Rhodococcus jostii Porin A (RjpA) Functions in Cholate Uptake

Abstract

RjpA in Rhodococcus jostii is the ortholog of a channel-forming porin, MspA. Deletion of rjpA delayed growth of R. jostii on cholate but not on cholesterol. Eventual growth on cholate involved increased expression of other porins, namely, RjpB, RjpC, and RjpD. Porins appear essential for the uptake of bile acids by mycolic acid bacteria.

TEXT

Steroids are important biomolecules occurring in all domains of life. Steroids are ubiquitous in the environment as a result of excretion and biomass decomposition, as well as industrial and municipal waste discharges. These steroids can be detrimental to both humans (1) and ecosystems (2). Some bacteria degrade steroids, and a few pathways permitting catabolism of steroids have been partially elucidated (3–6). Recent studies have also indicated that steroid uptake and metabolism play important roles in the virulence of both human (7, 8) and animal (9) pathogens. Despite the importance of steroids and their transformation by microorganisms, there remain substantial gaps in our knowledge of microbial steroid uptake and metabolism.

Rhodococcus jostii is a soil-dwelling, metabolically versatile member of the mycolic acid-containing actinobacteria that grow on a range of steroids (10). BLAST analysis of the R. jostii genome revealed four proteins, RjpA (Ro04074), RjpB (Ro03127), RjpC (Ro03156), and RjpD (R08561), with 33, 33, 31, and 36% identities, respectively, to a channel-forming porin from Mycobacterium smegmatis known as MspA (Mycobacterium smegmatis porin A). Furthermore, the presence of signal peptide cleavage sites in RjpA, RjpB, and RjpD indicates that these proteins, like MspA, target the membrane. RjpA and MspA are reciprocal best BLAST hits, indicating that they are orthologous. A structural homology search using RjpA as a query also retrieved MspA, suggesting that RjpA forms a channel similar to that formed by MspA. The MspA monomers oligomerize to form a large (>100-kDa) homo-octameric, goblet-shaped protein with a central pore spanning the outer membrane (11). Loop regions of the protein lining the pore eyelet undergo conformational change, which may affect the uptake of ions and solutes by the porin (12). In M. smegmatis, MspA is involved primarily in the uptake of hydrophilic solutes, such as glucose (13) and phosphates (14). MspA has been shown to be the main conduit for hydrophilic antibiotics, such as fluoroquinolones and chloramphenicol (15), in addition to nutrients. Deletion of MspA also caused a marked increase in resistance to hydrophobic antibiotics, such as rifampin and erythromycin (16). Furthermore, deletion of MspA resulted in a 3-fold reduction in the uptake of the bile acid steroid chenodeoxycholate (16).

The similarity of RjpA to the MspA porin and the effect of MspA deletion on the uptake of hydrophobic antibiotics and a steroid led to the hypothesis that RjpA is involved in the uptake of steroids by R. jostii, which this study investigated.

The rjpA gene was deleted from R. jostii, and the mutant was tested for its ability to grow on cholesterol or cholate. The rjpA gene of R. jostii was completely deleted, in frame and unmarked, using the sacB counterselection system described previously (17) and primers in Table 1. Deletion of rjpA did not affect the growth of R. jostii on cholesterol (Fig. 1). This result suggests that RjpA is not involved in the uptake of highly hydrophobic steroids, such as cholesterol. Both R. jostii and Mycobacterium tuberculosis take up cholesterol and, in the case of R. jostii, also β-sitosterol, via the Mce4 system, an unusually complex ATP-binding cassette transporter (7, 10). Functionally essential components of the Mce4 system include putatively extracytoplasmic Mce proteins, which we speculate may facilitate movement of hydrophobic steroids across the outer membrane and periplasm. The Mce4 system was not required for the uptake of, and growth on, cholate. In contrast to results with cholesterol, growth of the rjpA mutant on the more hydrophilic bile acid cholate was impaired. Initiation of growth of the mutant on cholate was 2 days later than that of the wild type (Fig. 2). Subsequently, the mutant grew at the same rate and to the same final cell density as the wild type. The 2-day difference in growth initiation was highly reproducible in independent experiments.

Table 1.

Primers and probes used in this study

No.	Primer	Sense or probe	5′–3′ sequencea	Function
1	ro04074	Forward	TGG TCA GGA TGC AGG GAG TT	Primers and probes for quantification of porin genes by RT-QPCR
2	ro04074	Probe rpJA	TGG CTC CAT CTC C
3	ro04074	Reverse	GTG ACC GGC CCG ATA GC
4	ro03127	Forward	GAG GCA ACC GGG TCG AA
5	ro03127	Probe rpJB	TCC TGC AAG GTG ATA CC
6	ro03127	Reverse	CGG CGG GAC CAC TTG ATA C
7	ro03156	Forward	CAG TGG CTC AAC GAC GTG AT
8	ro03156	Probe rpJC	CTC AAC GGC ACC CC
9	ro03156	Reverse	TGG ACT TGA TCG CGG TGT T
10	ro08561	Forward	CCG CCG CTG GAT GGT
11	ro08561	Probe rpJD	TTC CCA CCA GTG TCG AG
12	ro08561	Reverse	GGC GTA ACC GTT GTG GAA GA
13	ro04074_P1	Forward	CCCAAGCTTAGGCGATCAGCAAGCCGAGGACG	Upstream flanking region amplification primers
14	ro04074_P2	Reverse	TGCTCTAGACTGCGACTGCCACCGAGCGTG
15	ro04074_P3	Forward	TGCTCTAGAGGCCGTGACGGAGAACGGTG	Downstream flanking region amplification primers
16	ro04074_P4	Reverse	CGCGGATCCCGGTGTGCTGGGCGCG
17	ro04074_P7	Forward	CGGATTCGAGGAGCGGCAGGTG	rjpA deletion confirmation primers
18	ro04074_P8	Reverse	GGCGGTGAGGTGGTG GTGCATG
19	ro04074_P9	Forward	GGAATTCCATATGGGTCGGCATGACGGACATCAGC	Primers to amplify rjpA for complementation
20	ro04074_P10	Reverse	CCCAAGCTTCTAGTTGAGCTTCTGCGGCACACCG

^aRestriction sites are underlined.

Fig 1.

Growth of R. jostii (wild type [WT]) and the rjpA (Δro04074) mutant on cholesterol (n = 3; bars indicate standard errors). Cultures were grown on defined medium as previously described (22), with 2.0 mM cholesterol as the sole organic substrate. Because precipitated cholesterol interfered with optical density measurement, growth was measured as protein using the bicinchoninic acid (BCA) assay (Pierce) after hot alkaline lysis.

Fig 2.

Growth of R. jostii (WT), the rjpA mutant (Δro04074), and the rjpA complementation strain (Δro04074C) on cholate (n = 3; bars indicate standard errors). Growth conditions were as described for Fig. 1, except that 2.0 mM cholate was the sole organic substrate.

To verify that deletion of rjpA caused delayed growth on cholate, the mutation was complemented. The rjpA gene was cloned into the pTip-QC2 vector and expressed in the rjpA mutant strain, using previously described methods (10) and primers in Table 1. This complementation completely restored the wild-type growth phenotype on cholate (Fig. 2).

The ability of the rjpA mutant, following the 2-day delay, to grow on cholate at the same rate and to the same final density as the wild type suggested that R. jostii may compensate for the rjpA deletion by upregulating other porins. Three paralogs of rjpA, ro03127 (rjpB), ro03156 (rjpC), and ro08561 (rjpD), were identified by BLAST search. To determine whether the other rjp porins might compensate for the rjpA deletion, levels of expression of all four porin genes during exponential growth (optical density at 600 nm [OD₆₀₀] = 0.8) were compared using reverse transcriptase quantitative-PCR (RT-QPCR). The wild type expressed rjpA at much higher levels than the other porin genes on both pyruvate and cholate (Table 2), indicating that under our growth conditions, RjpA is the major porin in R. jostii. Compared to the wild type, the mutant greatly increased expression of rjpB, rjpC, and rjpD during growth on cholate, by 45-, 65-, and 26-fold, respectively. In contrast, on pyruvate, the rjpA mutant grew normally and did not increase expression of rjpB, rjpC, and rjpD, relative to that of the wild type. Thus, despite its high level of expression, RjpA does not play a critical role in growth on pyruvate. Expression of sigA is expected to be correlated with the cellular growth rate, and it did not vary by more than 2-fold among the assays. This indicates that the large differences in expression of rjp genes observed are not attributable to major differences in growth rates among the two strains on the two substrates.

Table 2.

Abundance of porin transcripts during exponential growth of R. jostii or the rjpA mutant strain on either pyruvate or cholate

Gene	No. of gene transcripts (no. of copies/ng of DNA) after growth ona:
	Pyruvate (20 mM)		Cholate (2 mM)
	R. jostii	Mutant	R. jostii	Mutant
rjpA	445,800 ± 195, 800	250 ± 160	28,300 ± 1,860	4 ± 4
rjpB	800 ± 320	180 ± 9	35 ± 4	1,500 ± 440
rjpC	50 ± 5	30 ± 1	22 ± 1	1,400 ± 140
rjpD	1,330 ± 240	350 ± 32	144 ± 39	3,800 ± 290
sigA	15,800 ± 330	30,800 ± 7,200	21,000 ± 720	27,300 ± 2,500

^aValues are means ± standard deviation (n = 3). Transcripts were measured by reverse transcriptase quantitative-PCR, as described previously (23), using the primers and probes in Table 1.

Overall, the results indicate that porins are essential for efficient cholate uptake. Of the porins, RjpA plays the main role in cholate uptake, but at least one of the other three porins can also serve that function and compensate for the loss of RjpA. A similar phenomenon was observed in M. smegmatis growing on glycerol (18). The mspA gene codes for the most highly expressed porin in M. smegmatis. Only mspA and, to a much lesser extent, mspC were expressed in wild-type M. smegmatis (18). In response to the deletion of mspA, expression of mspB and mspD was increased. Moreover, the deletion of mspA resulted in a significant decrease of nutrient permeability across the outer membrane, while the deletion of other porin genes caused no significant reduction in nutrient uptake (18).

We are beginning to understand steroid uptake by bacteria. This and other studies suggest that different uptake mechanisms are employed for the most hydrophobic steroids versus the more hydrophilic bile acids. So far, three systems that transport bile acids across the cytoplasmic membrane have been characterized. One is the well-characterized BaiG protein, a 50-kDa integral membrane protein from Eubacterium sp. strain VP1 12708 (19). BaiG cloned into Escherichia coli has been shown to transport unconjugated cholate and chenodeoxycholate (19). The second bile acid transporter was identified in Lactobacillus johnsonii and cloned into E. coli to demonstrate its ability to transport cholate (20). Recently, a transporter from Neisseria meningitidis homologous to the human apical sodium-dependent bile acid transporter (ASBT) was structurally characterized (21). The current study advances our understanding of the role of porins in transport of a bile acid across an outer membrane. Stephan et al. (16) proposed that the presence of porins might affect rates of diffusion of chenodeoxycholate through the lipids of the outer membrane of M. smegmatis. However, based on the essentiality of porins for growth of RHA1 on cholate, a more parsimonious conclusion may be that bile acids diffuse through porin channels. Clearly, further biochemical and structural studies are needed to elucidate the mechanism by which RjpA and other porins facilitate the uptake of bile acids. It is likely that bile acid uptake by mycolic acid bacteria, in general, requires porins. It is further possible that bile acid uptake by Gram-negative bacteria also requires porins, as their outer membranes are structurally analogous to those of mycolic acid bacteria.