n-Alkane Chain Length Alters Dietzia sp. Strain DQ12-45-1b Biosurfactant Production and Cell Surface Activity

Abstract

Upon growth on n-hexadecane (C₁₆), n-tetracosane (C₂₄), and n-hexatriacontane (C₃₆), Dietzia sp. strain DQ12-45-1b could produce different glycolipids, phospholipids, and lipopeptides. Interestingly, cultivation with C₃₆ increased cell surface hydrophobic activity, which attenuated the negative effect of the decline of the emulsification activity. These results suggest that the mechanisms of biosurfactant production and cell surface hydrophobicity are dependent upon the chain lengths of the n-alkanes used as carbon sources.

TEXT

Recently, the biodegradation of crude oil constituents, such as alkanes, through bioremediation of oil-polluted environments (1–3) and microbial enhanced oil recovery (MEOR) technology (4) has received worldwide attention. To date, a number of microorganisms have been reported to degrade alkanes of different chain lengths (5). A critical step in the biodegradation process requires microorganisms to access hydrophobic alkanes by at least two possible mechanisms. First, microorganisms produce surface-active materials, including glycolipids, phospholipids, and lipopeptides, to emulsify alkanes and achieve surfactant-mediated access (6–12). Second, they increase the hydrophobic activity of the cell surface to directly interact with alkanes (5, 13).

Although extensive research has been conducted on the production of different biosurfactants and the cell surface hydrophobic activities, these studies were mainly restricted to alkanes with chain lengths shorter than 18 carbon atoms (C₁₈) (14–18). This raises the question of how bacteria, such as those belonging to the genus Dietzia, access hydrocarbons with chain lengths longer than C₁₈. It is unclear whether accessing longer alkanes requires the production of surface-active materials that are similar to those produced when shorter alkanes (i.e., <C₁₈) are degraded. In addition, whether cell surface hydrophobicity contributes to the accession of longer alkanes is unknown. The aim of this study was, therefore, to address these questions because degradation of alkanes longer than C₁₈ is important for effective MEOR and bioremediation. We used a broad-spectrum alkane-degrading Dietzia sp. strain, DQ12-45-1b (19, 20), and the results of our study revealed that biosurfactant production and cell surface hydrophobic activity changed when different-chain-length n-alkanes were used as the sole carbon sources.

After Dietzia sp. strain DQ12-45-1b was incubated in mineral salt medium (MSM) (21) amended with 0.3% (vol/vol) n-hexadecane (C₁₆) and 0.05% (wt/vol) n-tetracosane (C₂₄) and n-hexatriacontane (C₃₆) as the sole carbon sources, respectively, the cultures were sampled at different time points and analyzed for bacterial growth, alkane degradation, cell surface hydrophobic activities, and emulsifying capacity of the culture broth. The biosurfactants were also extracted from the culture broth, and the moieties of the glycolipid-like biosurfactant were additionally analyzed. The transcripts of the glycolipid synthesis-related genes were also analyzed by real-time reverse transcription-PCR (RT-PCR). All experiments were performed in triplicate with various controls. Detailed experimental procedures and methods are described in the supplemental material.

Strain DQ12-45-1b could degrade C₁₆, C₂₄, and C₃₆ n-alkanes for growth (see Table S1 in the supplemental material), as was previously reported (19, 20). Along with the growth and degradation of alkanes, the surface tension of the culture broth decreased from approximately 62 mN m⁻¹ to 27.78 ± 1.97 and 45.97 ± 2.19 mN m⁻¹ at day 30 for C₁₆ and C₂₄ cultures, respectively, and to 48.66 ± 0.51 mN m⁻¹ at day 45 for the C₃₆ culture (see Fig. S1 in the supplemental material); this finding suggested that shorter alkanes resulted in a higher emulsifying activity, which allowed cells to access alkanes more easily. The C₁₆, C₂₄, and C₃₆ culture broths yielded 95.7, 25.4, and 15.8 mg liter⁻¹ of crude biosurfactant, respectively. Among them, 2 glycolipid compounds (16-A and 16-B) (Fig. 1A) were detected in the C₁₆ culture, while levels of phospholipid and lipopeptide materials were negligible in the C₁₆ culture (Fig. 1B and C). In the C₂₄ culture, glycolipid (24-C) (Fig. 1A) and phospholipid (Fig. 1B) compounds were detected at relatively equal amounts, with no instance of lipopeptide being detected (Fig. 1C). One glycolipid (36-D) (Fig. 1A) and 2 lipopeptide (Fig. 1C) compounds were detected in the C₃₆ culture; however, no phospholipids were detected (Fig. 1B). These results suggested that the production of biosurfactant by strain DQ12-45-1b was related to the length of the hydrocarbons. The biosurfactant activities of these materials were confirmed by oil displacement tests performed after they had been scraped out of the preparative thin-layer chromatography (TLC) plates (see the supplemental material).

Fig 1.

Thin-layer chromatography (TLC) analysis of crude biosurfactant extract from the cultures sampled at day 15. Panels A, B, and C show the presence and relative amounts of glycolipids, phospholipids, and lipopeptides from the n-hexadecane (C₁₆), n-tetracosane (C₂₄), and n-hexatriacontane (C₃₆) cultures, respectively. In panel A, 2 glycolipids (16-A, R_f, 0.59; 16-B, R_f, 0.51) were detected from the C₁₆ culture, whereas glycolipids 24-C (R_f, 0.60) and 36-D (R_f, 0.59) were detected from the C₂₄ and C₃₆ cultures, respectively. In panel B, phospholipids (R_f, 0.62) were detected only in the C₂₄ culture. In panel C, lipopeptides were detectable only in the C₃₆ culture, with 2 compounds detected: CL-a (R_f, 0.79) and CL-b (R_f, 0.56).

Since glycolipid-like materials were detected in all 3 cultures, the structures of the 4 glycolipids (16-A, 16-B, 24-C, and 36-D) were analyzed by gas chromatography-mass spectrometry (GC-MS). Three saccharide moieties were detected, including α-d-glucopyranoside-β-d-fructofuranose (disaccharide unit) for both 16-A and 16-B, 6-deoxy-mannose for 24-C, and 2-methoxime-gluconic acid for 36-D. The fatty acid moieties were also different among the glycolipids. In glycolipid 16-A, only hexadecanoic acid was detected, in contrast to the 5 acids detected in glycolipid 16-B, some of which had unsaturated bonds. Six fatty acids with chain lengths ranging from C₁₂ to C₁₉ and 11 fatty acids with chain lengths ranging from C₁₂ to C₂₄ were detected in 24-C and 36-D, respectively. These fatty acids also contained some unsaturated bonds (Table 1). Of note, the fatty acid moieties in 16-B contained fatty acids with chain lengths longer than C₁₆, indicating that they have originated from fatty acid synthesis processes.

Table 1.

Moieties of the glycolipids produced from C₁₆, C₂₄, and C₃₆ by strain DQ12-45-1b

Fatty acid	Moiety ofa:
	n-C16		n-C24 24-C	n-C36 36-D
	16-A	16-B
C12H24O2	ND	ND	+	+
C14H28O2	ND	+	+	+
C15H30O2	ND	ND	+	+
C16H30O2	ND	ND	ND	+ (7:1)
C16H32O2	+	+	+	+
C17H34O2	ND	ND	ND	+
C18H32O2	ND	+ (9:1, 12:1 and 9:1, 15:1)	ND	+ (9:1, 11:1)
C18H34O2	ND	+ (9:1)	ND	+ (8:1)
C18H36O2	ND	+	+	+
C19H38O2	ND	ND	+	+ (10-methyl)
C24H48O2	ND	ND	ND	+

^aThe saccharide moieties are as follows: n-C₁₆ 16-A and 16-B, α-d-glucopyranoside-β-d-fructofuranose; n-C₂₄ 24-C, 6-deoxymannose; and n-C₃₆ 36-D, 2-methoxime-gluconic acid. ND, not detected. The format “n:1” represents an unsaturated bond at the “nth” carbon atom.

The transcripts of phosphomannomutase (AlgC)-encoding gene homolog YMF1348 (accession no. JQ414011) and NADPH-dependent ketoacyl reductase (RhlG)-encoding gene homolog YMF0365 (accession no. JQ414010), which are key genes in glycolipid biosynthesis (see Table S2 in the supplemental material), were detected by real-time RT-PCR (see Fig. S2 in the supplemental material). Both YMF1348 and YMF0365 were significantly induced by n-alkanes compared to when cells were grown in a medium containing glucose. However, the 2 genes had different patterns of expression. In general, the transcriptional levels of both YMF1348 and YMF0365 were higher in cells grown on C₁₆, with the lowest values being obtained for cells grown on C₃₆, corresponding to the different amounts of glycolipids detected (see Fig. S2). The different transcription levels might be related to the different amounts of the precursors acetyl coenzyme A (acetyl-CoA) and malate in cells, which may suggest the different upregulation of genes in fatty acid biosynthesis, as detected in Alcanivorax borkumensis SK2 (22).

Although glycolipids, phospholipids, and lipopeptides have been reported as key microbial biosurfactants (6, 8, 10, 12), only 1 or 2 types (e.g., glycolipids, phospholipids, or lipopeptides) were simultaneously detected in the degradation of hydrocarbons with chain lengths of <C₁₈ (15, 16, 23). However, strain DQ12-45-1b could produce all the 3 types in various amounts when C₁₆, C₂₄, and C₃₆ were used as the sole carbon sources. Moreover, the simultaneous detection of different glycolipids, both with different saccharide and acid moieties, in a Dietzia strain has not been reported before.

The hydrophobicity of the cell surface, which is measured as the surface hydrophobicity rate, plays an important role in the microbial attachment onto other hydrophobic surfaces, such as solid n-alkanes (24). In this study, the cell surface hydrophobicity rates slightly increased with the incubation time from approximately 0.5% to 8.5% ± 5.5% and 8.3% ± 4.6% for C₁₆ and C₂₄ cultures, respectively, compared to approximately 4% in the control without any hydrocarbon. In contrast, the hydrophobicity rate in C₃₆ cultures could be as high as 23.9% ± 3.7% (Fig. 2), which caused more cells to attach to the solid surface of C₃₆ than to C₂₄ (see Fig. S3 in the supplemental material). These results could be attributed to the fact that glycolipids produced by C₃₆ contain longer organic acid moieties, resulting in higher hydrophobic activities, or that lipopeptides (including the surfactin, iturin, and fengycin classes) changed the cell hydrophobicity and cell contact with hydrocarbons (6); therefore, when C₃₆ was used as the sole carbon source, the decline in the emulsification activity (see Fig. S1 in the supplemental material) was compensated for by an increase in the cell surface hydrophobic activity (Fig. 2), which attenuated the negative impact of the long-chain length of C₃₆ and maintained cell growth, as indicated in our previous studies (19, 20).

Fig 2.

Change in cell surface hydrophobicity rates of Dietzia sp. strain DQ12-45-1b grown on n-hexadecane (C₁₆), n-tetracosane (C₂₄), and n-hexatriacontane (C₃₆), or grown on LB as the control.

On the basis of these results, we hypothesized that bacteria capable of degrading hydrocarbons of various chain lengths have similar functions for producing different biosurfactants and changing cell surface hydrophobic activity. These functions could be attributed to the unique regulation of different genes and pathways. Similar results were reported when investigating the influence of other environmental factors on biosurfactant production (22, 25–28). However, further investigation is required to verify this hypothesis.

Nucleotide sequence accession numbers.

The nucleotide sequences of YMF0365 and YMF1348 have been deposited in the GenBank database under accession no. JQ414010 and JQ414011, respectively.