Actinobaculum schaalii, a Common Uropathogen in Elderly Patients, Denmark

This organism is identified more often by PCR than by cultivation.

A ctinobaculum schaalii was fi rst described in 1997 and named after Klaus P. Schaalii, a German microbiologist specializing in actinomycete microbiology. The genus Actinobaculum includes A. schaalii, A. suis, A. massiliae, and A. urinale and is closely related to the genera Actinomyces and Arcanobacterium (1).
These bacteria are small, gram-positive, facultative anaerobic, CO 2 -requiring coccoid rods. They grow as dimorphic gray colonies <1 mm in diameter, are nonmotile and non-spore forming, and show weak β-hemolysis on agar plates containing 5% horse or sheep blood after 3-5 days of growth. They are catalase, oxidase, and urease negative and resistant to trimethoprim and ciprofl oxacin (2). Their habitat is probably the human genital or urinary tract (1).
Because of its slow growth and resemblance to the normal bacterial fl ora on skin and mucosa, A. schaalii is often overlooked or considered a contaminant. Furthermore, it is often overgrown by faster-growing commensal and pathogen bacteria. Most laboratories incubate urine samples only overnight in ambient air, which further impedes isolation of A. schaalii (2).
Diffi culties identifying A. schaalii by using traditional phenotypic tests have obscured its pathologic role for many years. However, A. schaalii can cause urinary tract infections (UTIs), some of which lead to serious illnesses such as urosepsis, osteomyelitis, and septicemia, mainly among the elderly and patients predisposed to UTIs (1)(2)(3)(4)(5)(6). We developed a TaqMan real-time quantitative PCR (qPCR) specifi c for the gyrase B (gyrB) gene for fast and sensitive detection of A. schaalii from urine and blood samples.

Patient and Control Groups
From October 2008 through January 2009, a total of 252 routine urine samples were randomly selected from patients of all ages from 3 hospitals and 150 medical practitioners in Viborg County, Denmark (population ≈230,000 persons). Seventy percent of patients were from hospitals. Urine collection was midstream, from bedpans, from catheters, or unspecifi ed in 41%, 19%, 18%, and 21% of cases, respectively. A total of 38 control urine samples were obtained from patients before they underwent elective surgery of hips or knees. These patients were 63-81 years of age and had negative results for leukocyte esterase and nitrate by a urine dipstick test (Roche Diagnostics Ltd., Burgess Hill, UK).
Samples tested by using PCR were simultaneously analyzed by using standard laboratory tests. These tests were wet smear microscopy and incubation on 5% Columbia sheep blood agar (Becton Dickinson, Heidelberg, Germany) in an atmosphere of 5% CO 2 at 35°C for 1 or 2 days.

Extraction of DNA
Bacteria were incubated anaerobically on 5% Columbia sheep blood agar in an atmosphere of CO 2 at 35°C for 2 days before harvesting. DNA was purifi ed by taking a swab of bacteria from the agar plate and transferring it to 1 mL of saline. The DNA from bacteria was extracted from 800 μL of saline by using the Kingfi sher mL magnetic particle processor (Thermo Electron Corporation, Waltham, MA, USA) according to the manufacturer's instructions, eluted in 100 μL elution buffer, and stored at 4°C until use. DNA was also obtained from 800-μL urine samples as described above.

Sequencing
Fourteen A. schaalii strains, including reference strain CCUG 27420, were used for sequencing. Universal primer pair UP-1 and UP-2r was used to amplify the gyrB gene from A. schaalii (Table 1). PCR was performed as described by Yamamoto and Harayama. (7). The PCR product was then gel purifi ed by using the QIAquick Gel Extraction Kit (QIAGEN, Hilgen Germany) and sequenced in an ABI 3130 XL genetic analyzer (Applied Biosystems, Foster City, CA, USA) according to the manufacturers' instructions. Sequencing primers UP-1S and UP-2Sr (Table  1) were used to sequence the purifi ed PCR product in both directions. Primers were synthesized by DNA Technology (Aarhus, Denmark).
The primer pair A.s-forward 5′-GGCCATGCAG TGGACCTC-3′ and A.s-reverse 5′-GCACATCATCA CCGGAAAGA-3′ amplifi ed a 185-bp fragment. The probe 5′-TCCGAATCGGTCAATACCTTCGC-3′ was labeled at the 5′ end with 6-carboxyfl uorescein and at the 3′ end with Black Hole Quencher 1. Primers and probe were synthesized by Sigma-Aldrich (St. Louis, MO, USA). primers, and 5 μL of template DNA. An internal control containing 1.25 μL of internal PCR control primer/probe mixture and 0.25 μL of internal PCR control DNA (Applied Biosystems) was also used. Samples were incubated for 1 cycle at 95°C for 2 min and 50 cycles at 95°C for 30 s and 60°C for 60 s. All samples were run in duplicate. DNA from A. schaalii CCUG 27420 was used as a positive control and was included in each PCR. Sterile water was used as a negative control. Results were analyzed by using the Mx3000P software package (Stratagene).

Detection Limit and Quantifi cation
The detection limit of the A. schaalii gyrB assay was determined by using a 10-fold serial dilution of known concentrations (1.5 × 10 1 to 1.5 × 10 8 CFU/mL) of A. schaalii CCUG 27420. Quantifi cation of A. schaalii in urine samples was performed by using the same dilution series.

Analytical Specifi city
To determine the analytical specifi city of the assay, we tested 36 clinical strains of A. schaalii and strain CCUG 27420. Phylogenetically related (1) and clinically relevant bacterial strains, including several Actinomyces spp., Ar-  (Table 2).

Verifi cation of TaqMan qPCR Assay Results
To verify results of this assay, 6 PCR products were sequenced. The fi rst 15 PCR-positive urine samples were cultivated, and isolates were identifi ed as described by Reinhard et. al. (2). Identity of isolated A schaalii strains was confi rmed by using a qPCR.

Purifi cation of DNA from Blood Cultures
Ten milliliters of blood and 1 mL of culture containing 2 × 10 7 , 2 × 10 5 , 2 × 10 3 , and 2 × 10 1 CFU/mL of A. schaalii reference strain CCUG 27420 were added to aerobic and anaerobic BACTEC culture vials (Becton Dickinson). DNA from bacteria-positive blood cultures was extracted from 800 μL of aerobic or anaerobic media and purifi ed by using the Kingfi sher processor as described above.
Because BACTEC culture vials contain sodium polyanetholesulfonate (SPS), a known PCR inhibitor, either DNA must be purifi ed from BACTEC culture vials by using specifi c purifi cation methods or purifi ed DNA must be diluted to prevent the SPS from inhibiting the PCR (8). Ten-fold serial dilutions of purifi ed DNA from positive BACTEC culture vials were made and tested by using the qPCR as described above. DNA was extracted from an anaerobic BACTEC culture vial from a patient sample from which A. schaalii had been isolated by cultivation.

Statistical Analysis
The χ 2 test was used to analyze differences in detection of A. schaalii. Statistical analyses were performed by using SPSS for Windows version 16.0 (SPSS Inc., Chicago, IL, USA).

Cultivation of PCR-Positive Samples
Isolates were obtained from 7 of the 15 urine samples cultured. The 7 isolates were confi rmed positive by our real-time PCR.

Detection Limit and Analytical Specifi city
Assay results were linear at bacterial concentrations from 1.5 × 10 4 to 1.5 × 10 8 CFU/mL with an R 2 value of 1.000 (Y = -3.296 × log(X) + 25.96). The detection limit of the assay was between 1.5 × 10 3 and 1.5 × 10 4 CFU/mL, which corresponds to 7.5-75 CFU/reaction. The assay amplifi ed DNA from all 37 isolates of A. schaalii tested. No PCR amplifi cation signal was detected when other species were tested (Table 2).

DNA Sequencing Analysis
The 6 PCR products amplifi ed from bacteria-positive urine samples had the expected size. Sequence alignment of the 6 PCR products showed homology to the sequenced gyrB gene from A. schaalii strains.

Identifi cation of A. schaalii from Blood Cultures
The 2 anaerobic BACTEC culture vials to which 1 mL of 2 × 10 7 CFU/mL and 2 × 10 5 CFU/mL had been added and 1 aerobic BACTEC culture vials to which 1 mL of 2 × 10 7 CFU/mL had been added showed positive results in the BACTEC 9240 blood culture system. There was no growth recorded with lower inoculum concentrations. PCR with undiluted and 10-fold diluted DNA was inhibited, probably by SPS. However, the 100-fold dilution of purifi ed DNA from the 2 anaerobic and 1 aerobic BACTEC culture vials was PCR positive. The 100-fold dilution of purifi ed DNA from a positive anaerobic BACTEC culture vial (patient specimen) was also PCR positive.

Analysis of Urine Samples
Of 252 urine samples, 41 (16%) were PCR positive with bacterial concentrations >10 4 CFU/mL. Of 155 urine samples from patients >60 years of age, 34 (22%) were PCR positive (Table 3), of which 31 (91%) harbored other common uropathogenic bacteria in addition to A. schaalii (Table 4). Species distribution of these common uropathogenic bacteria was comparable to that found in our microbiology department throughout the year. Treatment with antimicrobial drugs before specimens were obtained was reported by 19% of the patients.
The 41 PCR-positive urine samples were collected midstream from 37% of patients, from bedpans for 27%, from catheters for 12%, and by an unspecifi ed method for 24%. Among 177 hospitalized patients, 18% of samples from 104 patients >60 years of age and 10% of samples from 73 patients <60 years of age were PCR positive (p = 0.133). Among 75 urine samples obtained by practitioners, 30% of samples from 51 patients >60 years of age and none of the samples from 24 patients <60 years of age were PCR positive (p = 0.002). There was no signifi cant difference in the presence of A. schaalii by sex of the patients (p = 0.485). When the control group (patients who had had hip or knee surgery) was compared with patients >60 years of age, no signifi cant difference in the presence of A. schaalii was found (p = 0.227). In addition, we did not fi nd any detectable differences between PCR-positive and PCR-negative results for hospitalized patients concerning underlying urinary tract pathologic changes and concurrent conditions such as hypertension and diabetes.

Discussion
The real-time PCR assay confi rmed that infection with A. schaalii increases with age (2). More than 1 of 5 urine samples from patients >60 years of age were PCR positive, and A. schaalii was most common in patients who visited medical practitioners and who had an infection with ordinary urinary pathogens. In comparison, culture fi ndings in a study in our laboratory showed that 0.4% of cultured urine samples from patients >60 years of age had A. schaalii and that these patients had a broad spectrum of UTIs (2).
The present study shows that bacteria species, especially anaerobic or slow-growing species, are more common than what culture results indicate. Most likely, other pathogen bacteria exist that are even more diffi cult to identify by cultivation than is A. schaalii. Molecular biologic techniques such as real-time PCR can be valuable tools for identifi cation of these organisms. Pathogenic bacteria that are diffi cult to cultivate or identify by cultivation should not be underestimated.
Other common uropathogens were identifi ed by cultivation in 9 of 10 PCR-positive urine samples (Table 4). This fi nding indicates that A. schaalii is probably a common, undetected bacterial copathogen in many UTIs. Because most PCR-positive samples were from persons with multiple infections, determining which microorganism caused the UTI is diffi cult. However, results from our study support fi ndings in case reports (2,3,6) in which A. schaalii was often found in monoculture for patients who had UTIs and therefore considered the causative agent. Furthermore, PCR showed that A. schaalii is a more common pathogen than previously thought. However, it will be diffi cult to fulfi ll the last of Koch's criteria and prove with animal experiments that A. schaalii is a uropathogen.
Clinical microbiologists, clinicians, and medical practitioners should be aware of A. schaalii in patients predis- posed for UTIs or unexplained chronic UTIs, especially if initial fi ndings of wet smear microscopy for bacterial rods and leukocytes differ from negative growth under commonly used aerobic cultivation methods. For patients with suspected infections, urine should be sent to a department of clinical microbiology and incubated in an atmosphere of 5% CO 2 for 2 to 3 days. For patients with clinically verifi ed UTIs who do not respond to treatment with ciprofl oxacin or trimethoprim, infection with A. schaalii should be suspected. If A. schaalii is the cause of the infection, treatment with β-lactams, such as ampicillin or cephalosporins, should be given. The optimal duration of antimicrobial drug treatment with β-lactams is not clearly defi ned but several weeks of treatment may be required in severe cases.
Because A. schaalii can be diffi cult to identify even when cultured in an atmosphere of 5% CO 2 , the real-time PCR described in this report can be used for identifi cation in urine and blood cultures. Alternatively, if the bacteria can be isolated by cultivation, the API Coryne and Rapid ID32A test systems (bioMérieux, Marcy l'Etoile, France) can be used for identifi cation, as described by Reinhard et. al. (2). In conclusion, A. schaalii is an underestimated opportunistic copathogen that probably causes UTIs and urosepsis, particularly in elderly patients or patients predisposed for UTIs.