Molecular screening of unculturable bacteria present in men with reproductive failure

Abstract

Objective:  To determine the presence of unculturable bacteria using polymerase chain reaction (PCR) in infertile men with pyosperrmia.

Design:  Perspective clinical study.

Setting:  The study took place at the Department of Reproductive Medicine, Owaisi Hospital and Research Center; In vitro Fertilization Unit, Mahavir Hospital and Research Center; Center for Cellular and Molecular Biology; and Bharat Biotech Foundation, Hyderabad, India.

Patients:  A total of 68 infertile men and 15 donors, all with no symptoms of genito‐urinary tract infections and sterile semen cultures were included in the study.

Interventions:  None.

Main outcome measures:  Screening bacteria using routine bacterial cultures and PCR based screening with universal eubacterial primers.

Results:  The statistical analysis of all the semen parameters in asthenazoospermic, azoospermic, ceyptazoospermic, severe oligospermic and mild oligospermic patients were found to be significant compared with the controls. All the groups were found to be significant compared with the controls (P < 0.05) except for volume and pus cells in the cryptozoospermia group. The Student's t‐test also was significant for the seminal parameters before and after treatment of 68 selected individuals with pyospermia and sterile cultures. A total of 44.11% (30/68) samples were collected from the negative culture of pyospermic infertile men have shown the presence of bacteria on amplification using PCR with universal eubacterial primers. The DNA was purified and sequenced. The sequences were checked for homology using DNASTAR and Ribosomal DataBase Project II. A total of 90% of the samples have shown the nearest evolutionary relation to Pantoea P102 (AF394539) and 10% of samples have shown close relation with Burkholderia cepacia (AF042161).

Conclusion:  The routine bacteriological cultures were unable to detect certain bacterial species particularly with members of enterobacteriaceae family (Pantoea species). Polymerase chain reaction, when used for screening bacteria, can detect the unculturable form of bacteria in infertile men. No amplification for bacterial DNA was obtained in control samples (fertile men with sterile semen cultures.) (Reprod Med Biol 2004; 3: 77– 84)

INTRODUCTION

INFERTILITY AFFECTS APPROXIMATELY 15% couples worldwide. Approximately 50% of infertility is attributed to the male partner; however, only 2% are infertile because of severe defects in sperm production ¹ and the majority of infertile men are otherwise healthy with no known cause for these defects being identified with certainty. ² In such cases, the defects have often been ascribed to infection. Pyospermia or presence of leukocytes in semen is a very common observation among infertile men than fertile men. In most of the infertile men no known pathogen could be recovered from the semen or urine using traditional bacteriological techniques. The role of infection in male reproductive failure is an area of active debate and several studies have suggested that genitourinary tract infections may be associated with male reproductive failure. ³ In addition, several studies have suggested the existence of viable but non‐culturable bacterial species or unculturable bacteria in the environmental studies, in such cases the detection of bacterial ribosomal RNA (rRNA) genes as an indicator of the presence of bacteria is an established technique that has been used for the detection of both environmental and medically important bacteria. ⁴ , ⁵ Polymerase chain reaction (PCR) amplification of bacterial 16S rRNA genes has been used successfully to detect bacteria that cause a variety of infections including postoperative endophthalmitis, ⁶ , ⁷ septic arthritis, ⁸ and meningitis ⁹ etc. Therefore, in the present study we tried to identify the unculturable form of bacteria in infertile men with pyospermia.

A total of 50–83% of pyospermic men responded favorably to antibiotic therapy showing improved sperm parameters and decreased lymphocytes in semen even in the absence of culturable microbes, which is indicative of the presence of unculturable microbial infection. ¹⁰ The present study was carried out to identify and characterize these microbes using molecular biological techniques in order to establish an association between the role of infection and male reproductive failure. As molecular biological techniques characterize genotype rather than phenotype of microorganisms, these techniques can identify the unculturable microbes more effectively. ¹¹

As preliminary investigation, semen analysis is the most widely used test to estimate fertility potential of the male, which gives details of various parameters such as volume of the sample, concentration of spermatozoa and motility of spermatozoa and number of pus cells present in the sample. Other parameters such as vitality and osmotic fragility of the spermatozoa were also analyzed in the present study, for all the couples involved male factor infertility in total or in part. The data was statistically analyzed and compared with the control group.

MATERIALS AND METHODS

A TOTAL OF 864 couples were evaluated for infertility at the Assisted Conception Services Unit, Mahavir Hospital and Research Center (MHRC) and Owaisi Hospital and Research Center (OHRC) during the period from August 1998 to April 2002. The Assisted Conception Unit, MHRC and OHRC are referral centers that receive cases from all over Andhra Pradesh. A complete case history exploring all the facets of reproductive function, such as the history of married life, consanguinity nature and duration of infertility, history of contraceptive use, sexual history, details of occupation, history of medical disease, surgery, infection, medication, lifestyle and habits, diet, sociodemographic characteristics and family history of infertility/delayed conception, was recorded using a standard questionnaire.

The preliminary investigations of male partner were performed using hemocytometry. This gives the count, motility of the sperm and number of pus cells present in the sample. Sperm morphology was studied by Papanicolaou staining and the sperm head shape and size were studied.

Assessment of sperm vitality was performed using Eosin‐nigrosin test. Intact cells exclude eosin and dead cells take up the red color of eosin. Nigrosin was used as a counterstain to facilitate visualization of the unstained live cells. Sperm osmotic fragility was analyzed using the hypo osmotic swelling test.

All the above‐mentioned semen parameters were statistically analyzed using spss/PC + (10.0 version). A total of 26 fertile donors were also analyzed for the above parameters and were considered as controls.

A total of 68 patients with pyospermia (>1 million pus cells/mL) and sterile cultures were given empiric antibiotic therapy and were checked for the difference in their seminal parameters before and after antibiotic treatment. The semen parameters of the 68 individuals who had pyospermia with sterile cultures were statistically analyzed using the Student's t‐test for significance.

The molecular analysis of unculturable bacteria was carried out on the 68 patients (patients containing >1 million pus cells/mL of sample) selected for the present study. All patients with present or past history of urinary tract infection and those with positive bacterial cultures were excluded from the study. The semen specimens were collected under hygienic conditions; patients were instructed to follow the following precaution while collecting the sample:

• 1To wash their hands thoroughly with antiseptic soap, then their penis should be cleaned using an antiseptic solution (Septisole).
• 2The patients were also instructed to collect a semen sample by masturbation directly into the sterile container provided and to remove its lid only at the moment of collection and replace it immediately after completion of the collection to minimize contamination.

The samples for molecular analysis were stored at −20°C immediately after collection. Control semen specimens were also collected under similar conditions from 15 donors whose cultures were sterile.

Microbiological culture

All the specimens under the study were sent to a commercial laboratory for aerobic and anaerobic cultures along with chlamydeous, mycoplasma and neisseria cultures were performed as per standard protocols.

Isolation of bacterial DNA present in seminal fluid

The supernatant of semen (seminal plasma) was centrifuged at 12 000 r.p.m. for 5 mins. The pellet was suspended in Tris buffer and incubated at 80°C for 15 mins. Lysozyme was added to the solution and incubated for 2 h at 56°C. Proteinase K (150 mg mL) and sodium dodecylsulfate 2% were added, mixed thoroughly and this mixture was incubated for 4 h at 37°C.

The lysate was extracted with phenol, phenol chloroform (1:1) and chloroform isoamyl alcohol (24:1, v/v). The microbial DNA was precipitated by addition of 1/20th of solution acetate (pH, 5.2) and chilled ethanol. The microbial DNA was pelleted by centrifugation at 10 k for 15 min. The DNA was washed thrice with 70% ethanol and once with absolute alcohol, vacuum dried and dissolved in 300 mL Tris EDTA.

Measurment of DNA concentration

Two methods were used to measure the amount of nucleic acid in a preparation. A spectrophotometric measurement of the amount of ultraviolet (UV) irradiation absorbed by the bases is simple and accurate, but if the amount of DNA is very small or if the sample contains significant quantities of impurities, the amount of nucleic acid was estimated from the intensity of fluorescence emitted by ethidium bromide.

The UV induced fluorescence emitted by ethidium bromide molecule intercalated into the DNA. Therefore, the amount of fluorescence is proportional to the mass of DNA (1–5 ng of DNA can be detected by this method).

One microliter is used in a microliter reaction volume mixture for PCR amplification. Each reaction contained 20 pmole of each primer and standard amounts of Gene Amp reagent (GMI, Minnesota, MN, USA).

Polymerase chain reaction amplification of bacterial DNA

The purified DNA was used as template for PCR amplification of 16S rDNA using Perkin‐Elmer DNA thermocycler.

16S rDNA amplification

Approximately 800 base pairs (bp) of 16S rDNA amplified using PCR as described by Relman et al. ¹¹ The sequences of the primers for amplification are:

• • 5¹ primer: – GAGTTTGATCCTCCTGGCTCA;
• • 3¹ primer: – ACGGCTACCTTGTTACGACTT.

They are the conserved regions of 16S rRNA in all the Eubacteria from 0–864 bp position.

The conditions for PCR amplification: each PCR cycle for bacterial DNA amplification using the broad range eubacterial primers p8FPL and p806R consisted of 1 min denaturation at 94°C, annealing at 55°C for 1 min and extension at 72°C for 2 min in a programmable thermal controller. Positive and negative controls were set along with each set of samples. Small aliquot PCR product was checked through agarose gel electrophoresis and the product was then purified to remove unincorporated primers and deoxyribonucleoside triphosphate.

Determination of DNA sequence of polymerase chain reaction amplified products

The purified amplicons were then directly sequenced using the Bigdye terminator chemistry with an automated DNA sequencer Model ABI Prism 377 (Applied Biosystems, Foster City, CA, USA).

Sequencing reaction follows

Template DNA 5.0 µL

Big dye terminator ready mix 4.0 µL

Primer (5 pmoles) 1.0 µL

Total mix volume 10.0 µL

Cycle sequencing conditions

There were 30 cycles and each cycle consists of 96°C for 10 s of denaturation, 50°C for 5 s of annealing and 60°C for 4 s of extension.

Purification of extended product

After PCR 25 µL of alcohol and 1 µL of sodium acetate were added to each sample and kept for 10 min at room temperature. The samples were centrifuged for 20 min at 4500 r.p.m. The supernatant was removed by inverting the tubes. A total of 100 µL of 70% alcohol was added to each sample and centrifuged at 4500 r.p.m. for 10 min. Filter paper was placed on the plate and briefly centrifuged in an inverted position with filter paper beneath. A total of 10 µL of 50% Hi‐Di formamide was added to each sample and the samples were sequenced.

Sequence alignment and phylogenetic analysis

The sequences obtained using the forward and reverse primers were aligned using DNASTAR software (DNASTAR, Madison, WT, USA). The other 16S rDNA sequences used to determine the phylogenetic relationship were obtained from the individual sequences deposited in the ribosomal database, GeneBank and European Molecular Biology Laboratory (EMBL), etc. The phylogenetic analysis was done at Ribosomal DataBase Project II (RDP II) and also using DNASTAR software.

The alignment of the sequences, similarity determination and distance estimation was done using Clustal V. ¹³ Evolutionary tree was constructed with paup ¹⁴ and Phylip ¹⁵ software (Department of Genome Sciences at the University of Washington).

RESULTS

Descriptive statistics report for the five groups of men with reproductive failure

The statistical analysis of various seminal parameters analyzed for 312 men with reproductive failure were categorized into five different groups based on their sperm count (concentration of spermatozoa million/mL). They were compared with the semen parameters of 26 fertile donors by using spss/PC (10.0 version). Volume, sperm count, motility, morphology, vitality, osmoregularity, pus cells (minimum, maximum, average from the range recorded) were analyzed for means and standard deviations and the means were compared between the groups. Here the variances were compared by Levene's test for hetrogenity of variances. For all the variables the variances were found to be significant or heterogeneous. Hence, the data were analyzed as non‐parametric tests of significance (Mann–Whitney U‐test) between the control and other groups. All the variables in all groups were found to be significant compared with controls (P < 0.05) except for the volume and pus cells in the group cryptozoospermia (Table 1).

Table 1.

Descriptive statistics report for the five groups of men with reproductive failure

Etiological classics	N	Volume mL	Sperm count (million/mL)	Motility (%)	Normal morphology (%)	Vitality (%)	Osmo regularity (%)	Pus cell (min)	Pus cell (max)	Pus cell average
Asthenazoospemia	96	2.39 ± 0.554	47.88 ± 21.433	51.32 ± 10.78	51.32 ± 10.78	61.01 ± 10.69	59.28 ± 10.09	13.09 ± 6.29	16.88 ± 9.08	14.98 ± 7.291
Azoospermia	22	3.90 ± 1.734	0	0	0	0	0	4.68 ± 92	6.14 ± 6.61	5.41 ± 6.261
Cryptozoospermia	19	2.64 ± 0.612	0.43 ± 0.324	57.53 ± 11.51	62.32 ± 11.47	61.89 ± 9.05	60.26 ± 9.92	1.68 ± 1.20	3.26 ± 1.85	2.471 ± 1.495
Severe oligispermia	79	2.22 ± 0.472	4.47 ± 2.575	55.68 ± 10.50	59.27 ± 11.28	64.82 ± 12.26	63.92 ± 10.13	3.67 ± 3.78	5.11 ± 4.42	4.39 ± 4.081
Mild oligoospermia	96	2.58 ± 0.509	14.63 ± 3.027	60.89 ± 10.38	66.48 ± 8.11	68.04 ± 8.91	71.71 ± 6.23	2.07 ± 1.84	3.53 ± 2.06	2.80 ± 1.93
Cryptozospermia	26	2.93 ± 0.524	99.38 ± 21.896	71.5 ± 6.35	81.54 ± 5.37	86.42 ± 7.12	82.04 ± 5.94	1.23 ± 0.99	2.54 ± 1.33	1.88 ± 1.152
Normospermia

Sixty‐eight selected individuals who had pyospermia and whose cultures were sterile were given empiric antibiotic therapy. The seminal parameters of these individuals before and after antibiotic therapy were statistically analyzed using Student's t‐test (2, 3).

Table 2.

Representing the Student's t‐test values for all the five semen parameter analyzed for the 68 individuals selected for antibiotic therapy

	Paired differences
	Mean	Standard deviation	Standard error mean	95% confidence interval of the difference		Student's t‐test
Pair count 1
Count 1 pair	31.49	10.43	1.26	28.96	34.01	24.896
Motility % 2	36.85	7.34	0.89	35.08	38.53	41.421
Motility pair
Morphology % 3	13.90	7.40	0.90	12.11	15.69	15.488
Morphology pair
Vitality % 4 vitality 1 pair	4.74	5.83	0.71	3.32	6.15	6.701
Number of pus cells 5
Pus cells 1	−0.6662	2.0162	0.2445	1.1542	0.1782	−2.725

Table 3.

The significance value of seminal parameters analyzed

	Degrees of freedom	Significance (two‐tailed)
Pair count 1
Count	67	0.000
Pair motility 2
Motility	67	0.000
Pair morphology 3
Morphology	67	0.000
Pair vitality 4
Vitality	67	0.000
Pair no. pus cells 5
Pus cells	67	0.008

The results were highly significant (P < 0.001) (2, 3) with improvement in all the five seminal parameters analyzed, which suggested that the patients responded to antibiotic therapy with the improvement in their seminal parameters and decrease in their seminal leukocytes, in spite of their cultures being sterile. This suggested the presence of unculturable bacteria.

Out of the 68 infertile patients selected for the present study, 30 semen samples from the subjects have shown amplification for the bacterial DNA with the bacterial broad‐range 16S rRNA primers p8FPL and p806R and none of the control donor samples has shown any PCR amplification.

The amplicons obtained in the patients’ samples were purified and sequenced. The sequences thus obtained were compared with the published 16S rDNA data using BLASTN search and at RDP. A total of 90% (27/30) of the sequences have shown maximum homology to the following bacterial species: Pantoea 77000 (AF227896), Pantoea endophytica (AF130892), Pantoea P102 (AF394539) and Enterobacter agglomerans (AF130960) (4, 5).

Table 4.

Number of cases investigated by polymerase chain reaction

Number of cases investigated	Number of cases positive to polymerase chain reaction amplifications	Number of cases negative polymerase chain reaction amplification	Percentage of samples amplified by polymerase chain reaction (%)
68	30	38	43

Table 5.

Maximum homology of bacterial species

Name of the bacterial species obtained	Number of cases
*Salmonella Para typhi A
*Pantoea endophytica
*Pantoea 77000
*Enterobacter agglomerans	27
*Salmonella entrica
*Salmonella typhymurium
*Unculturable sheep mite bacterium well	3
*Burkholderia pseudomalla
*Burkholderia cepacia
*Burkholderia sardecolla

A total of 10% (3/30) of the sequences showed homology with the bacterial strains of Burkholderia cepacia (AF042161) and B. sordicola (AF148554).

16S rDNA sequence analysis

The DNA sequence match analysis revealed that the isolates share 90–100% similarity among themselves. All the sequences except three of the sequences show the closest relationship with the Pantoea P102 bacterial species than any other Enterobacteriaceae family members (Pantoea 77000, E. agglomerance, etc.) obtained closest homologous using the BLAST search. The partial DNA 16S rDNA sequences representing approximately 90% or less percent of the 16S rRNA gene similarity have been deposited in the Genebank database (accession numbers; Table 6).

Table 6.

The partial DNA 16S rDNA sequences deposited in the Genebank database

Serial number	Sequence	Gene Bank I.D.
1	16RA	483939
2	49R51	483941
3	83RA	480232
4	83RA51	478722
5	80R	482415
6	90R51	482431
7	19RA	478734
8	1R	482487
9	22R	482517
10	20RA	478768
11	23R	482539
12	23RA	478782
13	2R2	478784
14	3R2	478788
15	40RA	478792
16	4F	478796
17	5F	482579
18	5R	478832
19	22R	478838
20	16R53	478906
21	19RA	478908
22	1A29R	482701
23	22RA	482705
24	25RA	478912
25	2R53	482707
26	40R	482709

Twenty‐seven of the 30 sequences obtained shared a close relationship with the 16S rDNA sequence of Pantoea P102. Identical sequences were eliminated while constructing the phylogenetic tree and for evolutionary distance calculation. All these sequences shared a range of 99–88% similarity with Pantoea P102 species and 20–26% similarity with other Pantoea species (77000, endophylica) and E. agglomerance. Three of the sequences showed close relationship with Burkholderial species, which share 78.6–69.5% similarity and 22–26% similarity with other 16S rDNA sequences compared with it.

The evolutionary distances derived from the comparison of the 16S rDNA sequences (Fig. 1) of all the non‐identical isolates and the members of Enterobacteriaceae family that showed a close relationship was Pantoea P102 and the second closest relative was B. cepacia. This study was repeated with several different subdomains of 16S rDNA sequences and bootstrapping analysis was performed to avoid artifacts. Neighbor‐joining, parsimony and maximum‐likelihood analysis were then undertaken. In the analyses of the isolates, except for three, all were closely associated with Pantoea P102 strain than any other strain of the same genus and other related members of Enterobacteriaceae, the second closest relative of these isolated was B. cepacia. The remaining three isolates showed the closest relationship to B. cepacia.

Figure 1.

Descriptive statistics report for the five groups of men with reproductive failure

DISCUSSION

THE STATISTICAL ANALYSIS of various sperm parameters in men with pyospermia and sterile cultures, before and after antibiotic therapy was done and the Student's t‐test has shown high significance (P < 0.001), which suggests the presence of unculturable microbes in patients, causing asymptomatic bacterospermia, there by damaging the spermatozoa. The white blood cell count release reactive oxygen species and reactive oxygen species in the sperm result from the metabolism of oxygen by the sperm cell. These species include superoxide anion, hydrogen peroxide and nitric oxide. The excess amounts of these chemicals in the sperm combined with abnormally high concentrations of certain enzymes may indicate that the involved sperm are immature, damaged or abnormal. Excess levels of reactive oxygen species can produce oxidative damage to important components of the sperm, thereby effecting fertility potential of the spermatozoa.

Therefore, more efficient screening techniques (such as molecular biological techniques) should be developed for screening such fastidious organisms eluding from cultures, there by controlling the unwanted empiric antibiotic therapy.

Using molecular biological techniques it was found that Pantoea P102 was closely related to the 90% of the isolates. This bacterium belongs to Proteobacteria and the gamma subdivision of the Enterobacteriaceae family. The slight heterogeneity observed among the various isolates could be due to the presence of multiple, genetically distinct microbial populations.

Pantoea P102 was never before identified in infertile men. This bacterial species was also never before identified in human beings. The pathogenicity and culture condition of Pantoea P102 for human specimens is not known. This bacterial species was isolated and characterized as diazotropic endophyte, from grasses and its effect on plant growth by Riggs et al. ¹⁶ and no other published reports are as yet available on this species of Pantoea.

The other bacterial specieses showed homology with the bacterial isolates of the present study were also included for the phylogenetic analysis, were Pantoea 77000, whose 16S rDNA sequence analysis was carried out by Drancourt et al. ¹⁷ which was obtained from environmental studies. Pantoea enodophytica, characterized by phylogenetic analysis of endophytic populations, was isolated from plants by Rojas et al. ¹⁸ Pantoea agglomerans (AF157688) 16S rDNA was isolated by Raoult ¹⁹ and submitted (10‐Jun‐1999). Although all these bacteria were obtained as close relatives on homology studies, they were found to be distinctly related on phylogenetic analysis and only 20–25% similarity with the 16S rDNA isolates obtained in the present study.

The second nearest relative of all the isolates and closest relative of three bacterial isolates obtained in the present study was B. cepacia. The percentage similarity was 69–78%. Burkholderia, formerly called Pseudomonas cepacia, is a non‐fermenting Gram‐negative rod, first described by Bukholder in 1950, ²⁰ from rooting onions. It is occasionally isolated from soil and rooting bee trunks. Critically associated without breaks due to extrinsic sources‐infected blood gas analyzer, aibuterol neutralization solution and oral mouth wash. It is increasingly common organism associated with cystic fibrosis, an autosomal recessive genetic disease. It is less commonly associated with sinusitis, male infertility, cirrhosis of the liver, bowel obstruction, diarrhea, constipation and hemoptysis. Another important aspect of B. cepacia is it is extremely resistant to almost all antibiotics. Eventually, it may be killed by drugs such as Bacterium, but most B. cepacia strains cannot be killed by braomycin, piperacillin, ceptazidine and ciprofloxacin. This makes finding effective antibiotic treatment extremely difficult. Therefore, researchers should come forward to identify unculturable microbes causing silent damage to various organs and organ systems using molecular biological techniques, as empiric antibiotic therapy may give rise to many more forms of such fastidious organisms that are eluded from routine culture conditions and become resistant to most of the available antibiotics. Therefore, there is an urgent need to undertake further studies for a better understanding of these unculturable organisms, infecting the genito‐urinary tract and also other organ systems leading to unexplained health complications using molecular biological techniques as these can characterize an organism even if it is unculturable based on their genotype rather than their phenotype.