitoring of acid-associated pathogenicity of S. mutans. Figure S1 Schematic drawing of construction of pFW5ldh p -leading sequence -gfp-linker-spaP. (DOC)