Enhanced cell surface hydrophobicity favors the 9α‐hydroxylation of androstenedione by resting Rhodococcus sp. cells

Abstract

The achievement of an effective process of 9α‐hydroxylation of 4‐androstene‐3,17‐dione is of significant importance as it leads to the formation of the key intermediate 9α‐hydroxy‐4‐androstene‐3,17‐dione which is not possible by chemical means. In this study, the 9α‐hydroxylation of 4‐androstene‐3,17‐dione was carried out by resting Rhodococcus sp. cells. The ability of the naturally hydrophobic Rhodococcus to assimilate n‐alkanes was employed to obtain a cell depot with an intentionally increased cell surface hydrophobicity. The control Rhodococcus sp. cells were cultivated on medium containing glucose instead of n‐alkanes as a source of carbon and energy. Cells were harvested, washed from the cultivation media, and subjected to transformation of crystal androstenedione in buffer medium. The hydrophobicity of the n‐alkanes‐ and glucose‐grown cells, their total lipid content, and fatty acid composition were determined. The ultrastructure of the n‐alkanes‐ and glucose‐grown cells and their steroid hydroxylating activities were examined and compared. The results obtained in the present study showed that the intentionally achieved growth‐driven enhancement of the already hydrophobic Rhodococcus sp. cells made them even more compatible with the hydrophobic steroid substrate and enhanced its accessibility, which provided an increased steroid hydroxylating activity and lack of the accompanying product destruction.

Abbreviations

9α‐OH‐AD

9α‐hydroxy‐4‐androstene‐3,17‐dione

9α‐OH‐ADD

9α‐hydroxy‐4‐androstadiene‐3,17‐dione

AD

androstenedione; 4‐androstene‐3,17‐dione

ADD

4‐androstadiene‐3,17‐dione

isom

isomeric

Me

branched chain

SEM

Scanning electron microscopy

TEM

Transmission electron microscopy

1. Introduction

The achievement of an effective process of microbial 9α‐steroid hydroxylation is in the focus of interest of pharmaceutical industry as an essential step of the steroid drugs manufacturing allowing for a shorter way to the synthesis of fluorinated corticosteroids 1.

There are different approaches applied for achievement of an effective microbial transformation of steroid compounds 2. Most of them attempt to overcome the main bottlenecks in the process performance, either the low water solubility of the steroid substrate or the destruction of the desired product, or both. In our previous studies of the 9α‐hydroxylation of AD by resting Rhodococcus sp. cells we dissolved AD in water miscible organic solvents 3, applied it as water suspension obtained by ultrasonification 4, used non‐ionic surfactants as mediators of substrate solubility 5, aqueous‐organic solvent two‐phase systems as well as organic solvent as a sole component of the reaction medium 6. Studying inducibility of the Rhodococcus sp. 9α‐steroid hydroxylase we managed to avoid the destruction of the 9α‐OH‐AD and demonstrated almost complete transformation of AD into the desired product 7. Based on all our previous studies we are inclined to make the conclusion that the achievement of robust cell‐substrate contact is a significant factor allowing for increasing of their 9α‐steroid hydroxylating activity.

The present investigation is a continuation of our experiments to find conditions for an effective 9α‐hydroxylation of AD with resting Rhodococcus sp. cells. The hypothesis was that the preparation of a cell depot of Rhodococcus sp. cells with an additionally increased hydrophobicity and application of a crystal AD would favor the effectiveness of the 9α‐steroid hydroxylation reaction. To enhance hydrophobicity, we cultivated the Rhodococcus sp. cells on media containing n‐alkanes as a sole source of carbon and energy while control cells were cultivated on glucose. The hydrophobicity, the total lipid content, and fatty acid composition of the n‐alkanes‐ and glucose‐grown cells were determined. Their ultrastructure and steroid hydroxylating activities were examined and compared.

2. Materials and methods

The Rhodococcus sp. strain, medium and inoculum preparations, and cultivation conditions were described previously 4. n‐Alkanes, 2% (v/v), were added into the cultivation medium instead of glucose before its inoculation. The n‐alkanes mixture contained the following hydrocarbons: C9 ‐ 0.27%, C10 ‐ 1.21%, C11 ‐ 22.8%, C12 ‐ 37%, C13 ‐ 25.4%, C14 ‐ 0.22%, C15 ‐ 0.79%, C16 ‐ 0.4%, C17+C18 <0.1%, and traces of C19+C20.

Transformation of AD, analytical procedures, SEM and TEM procedures were described previously 5. The cell surface hydrophobicity was determined by microbial cell adhesion to hydrocarbons 8. The crystal AD was added in reaction medium in concentration 1 mg/mL.

The fatty acid composition was determined by gas chromatography‐mass spectrometry after extraction and transmethylation of the total lipids according to Cristie 9. An Agilent 6890 Plus System with 5973 mass selective detector and HP‐5MS capillary column, 60 m × 0.25 mm × 0.25 μm (Agilent Technologies, USA) were used. Helium was the carrier gas at a flow rate of 0.8 mL/min. Hydrocarbons were analysed by temperature gradient started from 80°C with 20 K/min to 150°C, then with 5 K/min to 320°C and held at this temperature for 15 min. Fatty acids methyl esters were measured by temperature gradient started from 150°C with 2 K/min to 210°C, then with 5 K/min to 280°C and 30 min held at that temperature. T_inj was 280°C and T_aux 300°C. Injection volume was 1 μL; split 20:1. Peaks identification was according to mass spectral library.

All numeric data presented are mean values of three independent experiments.

3. Results and discussion

The cell surface hydrophobicity of n‐alkanes‐grown cells was determined as 100%, of glucose‐grown ones it was 72%. The total amount of cell lipids was 63 mg/g in n‐alkanes‐grown cells and 11 mg/g in glucose‐grown ones. The same fatty acids are found in the cells, irrespectively of the carbon and energy source, albeit at different ratios (Table 1). The total amount of the saturated and unsaturated fatty acids and the isomeric forms are almost equal. The amount of the branched chain fatty acids, however, is 3.6 times higher in glucose‐grown cells. The even number fatty acids are better presented in glucose‐grown cells with the exception of 12:0, which is about six times more in the n‐alkanes‐grown ones. Odd number fatty acids in glucose‐grown cells are almost twice the amount in n‐alkanes‐grown ones with the most significant differences observed in 13:0 and isomeric 19:1.

Table 1.

Fatty acid composition of Rhodococcus sp. glucose‐ and n‐alkanes‐grown cells

Fatty acid	In glucose‐grown cells, %	In n‐alkanes‐grown cells, %
9:0	Traces	0.4
10:0	0.1	1.0
11:0	Traces	0.9
12:0	0.4	2.5
13:0	0.2	13.7
14:0	5.2	4.9
14:1	0.3	0.1
15:0	3.2	5.7
15:1	0.4	0.4
16:0	16.7	9.2
Me‐16:0	0.1	0.1
isom. 16:1	11.9	4.1
Me‐16:1	0.1	0.1
17:0	1.7	3.2
Me‐17:0	3.0	2.2
isom. 17:1	9.1	6.7
18:0	2.5	3.1
Me‐18:0	19.4	3.7
isom. 18:1	13.6	11.9
18:2	1.0	1.8
19:0	0.6	0.5
Me‐19:0	0.8	0.4
isom. 19:1	3.4	13.5
20:0	0.5	1.0
isom. 20:1	3.6	4.1
21:0	0.1	0.6
isom. 21:1	0.2	1.3
22:0	0.8	0.5
isom. 22:1	0.4	0.8
23:0	0.1	0.4
isom. 23:1	Traces	0.6
24:0	0.3	0.3
isom. 24:1	0.2	0.2
25:0	Traces	0.1
isom. 25:1	Traces	0.1
26:0	Traces	Traces
isom. 26:1	0.1	Traces

It is worth to relate the length of fatty acids detected in the n‐alkanes‐grown Rhodococcus sp. cells to the chain length of the hydrocarbons in the n‐alkanes mixture used as a source of carbon and energy in the cultivation medium. Thus, the hydrocarbons with 10 to 13 C atoms dominate in the n‐alkanes mixture and represent 86.4% of all hydrocarbons. The fatty acids with 10 to 13 C atoms in their chains represent 18.1% in the n‐alkanes‐grown cells and 0.7% of the whole fatty acids in glucose‐grown cells ones. This means that the amount of C10‐C13 fatty acids in n‐alkanes‐grown cells is 26 times higher compared to glucose‐grown ones. This observation is in a good correlation with the earlier reports that the hydrocarbon‐utilizing bacteria contained considerable amounts of fatty acids with chain lengths equivalent to those of the hydrocarbon substrate 10. Even long‐chained haloalkanes are incorporated into the cellular fatty acids by Rhodococcus rhodochrous cells 11.

As seen on SEM micrographs, the glucose‐grown cells (Fig. 1A and 1B) are with a dense surface and distinct even outlines. They are covered by an amorphous material, which is amassed on certain surface zones. Somewhere a separation of the amorphous material from the surface is also observed.The n‐alkanes‐grown cells (Fig. 1C and 1D) are with a reduced turgor. They are deformed and with curved uneven surfaces. The cells’ outlines are not smooth and some concave sites appeared on them. At some places, the surface membranes are disrupted.

Figure 1.

SEM of glucose‐ (A, B) and n‐alkanes‐grown (C, D) Rhodococcus sp. cells. Bar = 0.1 μm.

TEM investigations revealed some more differences in the ultrastructure of the Rhodococcus sp. cells. Cell surface membranes of glucose‐grown cells are fitted closely to the cytoplasm and an electron dense material is found to cover them (Fig. 2A and 2B). In few of the cells are seen small transparent zones located in the nucleoid area (Fig. 2A). Fibrilar structures (Fig. 2B) similar to those in Mycobacterium cells 12 are observed in some of the cells. A peculiar characteristic of the Rhodococcus sp. cells grown on n‐alkanes (Fig. 3A, B and C) is the presence of large electron‐transparent inclusion bodies in most of the cells. They are located in the nucleoid regions (Fig. 3A) as well as closely to the disrupted cell surface membranes (Fig. 3B) and in the area of mesosome (Fig. 3C). Our results are in correlation with reported earlier observations that the cultivation on hydrocarbons as a sole source of carbon and energy leads to appearance of large electron‐transparent inclusion bodies in Rhodococcus opacus 13 as well as in changes in the ultrastructure of Rhodococcus erythropolis 14.

Figure 2.

TEM of glucose‐grown Rhodococcus sp. cells. Electron dense cell surface (A) and fibrilar structures (B). Bar = 0.5 μm.

Figure 3.

TEM of n‐alkanes‐grown Rhodococcus sp. cells. Electron‐transparent inclusion bodies located in the nucleoid regions (A), closely to the disrupted cell surface membranes (B), and in the area of the mesosome (C). Bar = 0.5 μm.

Figure 4 presents the dynamics of the 9α‐steroid hydroxylation reaction carried out by resting glucose‐ and n‐alkanes‐grown Rhodococcus sp. cells. As seen from the curves, the two types of cells reach the maximum in the product accumulation for about 30 h. There is a lag‐phase in the curve of 9α‐OH‐AD accumulation by glucose‐grown cells and the amount of the accumulated 9α‐OH‐AD is twice as low as the one accumulated by n‐alkanes‐grown Rhodococcus sp. cells. After reaching the peak, the amount of the 9α‐OH‐AD accumulated by glucose‐grown cells gradually decreases which is due to the reported previously consecutive induction of the 9α‐hydroxylating activity and the accompanying product degrading Δ¹‐steroid dehydrogenating activity in the resting Rhodococcus sp. cells 7. It is known from the literature that bacteria are capable of assimilation of AD through a metabolic pathway proceeding either via 9α‐hydroxylation followed by  Δ¹‐ dehydrogenation or via Δ¹‐ dehydrogenation followed by 9α‐hydroxylation. In any case, the combined action of the two enzymes leads to formation of the unstable compound 9α‐OH‐ADD and spontaneous splitting the steroid ring which opens the process of complete AD degradation and its utilization as a single or an additional source of carbon and energy 15. As the activity of either 9α‐steroid hydroxylase or Δ¹‐steroid dehydrogenase prevails in different bacterial strains, this is successfully applied in the manufacturing of two key intermediates for steroid drug synthesis, 9α‐OH‐AD and ADD, respectively 16.

Figure 4.

Dynamics of 9α‐OH‐AD accumulation by glucose‐ and n‐alkanes‐grown Rhodococcus sp. cells.

The process of 9α‐hydroxylation carried out by n‐alkanes‐grown cells, on the contrary to the one carried out by glucose‐grown cells, starts almost immediately after adding the AD to the reaction mixture. We find reasonable to explain this lack of lag‐phase with the enhanced hydrophobicity of the n‐alkanes‐grown cells that in turn provides faster and stronger contact with the crystal steroid substrate. Such explanation is in accordance with the observations showing the growing naturally hydrophobic Mycobacterium vaccae 17 and Arthrobacter simplex 18 cells attached to the sitosterol particles. In case of the Mycobacterium fortuitum cells, they have even been co‐grinded with the steroid substrate to reinforce mechanically their contact as to achieve an effective transformation reaction 19.

We are inclined to suggest that in the n‐alkanes‐grown cells the enzyme pathway that allows for complete steroid mineralization may be impaired. Growth on hydrocarbons may lead to possible biochemical and physiological modifications resulting in significant prevalence of the steroid hydroxylation reaction over the reaction of steroid degradation and/or its complete inhibition manifested by the lack of the accompanying product destruction. In this context, possible explanation of our observation may come from the metabolic and physiological responses of Rhodococcus cells to the presence of hydrocarbons 20. Results obtained in our study are also in accordance with the recently published evidences for the positive correlation between the ability of some bacteria for both, assimilation of n‐alkanes and performing steroid transformation reactions 21, 22.

From this point of view, the ability of Rhodococcus sp. for growth on n‐alkanes and for carrying out an effective 9α‐steroid hydroxylation in growth restricting conditions can be considered as two‐parts process. The main peculiarity of this process is that each of its parts is based on a hydrophobic substrate. Thus, the cultivation of the naturally hydrophobic Rhodococcus sp. cells on the mixture of n‐alkanes as a source of carbon and energy in the first part (aiming at harvesting a cell depot with an intentionally increased cell surface hydrophobicity) is of paramount importance. The application of this cell depot in the second part of process, the transformation of crystal AD itself allows for a robust cell‐substrate contact and results in a highly effective reaction of 9α‐steroid hydroxylation without product destruction. In this context, the cultivation of the Rhodococcus sp. cells on a mixture of n‐alkanes can be perceived as an adaptation to the presence of the hydrophobic crystalline substrate to be transformed. That is why we believe it is worth to continue studies on the correlation between the ability of Rhodococcus to assimilate n‐alkanes and to transform crystal steroids in growth restricting conditions. Experiments in this area may reveal important knowledge with possible application in engineering of biotechnological steps in the steroid drug manufacturing avoiding both, application of solvents and manipulations usually applied to microbial cells.

4. Concluding remarks

An effective process of 9α‐steroid hydroxylation of androstenedione based solely on the natural characteristics and peculiarities of the Rhodococcus sp. cells is presented for the first time. Cells were cultivated on media containing n‐alkanes as a source of carbon and energy aiming at harvesting a cell depot with an additionally increased cell surface hydrophobicity. Steroid transformation process was carried out by resting n‐alkanes‐grown Rhodococcus sp. cells and androstenedione was added in a crystal form. As an outcome, the intentionally growth‐driven enhancement of the cell surface hydrophobicity (a consequence of the ability of the naturally hydrophobic Rhodococcus sp. cells to assimilate n‐alkanes) rendered them higher compatibility with the crystal hydrophobic steroid substrate and improved its accessibility in the process of transformation. The ultimate result was an achievement of a significantly increased 9 α‐steroid hydroxylating activity and lack of the accompanying product destruction.

Practical application

An effective process of 9α‐steroid hydroxylation of androstenedione was performed by Rhodococcus sp. cells with an intentionally increased cell surface hydrophobicity achieved by cultivation of the cells on a medium containing n‐alkanes as a source of carbon and energy. The transformation process itself was carried out by resting Rhodococcus cells in buffer medium where the steroid substrate was added in a crystal form.

The novelty of the suggested approach consists in the employment of the natural ability of bacteria from genus Rhodococcus for assimilation of n‐alkanes and adaptation to their presence via ultrastructural, physiological and metabolic modifications. Thus, via cultivation on a medium containing n‐alkanes as a sole source of carbon and energy was harvested a cell depot with intentionally increased cell surface hydrophobicity which in turn rendered these cells higher compatibility with the crystal steroid substrate and facilitated its accessibility.

The accomplishment by the n‐alkanes‐grown Rhodococcus sp. cells of an almost stoichiometrical transformation of androstenedione into its 9α‐hydroxy‐derivative without accompanying product destruction revealed the potential of the suggested approach.

The authors have declared no conflict of interest.