2024 Volume 49 Issue 4 Pages 271-276
When a test substance is poorly water-soluble, it can be adsorbed onto silica gel to facilitate dispersibility in a ready biodegradability test. To uniformly adsorb the test substance onto silica gel, the substance is dissolved in a solvent and then mixed with the silica gel. It is desirable for the solvent to completely evaporate afterward. In this study, we identified n-hexane as a suitable solvent for this purpose. Furthermore, through fluorescence observation, we revealed that the test substance adsorbed onto the silica gel adhered to activated sludge flocs. This is thought to improve contact between the test substance and microorganisms, thereby accelerating biodegradation.
Once persistent substances are released into the environment, they can remain for a long time and have the potential to accumulate in organisms or cause adverse effects on the environment.1) For appropriate management of chemical substances, it is important to understand their degradability. Under the Chemical Substances Control Law (CSCL) in Japan, the biodegradability of a test substance has been mainly evaluated based on biochemical oxygen demand (BOD) while the test substance and mixtures of microorganisms such as activated sludge are cultivated.2,3) The degradability test under the CSCL has been conducted in accordance with the Organisation for Economic Co-operation and Development (OECD) test guideline 301C.2,3) The 301C test requires that the concentration of the test substance shall be 100 mg L−1. Consequently, poorly water-soluble substances are being evaluated for their biodegradability without being sufficiently dissolved. After revision of the CSCL in 2019, the OECD test guideline 301F can also be used for the evaluation of biodegradability.3,4) In this test, auxiliary substances, referred to as additives, such as carriers, emulsifiers, and solvents are used to improve the contact between microorganisms and the test substance in the test medium.3,4) However, neither the CSCL nor the OECD 301 provides examples of the carriers; they are only listed in the International Organization for Standardization (ISO) 10634,5,6) which the OECD 301 refers to. The CSCL and the OECD 301 require that additive substances must not be biodegradable (here, this is referred to as the “ideal standard”). In contrast, although the ISO 10634 states that additive should be non-biodegradable, it adds that the biodegradation of additive does not exceed 10% compared to the test substance (here, this is referred to as the “realistic standard”).5,6)
The ISO 10634 uses silica gel as an example of a carrier and chloroform or acetone as an example of a solvent to dissolve test substances. We had revealed that the biodegradation of poorly water-soluble substances such as octabenzone (OB), anthraquinone, 2-ethylanthraquinone, bis(2-ethylhexyl)phthalate, and tris(2-ethylhexyl)trimellitate was accelerated by adsorption onto silica gel.7) According to the ISO 10634, the test substances were dissolved in chloroform and mixed with silica gel. After evaporation of chloroform, the biodegradability tests were conducted. However, since chloroform did not completely evaporate from silica gel, the increase in observed BOD was due to the residual chloroform only. It meant that the additive was biodegraded. This deviated from the ideal standard. There were two drawbacks even if it met the realistic standard. One problem was that the amount of the residual chloroform could not be controlled. Thus, the biodegradation rate of the test substance could not be evaluated based on the BOD because it was not possible to identify how much of the measured BOD came from the biodegradation of the test substance and how much came from chloroform. This problem also applies to residual solvents other than chloroform. Another problem was that some microorganisms that could biodegrade the residual solvent might grow in abundance. Should these microorganisms also biodegrade the test substance, it would suggest that the residual solvent is an accelerant of the test substance biodegradation process. This would imply that the true biodegradability of the test substance has not been properly evaluated.
In this study, we searched for appropriate organic solvents to be mixed with silica gel to meet the ideal standard or the realistic standard. We selected chloroform and acetone, which are listed in the ISO 10634, and n-hexane, which is not listed in the ISO 10634. As to the test substance, OB was selected because we had already reported that biodegradation of the substance was accelerated by adding silica gel.7) In addition, the metagenomic analysis was performed to determine if the microorganisms biodegrading the residual solvent are the same as those biodegrading the test substance (OB). Furthermore, the following hypotheses concerning why the addition of silica gel accelerates the biodegradation of the test substance were verified: 1) The increased surface area on which the test substance can dissolve, leads to an increased rate of dissolution. 2) Silica gel acts as a scaffold for microorganisms for promoting their growth. 3) Microorganisms adhere to the silica gel, allowing direct utilization of adsorbed test substance.
As a test substance OB was purchased from Tokyo Chemical Industry (Tokyo, Japan). Figure 1 shows the chemical structure of OB. n-Hexane (HPLC grade), K2HPO4, KH2PO4, Na2HPO4·12H2O, NH4Cl, MgSO4·7H2O, CaCl2, FeCl3·6H2O, 0.5% phosphate solution, and 1 M sodium hydroxide were purchased from FUJIFILM Wako Pure Chemical (Osaka, Japan). Acetone (guaranteed reagent grade) and chloroform (extra pure reagent) were purchased from Nacalai Tesque (Kyoto, Japan). Aniline was purchased from Kanto Chemical (Tokyo, Japan) as a reference substance in the biodegradability test. Anthracene and silica gel (5 to 25 µm particle size for thin-layer chromatography, surface area: ca. 500 m2 g−1) were purchased from Sigma-Aldrich (St. Louis, MO, USA).
The procedure for the adsorption of OB onto silica gel followed our previous report.8) Silica gel was mixed with either acetone, chloroform, or n-hexane. Each time after the mixing, the solvent was evaporated by using a rotary evaporator R-300 (Buchi Labortechnik AG, Flawil, Switzerland) and then a vacuum oven AVO-250NB (AS ONE Corp., Osaka, Japan) for 14 days. Until the ready biodegradation study was initiated, the silica gel was sealed in a glass vial and stored in a refrigerator EP-570 (Nihon Freezer, Tokyo, Japan) at 4±1°C.
3. Ready biodegradability testThe ready biodegradability test was carried out in accordance with OECD guideline 301F2) as follows: activated sludge was collected from a sludge return line at the municipal sewage treatment plant (Osaka, Japan), receiving predominantly domestic sewage. The collection days were Apr 14 in 2020 (for the ready biodegradability test with silica gel and acetone) and June 22 in 2021 (for the test with silica gel and chloroform or n-hexane). The test medium was prepared as described in our previous report,8) where the final concentrations of K2HPO4, KH2PO4, Na2HPO4·12H2O, NH4Cl, MgSO4·7H2O, CaCl2, FeCl3·6H2O were 85, 217.5, 672.1, 5, 22.5, 27.5, and 0.25 mg L−1, respectively. The final concentration of activated sludge in the test medium was 30 mg L−1, and the test substance was added up to a concentration of 100 mg L−1 either directly or adsorbed onto silica gel at a concentration of 1,333 mg L−1. The test medium was added up to 900 mL, stirred, and cultivated in darkness at 22±1°C. BOD was measured continuously with a coulometer OM7000A (Ohkura Electric, Saitama, Japan). The biodegradation rate of OB was calculated using Eq. (1) in the absence of silica gel.
![]() | (1) |
where BODtest is the BOD (measured value, mg) of a test vessel containing OB and activated sludge in the mineral medium. BODcontrol is the BOD (measured value, mg) of a control vessel containing activated sludge in the mineral medium. ThODtest is the theoretical oxygen demand (calculated value, mg), which is the total amount of oxygen required to oxidize OB (230 mg9)). The biodegradation rate of OB, when adsorbed onto silica gel, was calculated using Eq. (2).
![]() | (2) |
where BODtest+additive is the BOD (measured value, mg) of a vessel containing OB, silica gel, and activated sludge in the mineral medium (hereafter referred to as the “test+additive vessel”). The biodegradation rate of additive (i.e., residual organic solvent adsorbed onto silica gel) relative to ThODtest value was calculated using Eq. (3), in order to check if the rate met the realistic standard.
![]() | (3) |
where BODadditive is the BOD (measured value, mg) of an additive vessel containing silica gel and activated sludge in the mineral medium. The biodegradation rates in Figs. 2, 3, S1, and S2 were calculated using Eqs. (1)–(3).
The ready biodegradability tests to investigate the effect of mixing organic solvents with silica gel were conducted as follows: for acetone, two test vessels containing OB itself, two test+additive vessels containing OB adsorbed onto silica gel, one additive vessel containing silica gel, to which acetone had been added and then removed from the mixture by evaporation, and one control vessel were cultivated for 38 days. For chloroform and n-hexane, three test vessels, three test+additive vessels, two additive vessels, and two control vessels were cultivated for 42 days. For each test, one test vessel containing aniline as a reference substance was cultivated to confirm the biodegrading activity of activated sludge, ensuring that the validity criterion was fulfilled (biodegradation rate exceeding 60% after 14 days cultivation).
4. Metagenomic analysisActivated sludge containing about 4.5 mg of suspended solids was collected and centrifuged at 15,000 rpm for 20 min in a centrifuge (Eppendorf Himac CF16RXII, Ibaraki, Japan). After removing the supernatant, the sludge pellet was stored at −20°C. Bioengineering Lab (Kanagawa, Japan) conducted DNA extraction, PCR, and next-generation sequencing of the sludge pellet, as described in our previous report.8) The V3–V4 hypervariable region of 16S rRNA was sequenced by the MiSeq System (Illumina, San Diego, CA, USA) with read lengths of 2×300 bp, and representative sequences were output with QIIME2 (ver.2022.8).10) The EzBioCloud 16S database was used for taxonomic classification. Sequencing data were deposited in DNA Databank of Japan (DDBJ accession no. PRJDB16891). The quality of sequencing was as follows: the minimum number of valid pair reads was 3.2×104 (average 4.8×104) and the minimum Q20 and Q30 scores were 83.1% and 70.4% (average 89.6% and 79.7%), respectively.
To accurately identify the degraders of a test substance, our previous study8) showed that it was effective to examine the microbiota when the test substance was undergoing active biodegradation. Thus, activated sludge in the test or test+additive vessel was collected when each test substance was actively undergoing biodegradation, while activated sludge in the control vessel was collected at around the same time as activated sludge in the test+additive vessel. Figures S1 and S2 show the time points of activated sludge collection (closed diamond: ♦).
To investigate the microorganisms relevant to the biodegradation of OB or silica gel or both, the 10 dominant families were identified using the following procedures: 1) subtracting the relative abundance of each family in the control vessel from that in the test vessel, the test+additive vessel, or the additive vessel as determined by each ready biodegradation test; 2) selecting the top ten families on the basis of the highest subtracted relative abundance values obtained from testing biodegradability in the test vessels, the test+additive vessels, and the additive vessels used to investigate the effectiveness of additives as facilitators of OB degradation.
5. Distribution of test substance onto silica gel and microorganismsTo investigate the distribution of test substances and microorganisms adsorbed onto silica gel, fluorescence microscopy observation was conducted in accordance with our previous report,9) with anthracene as the test substance. The presence of anthracene only was detected using an excitation wavelength of 385 nm, and that of both anthracene and activated sludge was detected using excitation wavelengths of 385 nm and 475 nm. We used the method of adsorption onto silica gel described above and chloroform as the adsorption solvent due to the good solubility of anthracene in chloroform.
The average time-dependent biodegradation rates of OB, OB adsorbed onto silica gel with acetone, and silica gel with acetone for 38 days are shown in 1, 2a, and 3a in Fig. 2, respectively. Results of each replicate are shown in Fig. S1. OB (1 in Fig. 2) started to degrade at Day 29 (hereinafter defined as the time point when the biodegradation rate rose to 2%) and the rate reached 18.5% on average at Day 38. The biodegradation rate of OB adsorbed onto silica gel with acetone (2a in Fig. 2) increased gradually after Day 3, remarkably after Day 16, and reached 53.7% on average at Day 38. This result showed that the biodegradation rate of OB is accelerated by adsorption onto silica gel as reported previously.7) The biodegradation rate of silica gel with acetone (3a in Fig. 2), which was calculated using Eq. (3), increased gradually after Day 3 and reached 7.8% at Day 38. Since this value is below the realistic standard (i.e., 10%, see Introduction), use of acetone as a solvent for adsorption of poorly water-soluble substances onto silica gel may be applicable to the evaluation of the biodegradability of these substances. However, this method has the following two drawbacks. One problem is that the biodegradation rate of a test substance itself cannot be determined from BOD by this method. The biodegradations of 2a in Fig. 2 coincided with that of 3a in Fig. 2 after Day 3. The degradation after Day 3 was not observed for 1 in Fig. 2 or the control vessel (data not shown). Thus, the biodegradations after Day 3 were apparently caused by acetone. It is possible that the biodegradation rate of OB itself might be theoretically calculated by subtracting the biodegradation rate of 3a from that of 2a in Fig. 2. However, it is not guaranteed that the amount of residual acetone in the test+additive vessel is the same as that in the additive vessel because the amount after evaporation is uncontrollable. Another problem is that the increase in abundance of some microorganisms with the biodegradation of residual acetone might accelerate the biodegradation of the test substance.
The average time-dependent biodegradation rate of OB for 42 days is shown in 1 in Fig. 3. That of OB adsorbed onto silica gel with chloroform or n-hexane is shown in 2c or 2h in Fig. 3, respectively. That of silica gel with chloroform or n-hexane is shown in 3c or 3h in Fig. 3, respectively. Results of each replicate are shown in Fig. S2. The biodegradation rate of OB started to increase at Day 30, and reached 10±10% (1 in Fig. 3). This tendency was similar to that in Fig. 2. The biodegradation rates of 2c and 2h in Fig. 3 increased remarkably after Day 16, as in Fig. 2, and reached 71±4% and 70±10% for 2c and 2h in Fig. 3, respectively. These results revealed that any of the solvents (acetone, chloroform, or n-hexane) used for adsorption accelerated the biodegradation of OB to the similar extent. The biodegradation rate of 3c in Fig. 3 on average increased at Day 2 to 3.7% and reached a plateau. It was considered that the residual chloroform was biodegraded within two days. On the other hand, the biodegradation rate of 3h in Fig. 3 remained less than 0.6% on average. It meant that the amount of residual n-hexane was negligible or did not exist in the vessels. Thus, chloroform did not meet the ideal standard but met the realistic standard as a solvent for chemical adsorption onto silica gel. On the other hand, n-hexane was considered to meet both the ideal and realistic standards.
Henry’s law constant of acetone or chloroform is higher than that of n-hexane. However, a small amount of acetone and chloroform but not n-hexane remained after evaporation. It appeared that acetone and chloroform made dipole–dipole interaction and formed hydrogen bonds with silanol on the surface of silica gel. On the other hand, n-hexane did not form hydrogen bonds due to its nonpolar nature. Thus, almost no n-hexane remained on the silica gel surface. As a result, n-hexane is a suitable solvent to adsorb OB onto silica gel for the ready biodegradation test. By using n-hexane, the biodegradation rate of OB can be calculated based on the measured BODs. In addition, there is no suspicion that microorganisms that biodegrade n-hexane might accelerate the biodegradation of OB.
The OECD 301 states that solid carriers are not recommended for solid test substances but may be suitable for oily substances.2) Indeed, when the test substance is oily, it can be mixed directly with the carrier without dissolving in an organic solvent, thus allowing for an assessment of the biodegradability. On the other hand, when the test substance is solid, it needs to be dissolved in an organic solvent first. The ISO 10634 lists acetone and chloroform as examples of such solvents. However, as demonstrated in our study, these solvents could not be completely removed from the carrier. Nevertheless, our study found that when n-hexane is used as a solvent, it can be completely removed. Therefore, it has been shown that even if the test substance is a solid, its biodegradability can be reliably assessed by dissolving it in n-hexane and adsorbing it onto the carrier.
2. Metagenomic analysisTable 1 shows the relative abundances of the top ten families for OB. Variations in the relative abundance were examined at various taxonomic levels. Since the substantial difference was confirmed at the family level, the following discussion concerns differences in relative abundance at the family level. The relative abundances were calculated by subtracting the abundances of families in the control vessel and then averaging the corresponding replicates. The original relative abundances of the top ten families for the test, test+additive, additive, and control vessels as well as Day 0 vessel (i.e., before initiation of the test) are shown in Tables S1, S2, and S3 for acetone, chloroform, and n-hexane as the solvent, respectively. Dominant families that increased more than 5% through the biodegradation tests were Mycobacteriaceae, Nocardiaceae, PAC002126_f, Sphingomonadaceae, Comamonadaceae, PAC000016_f, or Saccharimonas_f (1, 2a, 2c, and 2h in Table 1). Several bacteria in these families were considered to have contributed to the biodegradation of OB. Although acetone was biodegraded in the additive vessel, none of families showed particularly increased relative abundance (3a in Table 1) because of the considerably biodegradable nature of acetone.11–14) In the case of chloroform, Flavobacteriaceae was increased up to 13.4% (3c in Table 1). This family could be a biodegrader of chloroform. However, this family was hardly expected to be a contributor to the biodegradation of OB because of the negligible increase in relative abundance of this family in the other vessels. In the case of n-hexane, none of families showed a specific increase in relative abundance (3h in Table 1). This finding supported that, with hardly any of it remaining in the vessels, n-hexane had no influence on the microbiota biodegrading the test substance. To sum up, acetone and chloroform remained in the test vessels and were biodegraded by some bacteria. Chloroform increased the relative abundance of a specific bacterial family that did not seem to contribute to the biodegradation of OB. Therefore, it was considered that the acceleration of OB biodegradation was due solely to the dispersibility of silica gel. n-Hexane was the more suitable organic solvent to use for mixing a test substance with silica gel because the biodegradation rate of OB itself could be determined from BOD.
Taxonomy | Relative abundancea) (%) | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
Phylum | Class | Order | Family | OB | +Silica gel | |||||
1 | 2a | (3a) | 2c | (3c) | 2h | (3h) | ||||
Actinobacteria | Actinobacteria_c | Corynebacteriales | Mycobacteriaceae | 7.3 | 5.6 | (−4.4) | 25.2 | (−1.8) | 7.4 | (−0.8) |
Actinobacteria | Actinobacteria_c | Corynebacteriales | Nocardiaceae | 0.9 | 0.0 | (0.0) | −0.1 | (0.1) | 6.0 | (−0.1) |
Bacteroidetes | Flavobacteria | Flavobacteriales | Flavobacteriaceae | 1.0 | 0.2 | (1.9) | −4.3 | (13.4) | −4.4 | (−3.2) |
Proteobacteria | Alphaproteobacteria | Caulobacterales | Caulobacteraceae | 1.2 | 1.6 | (1.5) | 0.0 | (1.4) | −0.5 | (0.7) |
Proteobacteria | Alphaproteobacteria | Rhizobiales | Bradyrhizobiaceae | −0.4 | −0.6 | (−0.5) | 1.4 | (0.1) | 4.6 | (−0.1) |
Proteobacteria | Alphaproteobacteria | Rhizobiales | PAC002126_f | 0.7 | 1.5 | (0.0) | 7.6 | (1.0) | 5.8 | (0.3) |
Proteobacteria | Alphaproteobacteria | Sphingomonadales | Sphingomonadaceae | 2.8 | 10.9 | (0.3) | 5.0 | (0.2) | 5.3 | (0.6) |
Proteobacteria | Betaproteobacteria | Burkholderiales | Comamonadaceae | 0.1 | 9.4 | (0.2) | 1.4 | (3.5) | 1.8 | (0.5) |
Saccharibacteria_TM7 | Saccharimonas_c | Saccharimonas_o | PAC000016_f | −0.8 | −0.7 | (−0.5) | 1.7 | (0.1) | 7.4 | (0.9) |
Saccharibacteria_TM7 | Saccharimonas_c | Saccharimonas_o | Saccharimonas_f | 12.7 | 16.9 | (0.0) | 0.6 | (0.0) | 16.8 | (0.0) |
a)Relative abundance on average subtracting the control of each test.
Figure 4 shows fluorescence images of anthracene adsorbed onto silica gel and microorganisms after five days cultivation. Under the excitation wavelength of 385 nm, the blue fluorescent anthracene on the silica gel existed in clumps in the test medium (Fig. 4A). When the stained microorganisms were also excited to emit green fluorescence by incident light of wavelength of 474 nm, the distribution of anthracene corresponded exactly with that of the activated sludge floc (i.e., aggregates of microorganisms, Fig. 4B). Thus, anthracene adsorbed onto silica gel seemed to adhere to the floc. When the transmitted light and fluorescence were merged, blue fluorescence was observed on the silica gel surface, and silica gel particles were observed to be distributed on the floc (Fig. 4C).
Based on these results, we investigated why the addition of silica gel accelerated the biodegradation of the test substance. We hypothesized three scenarios: 1) Increase in the surface area enhances the amount of test substance that can dissolve and leads to an increased rate of dissolution. 2) By acting as a scaffold for microorganisms, silica gel promotes their growth. 3) The adherence of microorganisms to the silica gel allows them to directly utilize the adsorbed test substance. As for 2), if microbial growth were enhanced, increases in BOD and changes in relative abundance would be observed; however, no such indicative results were obtained (3h of both Fig. 3 and Table 1). Regarding 3), since the test substance is adjacent to the microorganisms (Fig. 4), this possibility cannot be denied. As for 1), the increase in the dissolvable surface area can be mathematically explained as follows: First, the surface area of the test substance without the addition of silica gel can be calculated from the volume and density of the crystals, as the test substance is not dissolved. The crystal size of OB was estimated as 1×0.2×0.2 mm,15) which generally coincided with our crystal size estimates by microscopic examination (data not shown). The density of OB, calculated by ACD/Labs (ver. 11.02), is 1.07±0.06 g cm−3, so the surface area per 1 mg of the test substance is estimated to be 0.2 cm2 mg−1 [=0.0088÷(0.00004×1.07×1,000)]. Next, the surface area of the test substance when adsorbed onto silica gel can be calculated from the Brunauer–Emmett–Teller specific surface area of silica gel, which is about 500 m2 g−1. Assuming that the test substance is evenly adsorbed onto the silica gel, the surface area of silica gel per 1 mg of the test substance would be 7×104 cm2 mg−1[=1,333×(500×10)÷100]. Therefore, by adsorbing the test substance onto silica gel, the surface area of the test substance increases by an order of 105. However, in reality, the test substance was adsorbed unevenly onto silica gel (Fig. 4C), so the difference may not be as significant. In any case, it can be considered that the addition of silica gel significantly increased the surface area of test substance dissolution, thereby enhancing the supply of the test substance to microorganisms. As a result, the interaction between the test substance and microorganisms improved, which accelerated its biodegradation by microorganisms.
As for solvents facilitating adsorption onto silica gel, n-hexane was revealed to be suitable for OB and not only accelerated the biodegradation of OB but also met the ideal standard. Although small amounts of acetone and chloroform remained on silica gel surface, they met the realistic standard based on the extent of biodegradation. Based on the metagenomic analysis, acetone did not increase the abundance of any specific families, suggesting its potential use in the biodegradability test. On the other hand, chloroform increased the abundance of Flavobacteriaceae, suggesting it should not be used if this family contributes to the degradation of the test substance. Furthermore, it was revealed that the test substance, when adsorbed onto silica gel, adhered to the activated sludge floc. Adding silica gel increased the surface area where the test substance dissolved. This made more of the substance available to microorganisms. As a result, the interaction between the microorganisms and the substance improved, which accelerate its biodegradation by microorganisms.
The online version of this article contains supplementary material, which is available at https://www.jstage.jst.go.jp/browse/jpestics/.