A New Method for Albuminuria Measurement Using a Specific Reaction between Albumin and the Luciferin of the Firefly Squid Watasenia scintillans

This study demonstrates that the luciferin of the firefly squid Watasenia scintillans, which generally reacts with Watasenia luciferase, reacted with human albumin to emit light in proportion to the albumin concentration. The luminescence showed a peak wavelength at 540 nm and was eliminated by heat or protease treatment. We used urine samples collected from patients with diabetes to quantify urinary albumin concentration, which is essential for the early diagnosis of diabetic nephropathy. Consequently, we were able to measure urinary albumin concentrations by precipitating urinary proteins with acetone before the reaction with luciferin. A correlation was found with the result of the immunoturbidimetric method; however, the Watasenia luciferin method tended to produce lower albumin concentrations. This may be because the Watasenia luciferin reacts with only intact albumin. Therefore, the quantification method using Watasenia luciferin is a new principle of urinary albumin measurement that differs from already established methods such as immunoturbidimetry and high-performance liquid chromatography.


Introduction
The firefly squid Watasenia scintillans is a luminescent deep-sea squid that inhabits the Japan Sea. As it migrates from the deep sea to shallow waters to spawn every spring, the blue luminescence can be observed by the naked eye. The brightest light organ of Watasenia is located at the tip of the fourth arm that emits light. Other small organs scattered on the mantle emit weaker light. The luminescence is considered a counter-illumination system to escape from predators [1]. Like other luminescent organisms, W. scintillans has luciferase and luciferin in its body [2,3]. The luciferase of Watasenia has been found to be ATP-dependent and to exist in a crystalline structure [4,5]. In recent years, there have been attempts to identify Watasenia luciferase through mass spectrometry and transcriptome analysis; however, active luciferase has not yet been purified [6,7]. Watasenia luciferin has been identified and isolated from the light organ in arm photophores [8][9][10]. The structure of Watasenia luciferin is similar to that of coelenterazine, which is a luciferin present in a wide variety of marine luminescent organisms [11,12]. Although there are no reports of Watasenia luciferin being used for testing in medicine, this paper explores the potential use of the luciferin in medical testing.
The kidney has a functional component unit involved in urine production called the nephron, which is composed of the glomerulus and tubules, and it is known that albumin leaks into the urine when the glomerulus is damaged. Albuminuria is considered a risk factor for various diseases, among which diabetic nephropathy is known to be closely related to urinary albumin concentration. Diabetic nephropathy is a complication of diabetes and progresses through several stages. In its early stage, trace amounts of albumin are leaked into the patient's urine. Early diagnosis and appropriate glycemic and blood pressure control improve the prognosis of diabetic nephropathy [13][14][15][16][17]. Therefore, measuring urinary albumin levels is critical for suppressing diabetic nephropathy progression. Urinary albumin concentrations are very low compared with serum albumin concentrations, and they coexist with several molecules. Moreover, urine samples contain intact as well as half-degraded albumin, which is not observed in blood samples. These factors complicate the measurement of urinary albumin and the interpretation of its results.
The measurement of urinary albumin by dipstick is simple and rapid, but it is not highly quantitative. Immunoturbidimetry is a method used to quantify antigen-antibody complex precipitates in a liquid by measuring the optical density, and it is widely used to measure albumin concentrations in the urine of patients [18]. Although this method is simple and sensitive, methods based on antigen-antibody reactions exhibit a prozone (hook) effect, in which the signal is reversed at very high albumin concentrations [19]. Furthermore, several researchers have reported the presence of immuno-unreactive albumin in urine [20]. In methods based on high-performance liquid chromatography (HPLC), albumin is purified according to its molecular weight, and its concentration is measured; these methods are considered to be capable of measuring immuno-unreactive albumin [21]. Nevertheless, HPLC-based methods can also measure proteins of the same molecular weight as that of albumin. Therefore, HPLC-based methods tend to produce larger albumin concentrations than antibody-based methods [22]. In this study, we report that Watasenia luciferin reacts with human albumin and emits luminescence; we also attempt to establish the Watasenia luciferin method as a new quantitative technique for measuring urinary albumin levels in addition to the two major existing methods.

Specific Reaction and Luminescence of Watasenia Luciferin and Human Albumin
We synthesized Watasenia luciferin according to a previously reported method [23], with a slight modification (Schemes 1-4 and Supplementary Figures S1 and S2). We first planned to isolate Watasenia luciferase using the synthesized luciferin, but this was not successful. However, in this process, we discovered by chance that Watasenia luciferin reacted with albumin and emitted light, and then we began characterizing the albuminluciferin reaction.
Human albumin, bovine albumin, and several other proteins were mixed with Watasenia luciferin, and their luminescence was measured ( Figure 1). The strongest luminescence was observed in the reaction with human albumin, and the intensities of all other groups were compared against human albumin to determine whether the difference was significant. The luminescence levels of other proteins were similar to those observed without the protein. This indicates that luciferin reacts specifically with albumin and emits luminescence.   Human albumin, bovine albumin, and several other proteins were mixed with Watasenia luciferin, and their luminescence was measured ( Figure 1). The strongest luminescence was observed in the reaction with human albumin, and the intensities of all other groups were compared against human albumin to determine whether the difference was significant. The luminescence levels of other proteins were similar to those observed without the protein. This indicates that luciferin reacts specifically with albumin and emits luminescence.

Optimization of the Albumin-Luciferin Reaction
To optimize the albumin-luciferin reaction, Tris buffer solutions with several pH values in the reaction solution were evaluated. The strongest luminescence was observed when the reaction was performed using a buffer with pH 10 ( Figure 2A). As the luminescence of luciferin is considered to be due to oxidation [24], we evaluated whether the addition of H 2 O 2 to the solution would change the luminescence level. The results showed that the luminescence increased with the addition of albumin, even without H 2 O 2 ; however, in the presence of H 2 O 2 , the luminescence increased even further ( Figure 2B). Based on these findings, we measured the luminescence with H 2 O 2 in a buffer solution with pH 10.
To optimize the albumin-luciferin reaction, Tris buffer solutions with several pH values in the reaction solution were evaluated. The strongest luminescence was observed when the reaction was performed using a buffer with pH 10 ( Figure 2A). As the luminescence of luciferin is considered to be due to oxidation [24], we evaluated whether the addition of H2O2 to the solution would change the luminescence level. The results showed that the luminescence increased with the addition of albumin, even without H2O2; however, in the presence of H2O2, the luminescence increased even further ( Figure  2B). Based on these findings, we measured the luminescence with H2O2 in a buffer solution with pH 10.

Characterization of Albumin-Luciferin Bioluminescence
We investigated the relationship between the concentration of human albumin and the amount of luminescence. Luminescence exhibited a proportional relationship with the concentration of human albumin within the concentration range of 0−5 mg/mL (Figure 3). The correlation coefficient was 0.996, and the calculated detection limit was 20 μg/mL. The spectrum of the reaction between human albumin and Watasenia luciferin was measured and was found to peak at 540 nm, which differed from the blue luminescence of living W. scintillans ( Figure 4). (B) Relative changes in the luminescence intensity of the albumin-luciferin reaction with or without 200 mM hydrogen peroxide (mean ± SEM, n = 4; one-way ANOVA with Tukey's test, *** p < 0.001, n.s. = not significant).

Characterization of Albumin-Luciferin Bioluminescence
We investigated the relationship between the concentration of human albumin and the amount of luminescence. Luminescence exhibited a proportional relationship with the concentration of human albumin within the concentration range of 0−5 mg/mL ( Figure 3). The correlation coefficient was 0.996, and the calculated detection limit was 20 µg/mL. The spectrum of the reaction between human albumin and Watasenia luciferin was measured and was found to peak at 540 nm, which differed from the blue luminescence of living W. scintillans ( Figure 4).     We then determined the manipulations that would inhibit the luminescence to characterize the albumin-luciferin reaction. When the human albumin was heat-denatured, the amount of luminescence decreased depending on the duration of heat treatment ( Figure 5A). This indicates that albumin loses its luminescence ability when its higher-order structure is destroyed by thermal denaturation. However, the signal in the immunoturbidimetric assay showed no significant decrease, even with the albumin that was heat-denatured for 30 min ( Figure 5B).
When the albumin was treated with different concentrations of trypsin and a serine protease and reacted with luciferin, the amount of luminescence decreased according to the trypsin concentration ( Figure 5C). This indicates that the partial cleavage of the peptide by trypsin leads to the loss of luminescence ability. When the same trypsinized samples were measured through the immunoturbidimetric assay, the signal decreased depending on the trypsin concentration, but the decrease was milder than that in the reaction with luciferin ( Figure 5D). These results indicate that the immunoturbidimetric method is more resistant to the partial cleavage of albumin than the albumin-luciferin reaction. We then determined the manipulations that would inhibit the luminescence to characterize the albumin-luciferin reaction. When the human albumin was heat-denatured, the amount of luminescence decreased depending on the duration of heat treatment ( Figure 5A). This indicates that albumin loses its luminescence ability when its higher-order structure is destroyed by thermal denaturation. However, the signal in the immunoturbidimetric assay showed no significant decrease, even with the albumin that was heat-denatured for 30 min ( Figure 5B).

Detection of Albumin in the Urine of Diabetic Patients Using the Albumin-Luciferin Reaction
Next, because measuring albumin levels in urine is an essential method for the diagnosis of diabetic nephropathy, we evaluated the feasibility of using luciferin to measure trace albumin levels in the urine of 20 patients with diabetes. First, we reacted the urine sample directly with luciferin and measured the luminescence. However, the quantitative results from the direct reaction between urine and luciferin revealed a low When the albumin was treated with different concentrations of trypsin and a serine protease and reacted with luciferin, the amount of luminescence decreased according to the trypsin concentration ( Figure 5C). This indicates that the partial cleavage of the peptide by trypsin leads to the loss of luminescence ability. When the same trypsinized samples were measured through the immunoturbidimetric assay, the signal decreased depending on the trypsin concentration, but the decrease was milder than that in the reaction with luciferin ( Figure 5D). These results indicate that the immunoturbidimetric method is more resistant to the partial cleavage of albumin than the albumin-luciferin reaction.

Detection of Albumin in the Urine of Diabetic Patients Using the Albumin-Luciferin Reaction
Next, because measuring albumin levels in urine is an essential method for the diagnosis of diabetic nephropathy, we evaluated the feasibility of using luciferin to measure trace albumin levels in the urine of 20 patients with diabetes. First, we reacted the urine sample directly with luciferin and measured the luminescence. However, the quantitative results from the direct reaction between urine and luciferin revealed a low correlation with the immunoturbidimetric results ( Figure 6A,B and Figure 7A). This may be because coexisting molecules in urine inhibit the albumin-luciferin reaction. Next, we attempted to purify the protein in urine and react it with albumin to remove foreign substances in the urine. For this purpose, we used ammonium sulfate precipitation, polyethylene glycol (PEG) precipitation, and acetone precipitation.   In each method, the urine sample and standard albumin were precipitated, dissolved again in phosphate-buffered saline (PBS), and reacted with luciferin. The concentration of albumin was calculated and graphed based on the intensity of luminescence ( Figure 6C-E). The results were compared with those of the immunoturbidimetric assay ( Figure 6A) using urine samples; scatter plots were prepared, and correlation coefficients were calculated ( Figure 7B-D). Particularly, we observed a high correlation when acetone  In each method, the urine sample and standard albumin were precipitated, dissolved again in phosphate-buffered saline (PBS), and reacted with luciferin. The concentration of albumin was calculated and graphed based on the intensity of luminescence ( Figure 6C-E). The results were compared with those of the immunoturbidimetric assay ( Figure 6A) using urine samples; scatter plots were prepared, and correlation coefficients were calculated ( Figure 7B-D). Particularly, we observed a high correlation when acetone was used to precipitate the protein in the urine ( Figure 7D). Moreover, each purification method showed a lower value than the albumin quantitative value obtained using the immunoturbidimetric method ( Figure 7A-D). We analyzed the data used in Figure 7D again using the Bland-Altman plot. In urine containing high concentrations of albumin, the luciferin method tended to yield lower readings than immunoturbidimetry (Figure 8). The coefficient of variation (CV) value from the repeated measurements of luminescence from acetoneprecipitated samples of the same concentration was 5.2% (n = 8). In addition, a spike and recovery test was performed, and it was found that the recovery rate was 105%. This result means that the component in urine did not inhibit the reaction.

Discussion
This study demonstrated that Watasenia luciferin specifically reacted with albumin and emitted light (Figure 1). When reacted with purified albumin, it exhibited a very obvious proportional relationship (Figure 3). Since the calculated detection limit (20 μg/mL) was lower than the range of microalbuminuria (30-300 μg/mL), Watasenia lucif erin has a sufficient measurement range and sensitivity to measure trace albuminuria This is a new discovery and indicates that Watasenia luciferin can be used to measure albumin concentrations in liquids. Moreover, after exploring the optimal conditions fo the reaction, 200 mM H2O2 at pH 10 was found to be the best (Figure 2). This result i reasonable because it has been previously shown that oxygen is required for the lumi nescence of W. scintillans [24].
An earlier report described that coelenterazine, a molecule similar to Watasenia lu ciferin, reacts with albumin to emit light [25]. Although the underlying mechanism wa not confirmed, it was predicted that Sudlow site I, to which molecules such as warfarin bind [26], would be the binding site for coelenterazine based on the three-dimensiona structure of albumin. Watasenia luciferin may have a similar mechanism of action. Wa tasenia luciferin reacted more strongly with human albumin than with bovine albumin which is in contrast to the action of coelenterazine [25]. Minor structural differences from coelenterazine may increase the specificity for human albumin. Attempts are underway to modify coelenterazine to develop a molecule that reacts more strongly with albumin [27]. However, there have been no attempts to apply it to medical measurement purpose such as the measurement of albumin in urine.
As depicted in Figure 5, when the albumin was heat-denatured, the albumin

Discussion
This study demonstrated that Watasenia luciferin specifically reacted with albumin and emitted light (Figure 1). When reacted with purified albumin, it exhibited a very obvious proportional relationship (Figure 3). Since the calculated detection limit (20 µg/mL) was lower than the range of microalbuminuria (30-300 µg/mL), Watasenia luciferin has a sufficient measurement range and sensitivity to measure trace albuminuria. This is a new discovery and indicates that Watasenia luciferin can be used to measure albumin concentrations in liquids. Moreover, after exploring the optimal conditions for the reaction, 200 mM H 2 O 2 at pH 10 was found to be the best (Figure 2). This result is reasonable because it has been previously shown that oxygen is required for the luminescence of W. scintillans [24].
An earlier report described that coelenterazine, a molecule similar to Watasenia luciferin, reacts with albumin to emit light [25]. Although the underlying mechanism was not confirmed, it was predicted that Sudlow site I, to which molecules such as warfarin bind [26], would be the binding site for coelenterazine based on the three-dimensional structure of albumin. Watasenia luciferin may have a similar mechanism of action. Watasenia luciferin reacted more strongly with human albumin than with bovine albumin, which is in contrast to the action of coelenterazine [25]. Minor structural differences from coelenterazine may increase the specificity for human albumin. Attempts are underway to modify coelenterazine to develop a molecule that reacts more strongly with albumin [27]. However, there have been no attempts to apply it to medical measurement purposes, such as the measurement of albumin in urine.
As depicted in Figure 5, when the albumin was heat-denatured, the albumin-luciferin reaction significantly reduced the intensity of luminescence. By contrast, the immunoturbidimetric method did not reduce the signal, even with heat-denatured albumin. When the albumin was partially degraded by trypsin, the albumin-luciferin reaction was more susceptible, and the luminescence tended to decrease. These results indicate that Watasenia luciferin reacts with intact albumin but is less likely to react with partially degraded albumin. Specifically, the state of albumin recognized by Watasenia luciferin is slightly different from that recognized using the immunoturbidimetric and HPLC methods.
Among the protein precipitation methods evaluated in this study, the acetone method demonstrated the best correlation with the results of the immunoturbidimetric method ( Figure 7D). This may be because most proteins were precipitated by the acetone method, and foreign substances were successfully removed. The ammonium sulfate and PEG methods may have incompletely removed the coexisting molecules, or the albumin may not have been precipitated completely. Acetone precipitation is often used for the pretreatment of proteomes [28] and is seldom used with the expectation of protein activity after precipitation because it is an organic solvent. A surprising finding was that the ability of the binding to luciferin was retained after acetone precipitation. Although acetone precipitation was successful in this study, centrifugation is a time-consuming process and must be improved if a simpler process is available.
When the results of the immunoturbidimetric method were compared with those of the Watasenia luciferin method after acetone precipitation, a correlation (R = 0.880) was obtained ( Figure 7D); however, the Watasenia luciferin method tended to produce lower values. This is consistent with the fact that Watasenia luciferin is less likely to react with denatured or half-degraded albumin compared with the immunoturbidimetric method ( Figure 5). We found that the luciferin method using acetone precipitation yields lower values compared to immunoturbidimetry, especially in the urine that contains high concentrations of albumin ( Figure 8). Urine with high albumin concentrations may have a high percentage of half-degraded or modified albumin.
Currently, there is a lack of a perfect method for quantifying urinary albumin concentrations. Techniques such as the immunoturbidimetric assay, which is based on the antigen-antibody reaction, cannot detect peptide albumin and do not detect immunounreactive albumin [20]. However, the HPLC method, which is based on the principle of fractionation by molecular weight, can detect immuno-unreactive albumin, but it has the possibility of measuring proteins other than albumin that have similar molecular weights. The luciferin-based detection method we have developed in this study also tends to specifically detect intact albumin but not degenerated albumin. As these methods differ in terms of the state of albumin they detect, multiple methods of measurement can provide detailed information regarding the state of albumin in the urine. Urinary albumin has various states, such as intact, immuno-unreactive, peptide albumin, and dimers, and studies have reported that these states change according to the disease state [29][30][31][32]. A combination of multiple methods may help in clarifying the relationship between urinary albumin status and disease.
The luciferin-based albumin assay is a simple method based on the reaction with a single compound. This is an advantage, and there is no such thing as differences in antibody reactivity due to differences in immunized animals. If a method to purify urinary protein in a short time is developed, it is expected to become a simple and stable method for measuring urinary albumin.

Synthesis of Watasenia Luciferin
To obtain the desired Watasenia luciferin 1, we synthesized pyrazin-2-amine 5 and acetal derivative 8, according to a known method [33]. Commercially available pyrazine 2 was converted into dibromopyrazine 3 by the treatment of N-bromosuccinimide in dimethylsulfoxide and H 2 O. The Negishi cross-coupling reaction of 3 yielded the benzyl derivative 4, which was further transformed into the desired pyrazin-2-amine 5 by the Suzuki-Miyaura coupling reaction in the presence of a catalytic amount of (PhCN) 2 PdCl 2 and 1,4-bis(diphenylphosphino)butane (dppb) as a ligand, as shown in Scheme 1.
The phenolic hydroxyl group of p-hydroxybenzaldehyde 6 was protected with a tert-butyldimethylsilyl (TBS) group, followed by a reduction of the resulting TBS ether with NaBH 4 to produce the benzyl alcohol derivative 7 [34]. The benzyl chloride derivative derived from 7 was then converted into the corresponding Grignard reagent, which had to be further reacted with ethyl diethoxyacetate at −78 • C to obtain the desired acetal 8. The Grignard reagent was prepared according to the method described by Shrestha and Bossmann et al. [33], but the desired acetal derivative 8 was seldom obtained (7% over two steps). However, when the Grignard reagent was prepared in the presence of 1,2-dibromoethane, 75% of the desired acetal derivative 8 was synthesized over two steps, as shown in Scheme 2.
After obtaining the desired pyrazin-2-amine 5 and acetal derivative 8, we examined the synthesis of 1 according to the method described by S. Kojima et al., as shown in Scheme 3 [23]. The mixture of 5 and 8 in EtOH and 10% HCl solution were refluxed to yield 65% of the 3,7-dihydroimidazo [1,2a]pyrazine-3-one derivative 9. Finally, the sulfonation of 9 was performed with sulfur trioxide pyridine complex (SO 3 -pyridine), followed by treatment with sodium methoxide to form the sodium salt of 1. However, the desired sodium salt of 1 could not be obtained, and only a complex mixture was obtained (Scheme 3).
In 2003, M. Adamczyk et al. proposed a mechanism for luminescence via anion B when a series of derivatives A with a 3,7-dihydroimidazo [1,2a]pyrazine-3-one skeleton are deprotonated, as shown in Supplementary Figure S1 [35]. We believed that the failure to obtain the sodium salt of 1 was due to the decomposition of the corresponding bissulfonate anion of 9 by the addition of sodium methoxide, and we decided to attempt other sulfonation reaction conditions. We attempted the synthesis of 1 under acidic conditions, noting that the 3,7-dihydroimidazo [1,2a]pyrazine-3-one derivative 9 was obtained from 5 and 8 in a moderate yield after refluxing with 10% HCl solution. By treating 9 with an excess amount of chlorosulfonic acid, the desired 1 was successfully synthesized with a 65% yield (Scheme 4 and Supplementary Figure S2).
See the supplementary methods for more information.

Urine Samples and Other Materials
Urine samples were collected from 20 patients with diabetes from the University of Toyama Hospital and stored at −20 • C. Before usage, the urine samples were centrifuged at 17,800× g for 5 min to remove insoluble components.

Measurement of Luminescence
The intensity of luminescence was measured using a multiplate reader ARVO X2 (PerkinElmer). The solution used in the reaction consisted of 80 mM Tris (pH 10), 200 mM H 2 O 2 , 200 µM Watasenia luciferin, and 1 mg/mL protein sample (Figures 1, 2 and 5), which was performed in a 96-well plate (Nunclon Delta white microwell, Nunc). Spectrum measurements were performed using SpectraMax i3 (Molecular Device). The intensity of luminescence was measured at wavelengths at every 20 nm width. The solution consisted of 80 mM Tris (pH 10), 200 mM H 2 O 2 , 200 µM Watasenia luciferin, and 1 mg/mL human albumin.

Immunoturbidimetric Assay
N-assay TIA Micro Alb (Nittobo) was used to measure the albumin concentration in urine samples via immunoturbidimetry. According to the manufacturer's instructions, 3.5 µL of urine samples and standards was mixed with 150 µL of buffer R1 in a 96-well plate (BM6000, BM Bio) and incubated at 37 • C for 5 min. Then, 50 µL of buffer R2 was added to the mixture and incubated at 37 • C for 5 min. Absorbance was measured at wavelengths of 700 and 340 nm using a SpectraMax i3 multiplate reader. The signal intensity of the sample was obtained by the subtraction of the OD value at 700 nm from that at 340 nm. Purified human albumin (Nacalai) was used as a standard.

Acetone Precipitation
A mixture of 100 µL of the patient urine sample and 900 µL of the ice-cold acetone was cooled overnight at −20 • C. Next, the mixture was centrifuged at 17,800× g for 30 min, and the resulting precipitate was dissolved in 50 µL of PBS. Purified human albumin was treated in the same manner to be used as a standard.

Ammonium Sulfate Precipitation
Saturated ammonium sulfate solution (900 µL) was mixed with 100 µL of the urine sample and allowed to stand on ice for 1 h. It was then centrifuged at 17,800× g for 30 min, and the resulting precipitate was dissolved in 50 µL of PBS.

PEG Precipitation
PEG 8000 (MP Biomedicals) was dissolved in water to prepare a 50% (w/w) solution. Then, 100 µL of the PEG solution was mixed with 100 µL of the urine sample, placed on ice for 1 h, and centrifuged at 17,800× g for 30 min to precipitate the protein. The precipitate was dissolved in PBS and used for the reaction with luciferase.

Spike-Recovery Test
HSA (100 µg) was added to the patient urine (100 µL), acetone-precipitated, and reacted with luciferin to measure luminescence. The difference between the luminescence of urine with and without HSA was calculated. The same procedure was performed with PBS, and the recovery rate was calculated. This confirmed whether the components in the urine inhibited the reaction of HSA with luciferin.

Statistical Analysis
One-way ANOVA with Dunnett's test or Tukey's test was conducted to determine the statistical significance. The data were expressed as the mean ± SEM. p < 0.05 was considered to indicate statistical significance. In Figures 3 and 7, the approximate straight line was computed by the least-squares method.

Conclusions
Watasenia luciferin specifically reacts with albumin and emits light in proportion to its concentration. We demonstrated that this luminescence can be used as an indicator to detect urinary albumin after acetone precipitation. This is a new principle for quantifying urinary albumin.