Detection of Ureaplasma Biovars and Subtyping of Ureaplasma parvum among Women Referring to a University Hospital in Morocco

Objectives The aim of this study was to determine the prevalence of Ureaplasma biovars and Ureaplasma parvum (U. parvum) serovars, their associated risk factors, and genital STI-related symptoms. Methods DNA obtained from cervical samples of 1053 women attending the department of Obstetrics and Gynecology and the laboratory of pathological anatomy of Hassan II university hospital of Fez, Morocco, was used to detect Ureaplasma biovars (U. urealyticum and U. parvum) and to subtype U. parvum by polymerase chain reaction (PCR). Results Of the 1053 women examined, 25.4% (268/1053) were Ureaplasma positives. The rates of U. urealyticum and U. parvum were 12.1% (128/1053) and 7% (74/1053), respectively, and the copresence of these biovars was noted in 6.3% (66/1053) cases. The U. parvum subtyping revealed a predominance of the serovar 3/14 (61.4%). The association of demographics variables with Ureaplasma biovars was studied and shows that the age (“<30” years) seems to be a risk factor of Ureaplasma spp. and U. urealyticum carriage (OR 1.729, 95% CI [1.113-2.687] and OR 1.848, 95% CI [1.026-3.330], respectively). There was no difference in the prevalence of Ureaplasma type regarding symptoms. However, a significant association was found between U. parvum serovar 1 and infertility (P = 0.011). Conclusion This first study conducted in Morocco provides an idea on Ureaplasma biovars and U. parvum serovars circulating in this region, their associated risk factors, and genital STI-related symptoms. Therefore, further studies are required to clarify and confirm the pathogenic role of these Ureaplasma species.


Introduction
Urogenital Ureaplasma belongs to the normal commensal flora of the human genital tract [1,2]. However, it can be pathogenic when its bacterial load is ≥10 4 organisms per ml (infectious dose). This rate is commonly accepted as a burden indicating an infection that should be treated [2,3]. In fact, it can be associated with many gynaecological or obstetric pathologies such as nongonococcal urethritis, pelvic inflammatory disease, premature birth or late abortion, and infertility [4].
Human Ureaplasma spp. include two human pathogen species: Ureaplasma urealyticum (U. urealyticum) (biovar 2) and Ureaplasma parvum (U. parvum) (biovar 1). The specific identification of each species is based on molecular methods [5], and some studies shows that U. parvum is more common than the most pathogenic U. urealyticum [6]. The colonization rate in healthy women is about 18-87% for U. parvum and about 6-10% for U. urealyticum [7].
Based on biochemical and genetic characteristics [5], U. parvum (parvo biovar) is divided on four serovars (1, 3, 6, and 14) and U. urealyticum (biovar T960T) is separated on three subtypes. Subtype 1 includes serovars 2, 5, 8, and 9; subtype 2 comprises serovars 4, 10, 12, and 13; and subtype 3 contains serovars 7 and 11. A Polishian study showed that infection of the upper genitourinary tract with U. parvum is more common in infertile women than in fertile women [2]. Several studies suggest that the pathogenicity of Ureaplasma may be serotype-dependent, and others have shown that some serotypes are more frequently associated with syndromes than others [8][9][10]. U. parvum serovar 3 is the most frequently detected in infertile women and men, and some studies have revealed that the infiltration of U. parvum serovars 1, 3, and 6 caused morphological changes of the external genitalia in female mice, which can cause disorders of the superior genital tract that can lead to infertility [11]. However, U. parvum serovar 6 is the second most prevalent (in both women and men) and is associated with premature birth [12][13][14]. It is the leading cause of death among children under five years in developing countries [15].
In Morocco, there is no information about Ureaplasma biovars prevalence, serovars distribution of U. parvum, and their association with the risk factors and genital sexually transmitted infection-(STI-) related symptoms.
The present study aims at determining the prevalence of Ureaplasma biovars and U. parvum serovars, their associated risk factors, and genital STI-related symptoms.

Patients and Sampling.
A prospective study was conducted from 2013 to 2015 among women attending the Department of Obstetrics and Gynecology and the Laboratory of Pathological Anatomy of Hassan II University Hospital of Fez, Morocco, to determine the prevalence of bacterial STI and to characterize species [16]. All collected cervical samples were used to determine the prevalence of Ureaplasma biovars and to subtype U. parvum. All demographics and clinical data of the patients are available. Patients were divided in two groups, asymptomatic women who came to the gynecological examination because of a routine checkup and symptomatic women who had at least one of the following symptoms: leucorrhoea, pelvic pain/dyspareunia, pruritus, menorrhagia, metrorrhagia, or dysuria.

Ethics.
The study was approved by the Institutional Review Board of Fez, Morocco (No. 02/15), and a written consent was obtained from all women.

Ureaplasma Biovars
Detection and U. parvum Subtyping. U. urealyticum and U. parvum were detected in cervical samples by PCR using UreaA-B (F and R) and UMS-57/UMA222 primers, respectively, as described previously [5,17]. The samples that were U. parvum positives were further typed, and their serovars were determined using a previously described PCR targeting the mba (multiband antigen) gene [5]. The specific primer pairs used as well as the size of the generated products were described in Table 1. 2.4. Statistical Analysis. Statistical analysis was performed using the SPSS (version 20) software. The different correlations were made using the chi-squared or Fisher's exact tests. The multivariate analysis was carried by binary logistic regression to determine the risk factors including all the variables with P ≤ 0:20 in the initial model. The results were expressed as odds ratio (OR), 95% confidence intervals (CIs), and P values. In all tests, a P value < 0.05 was considered as significant.  The U. parvum typing shows that the serovar 3/14 was the most frequent (80/140, 57.1%), followed by the serovar 1 and the serovar 6 (20%, 28/140 and 17.9%, 25/140, respectively). This distribution was different when considering the pregnant women only. Thus, the serovar 3/14 is still the most predominant (64.5%) followed by serovar 6 (25.8%) and then the serovar 1 (9.7%) ( Figure 1).

Risk Factors Associated with Ureaplasma Biovars and U. parvum Serovars.
To determine the risk factors associated with Ureaplasma biovars and U. parvum serovars, a univariate analysis was performed by excluding the mixed cases of U. urealyticum/U. parvum (N = 66) and U. parvum serovars (N = 7).
Variables used for this analysis include sociodemographic factors, medical history, and sexual behaviour. The results of this analysis are presented in Table 3.
The correlation of Ureaplasma biovars and U. parvum serovars to the age shows that Ureaplasma spp., U. urealyticum, and U. parvum serovar 3/14 had a peak of prevalence in women "<30" years-old. However, U. parvum serovar 6 was found maximally in the "30-50" years-old women and U. parvum serovar 1was more frequent in the ">50" yearsold age group ( Figure 2). Thus, the univariate analysis shows a significant association between Ureaplasma and "<30" years of age (P = 0:048) ( Table 3).
All the variables with P value ≤ 0.2 were used in a multivariate analysis, and the results are presented in Table 4. Age ("<30" years) seems to be a risk factor of Ureaplasma spp. and U. urealyticum carriage (

Correlation between Ureaplasma Biovars and U. parvum
Serovars and Genital STI-Related Symptoms. In order to study the correlation between genital STIs-related symptoms and Ureaplasma biovars, a statistical analysis was performed. For this analysis, Ureaplasma negative cases and the coinfection of Ureaplasma spp. with other STI (Human    Table 5. Thus, 36 women were with genital STI-related symptoms vs. 51 of asymptomatic women, and 5 women were infertile vs. 82 of fertile women. U. urealyticum and U. parvum were equally distributed among symptomatic and asymptomatic patients. However, a higher frequency of U. parvum was observed in infertile women (Table 5).

Infectious Diseases in Obstetrics and Gynecology
To determine the correlation between the presence of genital STI-related symptoms and U. parvum serovars, cases with multiples serovars (N = 7) were excluded. The results were presented in Table 6. The results show a significant association between U. parvum serovar 1 and infertility (P = 0:011) ( Table 6).

Discussion
Ureaplasma spp. mainly resides on the mucous surfaces of the urogenital tract in adults or in the respiratory tract of    [18]. These species can be responsible of nongonococcal urethritis and pregnancy complications. Their incidence is higher in women compared to men [19], and their colonization was related to age, low socio-economic level, multiplicity of sexual partners, ethnicity, and oral contraceptives uses [20].
To the best of our knowledge, this is the first study conducted in Morocco with the aim at determining the distribution, the prevalence of Ureaplasma biovars and the circulating U. parvum serovars, and their association with risk factors and genital STI-related symptoms.
The present study shows that 25% of women harbored Ureaplasma spp. This rate is slightly similar to that obtained in an Italian study (23%) [21], but lower than the rate obtained in Croatian study (34%) [3] (Table 7). Nevertheless, in the Italian study, the obtained prevalence represents only cases of Ureaplasma culture-positive with a bacterial load of ≥10 4 CFU (infection cases). This can explain the low rate obtained compared to our results, where U. urealyticum detection has been made directly by PCR without bacterial quantification.
Regarding Ureaplasma biovars, our results showed a predominance of U. urealyticum (18%; 194/1053) compared to U. parvum (13%; 140/1053) (considering the mixed infection/colonization). This distribution seems to be inversed compared to that reported on other geographical area such as in Italia and Croatia [3,21]. The U. parvum subtyping results show that U. parvum serovar 3/14 was the most prevalent followed by serovars 1 and 6, respectively. Similar distribution was observed in the Italian study even if the rates are different [21] (Table 7). The study population and geographic location can explain this difference.
In nonpregnant women group and independently of their fertility status, the prevalence of U. urealyticum and U. parvum was 11.6% (99/856) and 6.8% (58/856), respectively (excluding cases of mixed detection). These distribution and prevalence are different from that reported in a Brazilian study, which was marked by high prevalence of U. parvum (60.6%) [22] (Table 7). The difference is notably observed at U. parvum serovar distribution level and was marked [22], in comparison with our result, by the lowest rate of serovar 1 and higher cases of coinfection/colonization [22] (Table 7). This difference may be related to geographic area, sample size, and the sexual activity.
In pregnant women group, the U. urealyticum is more predominant than U. parvum (14.7% (29/197) vs. 8.1% (16/197)), respectively (and considering simple infection/colonization only), inversely to results obtained on Australian women [14] (Table 7). The difference was also obtained in U. parvum type distribution that even if the serovar 3/14 is predominant in both populations, the serovar 1 is largely prevalent than serovar 6 in the Australian study [5] ( Table 7). Nevertheless, our serovars distribution is similar to that recently obtained in an American and another Australian studies (where serovars 3 and 6 are the most frequent) [14,23] (Table 7). This can raises the question of epidemiological evolution or changes of species distribution over the time. The fact that in these later studies, serovar 6 was significantly associated with preterm delivery [14,23] lets us suppose that pregnant women of our series, carrying serovar 6 (25.8%), were probably at risk of premature delivery. Thus, it is necessary to supervise their pregnancy progress in order to verify this hypothesis and to strengthen screening programs for Ureaplasma to prevent complications.
In our study, the presence of multiple U. parvum serovars was not detected in the group of pregnant women while it was reported in 8% and 2.8% of cases in Australian studies [5,14] (Table 7). This difference may be related to the geographic location and sexual behaviors of women as well as that of their partners.
The correlation between U. parvum serovars and age has been studied and the results show that high levels of U. parvum serovar 3/14 and serovar 6 were mainly obtained in young women "≤50 years" (which can be related to sexual activity), while the U. parvum serovar 1 was more common among women aged ">50" years. This can be related to the hormonal status (menopause) of these women as reported by Iwasaka et al. who described the presence of Ureaplasma spp. in vaginal flora of 25% of postmenopausal women [24]. In fact, menopause may be associated with vaginal atrophy, thinning of the vaginal wall, dryness, and changes in pH due to a lack of estrogen, factors that may affect sexual function, and promotes infection/colonization [25].
Using logistic regression models, a significant association was obtained between the age group "<30 years" and Ureaplasma spp. (P = 0:048). These results are not surprising and confirm the results of previous studies [26,27] as well as the fact that genital mycoplasmas are linked to sexual activity. In our series, most of the recruited patients declared that they had only one sexual partner during their life (their husband) which is related to socio-cultural and religious context. Thus, the association between sexual behavior and Ureaplasma biovars cannot be performed.
The correlation between Ureaplasma biovars and genital STI-related symptoms shows no association neither for U. parvum nor for U. urealyticum. These results are consistent with those of other studies conducted in Australia, Slovenia, and Croatia [3,28,29]. However, an Italian study reported that U. urealyticum and U. parvum serovar 3/14 were   [21]. In our study, a significant association between U. parvum serovar 1 and infertility was obtained (P = 0:011). To confirm this data, the study of larger population and refinement of Ureaplasma spp. molecular characterization will be of interest.
There were some limitations of the present study: firstly, the patients participating in this study may not be truly representative of the Moroccan population; in other words, the data derived from this region may not reflect the situation of other geographical regions and studies in the other Moroccan regions are required. Secondly, following the successive recruitment of patients, the sample size in the groups of asymptomatic and symptomatic women was different. Likewise, the cultivation and quantification of Ureaplasma spp. were not carried out. Therefore, simple colonization cannot be distinguished from infection. However, our main objective was to determine the prevalence of Ureaplasma biovars and U. parvum serovars, their associated risk factors, and genital STI-related symptoms.

Conclusion
In this study, the distribution and prevalence of Ureaplasma biovars and each of the U. parvum serovars have been determined. Thus, the results showed the predominance of U. urealyticum compared to U. parvum. Moreover, the U. parvum subtyping revealed that the serovar 3/14 is the most prevalent followed by serovars 1 and 6 and a significant association between serovar 1 and infertility. This lets suggest the need of testing this species in the infertility cases.
The obtained results provide an idea on Ureaplasma biovars and U. parvum serovars distribution in Morocco and especially in Fez region area, their associated risk factors, and genital STI-related symptoms. Further studies are needed to clarify and confirm the pathogenic role of these species.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.

Ethical Approval
This study was approved by the Institutional Review Board of Fez, Morocco (No. 02/15).