Characterization and diversity of methicillin-resistant Staphylococcus aureus in two hospitals in Gabon

Abstract

Background

Methicillin-resistant Staphylococcus aureus (MRSA) poses a significant global health threat, particularly within the ESKAPE pathogens. Surveillance in Sub-Saharan Africa, including Gabon, remains limited despite reported high rates of Panton-Valentine leukocidin (PVL)-positive S. aureus and emerging MRSA clones.

Methods

A prospective longitudinal study was conducted at two hospitals in Northern (CHRO) and Southeastern (CHUAB) Gabon. Nasal swabs were collected from 200 patients (100 per hospital) upon admission (D0) and serially over 6 days. Environmental swabs were also taken. S. aureus isolates were identified, and MRSA confirmed of mecA gene detection. Characterization included spa typing, SCCmec typing, the lukS/F-PV gene detection, antimicrobial susceptibility testing (EUCAST), and detection of the dfrG trimethoprim resistance gene. Clonal complexes (CC) were inferred.

Results

MRSA nasal carriage prevalence was worrying high (60% of S. aureus carriers overall), reaching 70% at CHRO. PVL genes (65% of MRSA isolates) were occurring equally in community-associated (CA, 63%) and hospital-associated (HA, 69%) MRSA. CC5-ST5 (primarily spa t311/t003, SCCmec IV) dominated CA-MRSA (37%) and environmental isolates. Other virulent CA-MRSA clones (CC8, CC80, CC152) were detected. A high proportion of strains were non-typeable by MLST-CC (72% patient, 78% environmental) and SCCmec (37% CA-MRSA), suggesting novel lineages. MRSA exhibited resistance: tetracycline (59%), trimethoprim-sulfamethoxazole (SXT, 31%; 94% mediated by dfrG), ciprofloxacin (26%). No significant resistance difference existed between CA- and HA-MRSA. Environmental contamination was extensive (72% of environmental S. aureus were MRSA), with overlapping clones (CC5-ST5, CC121, spa t653) implicating cross-transmission. Cumulative incidence indicated a 17.1% risk of acquiring new colonization within 6 days of admission. MRSA burden was significantly higher at the northern border hospital (CHRO).

Conclusion

This study reveals a critical public health threat in Gabonese hospitals, characterized by virulent (high PVL), resistant MRSA clones circulating actively between patients and the environment. The dominance of CC5-ST5 in CA-MRSA, high rates of non-typeable strains highlight unique epidemiological features.

Introduction

Staphylococcus aureus is a part of ESKAPE (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomona aeroginosa and Enterobacter spp), a group of priority pathogens in hospital, responsible for 30% of global mortality. This bacterium causes diverse infections, including skin and soft tissue infections, bloodstream infection (sepsis), lower respiratory tract infection, catheter-associated infection and toxic shock syndrome [1–4]. Surveillance of S. aureus clones in Sub-Saharan Africa remains insufficient despite its clinical significance. However, studies conducted in this region reported the presence of methicillin-resistant Staphylococcus aureus (MRSA) clones [5]. These clones were also resistant to other antibiotic including those the latest generations. This trend arises from the ability to withstand antibiotic pressure through multiple mechanisms, notably the acquisition of antibiotic resistance genes. This could pose a major public health risk in the absence of surveillance in the sub-region. S. aureus integrates mobile genetic elements like Staphylococcal cassette chromosome mec (SCCmec), conferring resistance to beta-lactam antibiotics [6]. For example, clones acquiring SCCmec type IV illustrate this capacity for adaptation and evolutionary plasticity. This spread is realized through horizontal gene transfer of the SCCmec or transmitting existing clones between humans and animals. Colonization by MRSA increases the risk of subsequent infection. MRSA strains typically exhibit multidrug resistance including resistance to penicillin, methicillin, oxacillin, cefoxitin, amoxicillin-clavulanic acid, amoxicillin-sulbactam, ciprofloxacin, erythromycin, ceftriaxone, tetracycline and chloramphenicol [7].

The prevalence, genetic strains, and antibiotic resistance patterns of MRSA differ significantly across geographic regions worldwide. In Europe, Asia and Americas Hospital-Acquired MRSA (HA-MRSA) prevalence ranges from 38.9% in Norway and 58.4% in Portugal [8] while Community-Acquired MRSA (CA-MRSA) dominates in Japan (79%), Australia (84.9%) and India (64.7%) [9]. In contrast, Sub-Saharan Africa lacks systematic MRSA surveillance, though sporadic reports from Angola and Sao Tomé and Principe, have reported the circulation of clones CC5-MRSA-IV or CC88-MRSA-IV in hospitals [10]. Regional data suggest rising MRSA prevalence with heterogeneous clonal distributions, potentially linked to variable implementation of infection prevention measures [11].

In Gabon, prior studies report high rates of Panton-Valentin leucocidin (PVL) positive S. aureus [12] but low MRSA prevalence [13, 14]. However, no study has quantified hospital-acquired MRSA infection rates or associated patient risks. To address this gap, we characterized CA-MRSA, HA-MRSA, and environmental MRSA isolates from two hospitals in northern (CHRO) and southeastern (CHUAB) Gabon.

Materials and methods

Ethical consideration

This study is part of the ECOPAR (Parasitic Ecology) project, which has been approved by Gabon’s National Ethics and Research Committee (CNER). Although the project respecting the Helsinki Declaration has been validated agreement no. (PROT/0020/2013/SG/CNE), we obtained authorization from the director of Centre Hospitalier Regional Oyem (CHRO), Oyem, North Gabon and Centre Hospitalier Universitaire Amissa Bongo de Franceville (CHUAB), Franceville, South East Gabon. Each participant was informed about the study’s relevance to monitoring hospital-acquired strains and the sample collection procedure. We obtained oral informed consent from each participant and for children, parental assent is obtained.

Sampling sites

This prospective longitudinal study was conducted at the Centre Hospitalier Régional d’Oyem (CHRO) and the Centre Hospitalier Universitaire Amissa Bongo de Franceville (CHUAB) in Gabon. It is worth noting that both facilities were undergoing renovation during the sampling period. The town of Oyem is located in Northern Gabon, and the CHRO provides medical services, has a capacity of 170 beds divided into four sectors: general medicine, pediatrics, maternity-gynecology and surgery. The town also borders Cameroon and Equatorial Guinea, and is a major exchange point between Gabon and these two countries. Franceville is the country’s third largest city in terms of population, and is located in the south-east of Gabon, on the border with the Republic of Congo. The CHUAB is the referral hospital for the south-east region of Gabon, with a capacity of around 100 beds. It includes general medicine, pediatrics, maternity-gynecology and surgery departments (Fig. 1).

Fig. 1.

Localization of the two study sites in Gabon

Patient recruitment and sampling

Samples were taken over a one-month period, from May to June 2024 at the CHUAB and from June to July 2024 at the CHRO. Demographic and clinical data were collected, including sex, age, date of hospitalization, reason for hospitalization, antibiotic use in the last six months prior to hospitalization. Patients were included if the nasal swab was taken within 24 h of admission to hospital (D0 < 24 h). Exclusion criteria were comprised of children under 5 years, immunocompromised status and intubated patients. Nasal swabs were taken on admission from 100 patients in each hospital at D0 < 24 h, and followed (Fig. 2). Samples were also taken from the environment of wards, operating theatres and intensive care units, where we sample from beds, gurneys, tables, doors, pumps, toilets, walls, nursing equipment, stethoscopes and oxygen masks. All sample were homogenized and placed in Brain Heart Infusion (BHI) for 48 h at 37 °C, then stored at −20 °C after addition of 50% glycerol.

Fig. 2.

Scheduled swabs taken from each patient in both hospitals. A total of 328 nasal swabs were taken from 200 patients on admission, 100 from each hospital. The number of patients remaining at each timepoint is given. D0 = Day 0, D2 = D0 + 48 h, D4 = D0 + 96 h, D6 = DO + 144 h

Classification criteria for community-acquired and hospital-acquired MRSA

Samples collected on admission were classified as Community-colonization (CA) if the patient had no prior hospitalization history. Samples obtained after 48 h of admission were classified as Hospital-colonization (HA). MRSA strains were categorized as HA-MRSA or CA-MRSA based on SCCmec type: SCCmec II or III for HA-MRSA, and SCCmec IV, V, or VI for CA-MRSA.

Isolation and identification of S. aureus

Bacterial suspensions were plated on mannitol agar (BioMérieux, France). Colonies consistent with characteristic S. aureus were subculture onto trypticase soy agar (BioMérieux France). All cultures on mannitol agar were incubated at 37 °C for 48 h, and on trypticase soy agar from 18 h to 24 h.

Extraction and amplification of S. aureus, mecA, and lukS/F-PV genes

DNA extraction was performed using the heat shock method [15]. PCR were conducted to detect the 16 S rRNA, nuc, mecA, lukS/F-PV and spa genes associated with Staphylococcus aureus. The specific primers used for amplification and detection of the different genes can be found on Table 1. PCR amplification to identify S. aureus was carried out in a total volume of 25 µl reaction mixture containing 17 µl of ultra-purified water without DNAse-RNAse, 2.5 µl of 10X buffer, 0.5 µl of dNTP, 0.75 µl of MgCl2, 0.5 µl of each primer (10µM), 0.25 µl of Platinum Taq (Invitrogen, France) and 3 µl of DNA (Table 1). Amplification of the mecA, lukS/F-PVand spa gene for spa typing was carried out in a total volume of 25 µl made up of 12.5 µl of AmpliTaq Gold master mix (Applied Biosystems), 6 µl of ultra-purified water and 0.75 µl of each primer (20µM). Detection of the lukS/F-PV gene was performed on all strains identified as MRSA using the conditions defined for amplification of the specific S. aureus gene. The specific primers used for amplification and detection of the different genes can be found Table 1.

Table 1.

Used primer sequences

Target Gene	Primer Sequence (5’ → 3’)	ToC	Product Size (bp)	Reference
16 S rRNA	AAAGCGATTGATGGTGATACGGTT	56	600	This study
	TGCTTTGTTTCAGGTGTATCAACCA
mecA	GTGAAGATATACCAAGTGATT	61	533	[16]
	ATGCGCTATAGATTGAAAGGAT
LukS/F-PV	ATCATTAGGTAAAATGTCTGGACATGATCCA	56	433	[17]
	GCATCAAGTGTATTGGATAGCAAAAGC
spa	AAGACGATCCTTCGGTGAGC	59	variable	[16]
	GGTTTTGCAATGTCATTTACTG
nuc	GCGATTGATGGTGATACGGTT	58	270	[16]
	AGCCAAGCCTTGACGAACTAAAGC

Spa typing and cluster analysis

S.aureus isolates were typed by DNA sequence analysis of the polymorphic X, or short sequence repeat (SSR), region of the protein A gene (spa typing) [18, 19]. PCR amplification of the SSR region of the spa gene was performed as described previously [19, 20]. spa typing was performed by Sanger sequencing [21]. The sequences obtained were analyzed and aligned using MEGA 11 version 11.0.13. Sequences assembled were verified using the online BLAST database (http://blast.ncbi.nlm.nih.gov/Blast.cg). spa sequence analysis was performed using SPA TYPER (version 0.3.3) and compared with spa types available in the Ridom database (version 10.0.4). The resulting spa types were then grouped into spa clonal complexes (CCs) using the BURP algorithm with parameters set to exclude spa types that are shorter than five repeats [22]. We analyzed the relationship between the S. aureus subtypes (ST) described in this study and other common lineages reported in Africa was analyzed as previously described by Abdulgader and collaborators [23].

Antimicrobial susceptibility testing

Antimicrobial susceptibility testing for all MRSA isolates was performed according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST, 2024) by using the method of disk diffusion [24]. The following antibiotics were used: Ciprofloxacin (CIP, 5 µg), Tetracycline (TET, 30 µg) and Trimethoprim-sulfamethoxazole (SXT,1.25/23.75 µg).

Characterization of staphylococcal cassette chromosome mec and dfrG gene

All MSRA isolates resistant to trimethoprim-sulfamethoxazole were screened for the dfrG gene, which is the most common gene for resistance to trimethropim in S. aureus causing infections in humans [25]. The multiplex-PCR (duplex II and III, triplex IV, V et VI) protocol described by Milheiriço and collaborators [26] was used to determine if isolates harbored segments of staphylococcal cassette chromosome mec (SCCmec) elements I to VI. All PCR assays were performed according to previously protocols.

Statistical analysis

All data were analyzed using R software (version 4.2.3). Categorical variables were expressed as counts and percentages and compared using the Chi-Square test with Yates’ correction to assess significant associations between variables of interest. In addition, cumulative incidence was used to evaluate how the length of hospital stay influenced the risk of infection from the time of admission.

Results

S. aureus and MRSA prevalence in patient and environment swab samples

A total of 200 patients were included in this study, 100 recruited at the CHUAB and 100 recruited at the CHRO, and 97 S. aureus were obtained. The study population was composed of 52% (n = 103/200) of men and 48% (n = 97/200) of women (Table 2). Of the S. aureus carriers (excluding environmental isolates) detected at once during the study period, 40% (n = 40/100) were from CHUAB (four patients were colonized by two different isolates) and 53% (n = 53/100) from the CHRO. The prevalence of MRSA carriage was significantly higher (p = 0.02) at the CHRO (70%, n = 37/53) than at the CHUAB (48%, n = 19/40). Nasal carriage was similar in men (49%; n = 50/200) and females (44%; n = 43/200) and 60% of these carriers were identified as MRSA-positive in both men and women. The colonization rate did not change significantly with age. The rate for children was 51% (n = 38/75), 63% (n = 24/38) of whom were MRSA carriers. In adults and the elderly, the colonization rate was 44%, with MRSA detected in 58% (n = 28/48) and 57% (n = 4/7) of cases, respectively. MRSA carriage was not significantly associated with previous antibiotic use, level of education or hospital ward (Table 2).

Table 2.

Characteristics of individuals carrying S. aureus and MRSA

	Characteristics	n (%)	S. aureus	p-value	MRSA	p-value
Hopital	CHUAB	100(50%)	40(40%)	0.06	19(48%)	0.02*
	CHRO	100(50%)	53(53%)		37(70%)
Sex	Male	103(52%)	50(49%)	0.55	30(60%)	0.96
	Female	97(48%)	43(44%)		26(60%)
Age	Children (6–18 years)	75(37%)	38(51%)	0.65	24(63%)	0.88
	Adults (18–65 yars)	109(55%)	48(44%)		28(58%)
	Elderly (> 65 years)	16(8%)	7(44%)		4(57%)
Exposure to antibiotics	Yes	31(15%)	15(48%)	0.81	10(67%)	0.57
	No	169(85%)	78(46%)		46(59%)
Education	Student	98(49%)	47(48%)	0.26	29(62%)	0.94
	Worker	42(21%)	15(36%)		9(60%)
	Out of work	60(30%)	31(52%)		18(58%)
Department	Emergency	88(44%)	41(47%)	0.71	23(56%)	0.63
	Medicine	24(12%)	14(58%)		8(57%)
	Pediatric	34(17%)	15(44%)		10(67%)
	Surgery	26(13%)	10(38%)		5(50%)
	Maternity/gynecology	28(14%)	13(46%)		10(78%)

A total of 199 hospital swabs were obtained. S. aureus prevalence was highest in Architectural Components (including floor, wall, door, door handle, AC unit) (46.9%, 30/64 samples), followed by Medical Equipment and Furniture (including bed, gallows, table, medical cart, incubator, CO₂ tube, BP monitor, basin, bedpan) (42.1%, 48/114), and lowest in Hygiene Stations (including ‘sink, shower, toilet’) (40.0%, 10/25). Similarly, MRSA prevalence was highest in Architectural Components (37.5%, 24/64), followed by Medical Equipment and Furniture (28.9%, 33/114), and lowest in Hygiene Stations (28.0%, 7/25). However, chi-square tests indicated no statistically significant differences in prevalence across the three categories for either S. aureus (χ² = 0.51, p = 0.77) or MRSA (χ² = 1.55, p = 0.46) (Table 3).

Table 3.

S. aureus and MRSA prevalence in hospital environment

Category	Total Sample	S. aureus (n, %)	p- value	MRSA (n, %)	p- value
Medical Equipment/Furniture	110	48 (43%)	0.77	33 (29%)	0.46
Hygiene Station	25	10 (40%)		7 (28%)
Architectural Components	64	30 (47%)		24 (37%)

Distribution of S. aureus isolates

A total of 97 Staphylococcus aureus isolates were obtained in this study, 76% (n = 74/97) from the community (patient colonized on admission) and 24% (n = 23/97) from hospitals (patient colonized 48 h after admission). Of these isolates, 60% (n = 58/97) were identified as MRSA, of which 60% (n = 35/58) of CA-MRSA and 40% (n = 23/58) HA-MRSA. In addition, 88 S. aureus isolates were recovered from the hospital environment, of which 64 (72%) were MRSA. In total, 65% (n = 38/58) of PVL was obtained in this study. There was no significant difference in the prevalence of PVL genes between MRSA isolates were from in hospital (69%, n = 16/23) and MRSA isolates were from in the community (63%, n = 22/35) isolates (Fig. 3).

Fig. 3.

Distribution of MRSA and PVL isolates from nasal swabs and environmental swabs

Risk of acquiring a hospital strain

A total of 22 new cases of colonization were recorded during follow-up among the 129 patients who were uninfected on arrival at hospital. Lost to follow-up included patients discharged from hospital or who died before the end of follow-up (Table 4). The cumulative incidence over the 6-day period was 17.1%, indicating that approximately one in six initially uninfected individuals acquired the colonization during their hospital stay. At the same time, the incidence rate was estimated at 21.9 per 1,000 person-days, reflecting the speed at which cases appear in the population at risk, considering individual follow-up times. Due to high attrition (93.5% loss by D6), cumulative incidence (17.1%) and incidence rate (21.9/1,000 person-days) should be interpreted with caution.

Table 4.

Patients not-colonized (*Patients discharged or deceased before study completion)

Days	Population	Case	*Lost to Follow-up	Not colonized
D0	129	0	0	129
D2	82	11	47	71
D4	33	9	49	24
D6	13	2	20	11

Genotypic characteristics and classification of MRSA clonal complex

All 58 MRSA isolates from patients were typed. Only 31 spa repeat profiles were identified, of which 21 isolates were of known spa types and 10 isolates of unknown spa types. The assignment of the 21 investigated isolates to CCs and STs is presented in Table 4. The CC5-ST5 clone were the predominant lineage among community-associated MRSA (CA-MRSA) isolates, representing 37% (n = 7/19) of cases. This clone primarily featured the spa type t311 (4 isolates) coupled with SCCmec IV, though PVL positivity was variable (50% of CC5-ST5 isolates). What’s more, a high prevalence of PVL genes was observed across CA-MRSA isolates, with 63% (n = 12/19) testing positive. This included all isolates from clonal complexes CC8 (spa t121/t1476) and CC152 (spa t7568). However, non-typeable strains which could not be assigned to a known MLST clonal complex, accounted for 21% (4/19) of CA-MRSA isolates. These predominantly exhibited spa types t653 (SCCmec VI/PVL+) and t795 (SCCmec IV/PVL+). SCCmec diversity revealed distinct patterns between community and hospital isolates. Among CA-MRSA, SCCmec IV dominated (58%, n = 11/19) while 37% (n = 7/19) were non-typeable, suggesting possible novel cassettes. In constrast, hospital associated MRSA (HA-MRSA) isolates carried SCCmec II/III, these HA-MRSA strains were linked to CC121 (spa t1991) and NT-t653 lineages (Table 5).

Table 5.

Genotypic characteristics of MRSA isolates in patients: NT = non-typeable, + = Positive, – = Negative

Sources	MLST-CC	STs	spa type	spa repeat profiles	SCCmec	PVL
CA-MRSA	CC5	ST5	t311	26-23-17-34-20-17-12-17-16	IV	+
	CC5	ST5	t003	26-17-20-17-12-17-17-16	IV	-
	CC5	ST5	t304	11-10-21-17-34-24-34-22-25	IV	+
	CC8	ST8	t121	11-19-21-17-34-24-34-22-25	IV	+
	CC8	ST8	t1476	11-10-17-34-24-34-22-25	IV	+
	CC45	ST45	t939	04-16-34-12-34-12	IV	-
	CC121	ST121	t4198	08-23-18-17	IV	+
	CC5	ST5	t311	26-23-17-34-20-17-12-17-16	IV	-
	CC121	ST121	t314	08-17-23-18-17	IV	+
	CC152	ST152	t7568	07-23-12-17-34-12-23-02-12-23	IV	+
	NT	NT	t653	26-17-12-17-16	VI	+
	NT	NT	t795	07-23-12-21-12-12-12-17	IV	+
	CC5	ST5	t311	26-23-17-34-20-17-12-17-16	NT	-
	CC5	ST5	t311	26-23-17-34-20-17-12-17-16	NT	+
	CC8	ST8	t121	11-19-21-17-34-24-34-22-25	NT	+
	CC121	ST121	t1991	08–17.	NT	+
	CC121	ST121	t4198	08-23-18-17	NT	-
	NT	NT	t653	26-17-12-17-16	NT	-
HA-MRSA	CC121	ST121	t1991	08–17.	II	-
	NT	NT	t653	26-17-12-17-16	III	-
	NT	NT	t653	26-17-12-17-16	III	+

Among the 64 environmental MRSA isolates, only 27 spa repeat profiles were identified and 8 isolates of unknown spa types. The CC5-ST5 clone dominated environmental reservoirs, comprising 37% (n = 10/27) of isolates. This lineage primarily featured two spa types: t311 (5 isolates) and t003 (5 isolates). SCCmec typing revealed type IV in 60% (n = 6/10) and type VI in 20% (n = 2/10) of these CC5 strains. Notably, 80% (n = 4/5) of spa t311 isolates carried the PVL virulence gene. Among other significant clones, CC80-ST80 (t042) emerged as a highly virulent community-adapted strain. All isolates of this lineage carried either SCCmec IV or VI and were uniformly PVL-positive. Equally concerning was the detection of CC1-ST188 (t189), a rarely reported lineage in Africa. This isolate exhibited SCCmec V typically linked to hospital settings and was PVL-positive, suggesting a novel hospital-adapted variant with enhanced virulence. Non-typeable (NT) strains represented a substantial proportion (44%, n = 12/27) of environmental isolates. The majority (58%, n = 7/12) belonged to spa type t653 and displayed remarkable SCCmec diversity: types II, III, IV, and NT were all identified (Table 6).

Table 6.

Genotypic characteristics of MRSA in the environment: NT = non-typeable, + = Positive, – = Negative

Sources	MLST-CC	STs	spa type	SCCmec	PVL
Environment	CC1	ST188	t189	V	+
	CC5	ST5	t311 (5)	III, IV, IV, IV, VI	-,+,+,+,-
			t003 (5)	IV, IV, IV, VI, VI	+,+,-,-,+
	CC8	ST121	t314	IV	+
	CC45	ST45	t939	IV	+
	CC80	ST80	t042 (2)	VI, IV	+,+
	NT	NT	t653 (7)	III, NT, IV, II, II, II, II	-,+,+,-,-,-,-
			t10744	NT	-
			t2104	NT	+
			t2304	II	+
			t605	NT	-
			t1623	IV	-

Genetic diversity in the population

In this study, the clonal complexes: CC5, CC8, C121 were identified in the different age groups but with different associated spa types. CC45 was present only in children and the only clonal complex identified in the elderly was CC121 associated with t4198 also found in children (Table 7).

Table 7.

Typical spa traffic by age group

Age group	MLST-CC Associated	spa type	SCCmec	PVL
Children (6-18 years)	CC5	t311	NT	-
		t003	IV	-
	CC8	t1476	IV	+
	CC45	t939	IV	-
	CC121	t1991	II	-
		t1991	NT	+
		t4198	NT	-
	NT	t653	VI	+
		t653	NT	-
		t795	IV	+
Adults (18-65 yars)	CC5	t311	NT	+
		t304	IV	+
		t311	IV	-
		t311	IV	+
	CC8	t121	IV	+
		t121	NT	+
	CC121	t314	IV	+
	NT	t7568	IV	+
		t653	III	-
		t653	III	+
Elderly (>65 years)	CC121	t4198	IV	+

Analysis of CC and SCCmec profiles in patients and the environment

The distribution of clonal complexes (CC) showed that nontypeable (NT) strains predominated both in patients (72%, n = 42/58) and in the environment (78%, n = 50/64), although this dominance was not statistically significant (p = 0.48). Among the typed strains, CC121 showed a 9% (n = 5/58) tendency to associate with patients versus 2% (n = 1/64) in the environment (p = 0.10), while CC8 (5%, n = 3/58) and CC152 (2%, n = 1/58) were exclusively associated with patients. Conversely, CC5 was found both in the environment (15%, n = 10/64) and in patients (10%, n = 6/58). Rare CC45 clones and environmental exclusives CC1 and CC80 were described (Fig. 4).

Fig. 4.

Clonal complex distribution

The distribution of SCCmec in patients and in the environment showed that SCCmec II belonging to HA-MRSA was widespread both in patients and in the environment (21%, n = 12/58 and 22%, n = 14/64) respectively. In contrast, SCCmec III was more prevalent in the population (19%, n = 11/58; p = 0.04) than in the environment (8%, n = 5/64). However, SCCmec IV and non-typeable (NT) strains were almost identical in abundance. SCCmec V was exclusively environmental (6%, n = 4/58), while SCCmec VI was strongly influenced by the environment (9%, n = 6/64) (Fig. 5).

Fig. 5.

SCCmec distribution

Antimicrobial susceptibility testing

The drug resistance characteristics of CA-MRSA and HA-MRSA isolates are shown in Fig. 6A. MRSA isolates exhibited high level of resistance to tetracycline (59%, n = 34/58) but moderate for trimethoprim-sulfamethoxazole (31%, n = 18/58) and ciprofloxacin (26%, n = 15/58) (Fig. 6B). dfrG accounted for 94% (n = 17/18) of all trimethoprim-sulfamethoxazole resistant MRSA isolates. No significant difference (p = 0.05) in antibiotic resistance existed between CA-MRSA and HA-MRSA.

Fig. 6.

Sensitivity of methicillin-resistant isolates (n = 58) to different antimicrobials. (A) Distribution of antimicrobial resistance between CA-MRSA and HA-MRSA isolates, CIP Ciprofloxacin, TET Tetracycline, SXT Trimethoprim-sulfamethoxazole. (B) Bar chart representing the resistance and susceptibility profiles of different antimicrobials against methicillin-resistant isolates

Discussion

The aim of this study was to characterize of CA-MRSA, HA-MRSA and environmental MRSA for two hospitals in north (CHRO) and south-west (CHUAB) of Gabon. This study provides important insights into MRSA epidemiology in Gabonese hospitals, revealing distinct patterns compared to global trends and highlighting significant public health concerns. We observed an high MRSA carriage rate (60% of S. aureus carriers overall, reaching 70% at CHRO), exceeding rates reported globally and in prior Gabonese studies [14, 15]. This suggests a worsening MRSA epidemic in these settings. Equally striking was the high prevalence (65%) of the Panton-Valentine Leucocidin (PVL) toxin gene among MRSA isolates. PVL, a cytotoxin associated with severe skin/soft tissue infections and necrotizing pneumonia [13, 27, 28] was prevalent in both CA-MRSA (63%) and HA-MRSA (69%) isolates. This blurs the traditional distinction between community and hospital strains and significantly amplifies the virulence potential of circulating clones [13, 24, 28].

Genotyping revealed CC5-ST5 (spa t311/t003) as the predominant lineage among CA-MRSA isolates (37%) and environmental samples. This is notable as CC5 is a globally disseminated lineage typically associated with healthcare settings (HA-MRSA). Its dominance in CA-MRSA, primarily carrying SCCmec IV (traditionally a CA-MRSA marker), suggests successful adaptation and spread of healthcare-associated clones into the community, or vice-versa, reflecting blurred epidemiological boundaries [6, 11, 29]. We also detected recognized virulent CA-MRSA clones (CC8-ST8 [t121/t1476], CC80-ST80 [t042 in environment], CC152-ST152 [t7568]), aligning with findings elsewhere in Africa and globally [10, 24, 30]. Their presence in patients and the environment underscores their transmissibility and potential to cause severe infections across settings. CC121-ST121 demonstrated adaptability, found in both patients (CA and HA) and the environment with varied spa types (t1991, t314, t4198) and SCCmec types (II, IV, NT), highlighting its role in potential nosocomial transmission [22]. A high proportion of non-typeable (NT) strains by MLST-CC (72% patients, 78% environment) and non-typeable SCCmec cassettes (prominent in CA-MRSA and environment) suggests novel or under-characterized lineages circulating in Gabon. This includes the frequent spa t653 and the rarely reported CC1-ST188(t189/SCCmec V/PVL+) in the environment, warranting close monitoring [23].

While HA-MRSA isolates carried SCCmec II/III, CA-MRSA was dominated by SCCmec IV (58%) but also featured a high rate of non-typeable cassettes (37%). SCCmec IV was the most common type overall, found abundantly in CA-MRSA and environmental isolates, reflecting its global dissemination facilitated by easier horizontal transfer [7, 31]. The presence of SCCmec II in the environment (22%) mirrored patient prevalence (21%), confirming environmental contamination with typical HA-MRSA strains. The higher prevalence of SCCmec III in patients (19% and 8% environment) might indicate patient-specific colonization pressures. Detection of SCCmec V (exclusively environmental) and SCCmec VI further illustrates diverse circulating resistance mechanisms.

MRSA isolates exhibited concerning high resistance to tetracycline (59%) and moderate resistance to trimethoprim-sulfamethoxazole (SXT, 31%) and ciprofloxacin (26%). The near-total presence of the dfrG gene (94%) in SXT-resistant isolates confirms its role as the primary trimethoprim resistance determinant in human infections [25], severely limiting this therapeutic alternative. The lack of significant resistance differences between CA-MRSA and HA-MRSA emphasizes the convergence of resistance profiles in circulating strains.

The estimated cumulative incidence (17.1% over 6 days) and incidence rate (21.9 per 1000 person-days), despite limitations from high attrition, indicate a substantial risk of acquiring MRSA colonization during hospitalization. This is likely facilitated by extensive environmental contamination. That’s right, the high MRSA recovery rate (72%) from hospital surfaces demonstrates a significant environmental reservoir [32]. The overlap of clones (CC5-ST5, CC121, spa t653) between patients and the environment provides evidence for potential cross-transmission. But also, geographical context with the higher MRSA prevalence at CHRO, located near Cameroon and Equatorial Guinea, suggests potential cross-border influences on strain introduction and spread.

The interpretation of our findings is constrained by several factors. First, high patient attrition during follow-up limited the precision of hospital-acquired MRSA (HA-MRSA) acquisition estimates. Second, small sample sizes, particularly for elderly patients restricted robust subgroup analyses. Finally, the inability to determine multilocus sequence typing clonal complexes (MLST-CC) for 72% of patient isolates and 78% of environmental isolates, along with non-typeable (NT) SCCmec cassettes in 37% of CA-MRSA and many environmental strains, highlights significant gaps in current databases for African MRSA lineages and suggests the potential circulation of novel, uncharacterized strains in this region.

Conclusion

This study reveals a public health threat within Gabonese healthcare settings, characterized by virulent and resistant MRSA clones actively circulating between patients and the hospital environment. Key findings include a worrying high MRSA nasal colonization rate, extensive environmental contamination and high PVL prevalence blurring traditional community/hospital strain boundaries. The dominance of CC5-ST5 in community-onset colonization, coupled with high rates of non-typeable strains suggests unique epidemiological adaptations and potentially novel lineages circulating in Gabon. Resistance particularly to tetracycline (59%) and trimethoprim-sulfamethoxazole severely limits therapeutic options. The high cumulative incidence of new colonization and overlapping clones between patients et environments confirm active nosocomial transmission, exacerbated by the regional burden disparity observed at the northern border hospital (CHRO). These findings underscore an urgent need for enhanced genomic surveillance, stringent infection control protocols, and robust antibiotic stewardship programs to mitigate the spread of these virulent, resistant clones in Gabon and neighboring regions.

Abbreviations

CC

Clonal complex

MRSA

Methicillin Resistant staphylococcus aureus

PVL

Panton-Valentine Leukocidin

CA-MRSA

Community-Acquired methicillin-resistant staphylococcus aureus

HA-MRSA

Hospital-Acquired methicillin-resistant staphylococcus aureus

SCCmec

Staphylococcal cassette chromosome mec

MDR

Multidrug-resistant

MLST

Multi locus sequence typing

PCR

Polymerase chain reaction

Spa

Staphylococcal protein A

Authors’ contributions

Methodology : BFGN, DOE, TAT, ICMK, Investigation : BFGN, DOE, YO, ECTM, JNN, Analysis : BFGN, DOE, AD, JNN, NMLP, OM, FF, DP, BN, Original Draft : BFGN, JNN, ECTM, Preparation : BFGN, DOE, TAT, ICMK, YO, Review and Editing : BF GN, DOE, BN, Conceptualization : BFGN, DOE, BN, Supervision : DOE and BN, Project administration : AD, SED, DOE, BN and Manuscript Revision : BFGN, DOE, DP, BN.

Funding

This work was supported by the Centre Interdisciplinaire de Recherches Médicales de Franceville (CIRMF) through the ECOPAR project, and by the ANR LABEX ECOFECT (ANR-11-LABX-0048) of the Université de Lyon, within the Investissements d’Avenir program (ANR-11-IDEX-0007) managed by the French National Research Agency.

Data availability

The datasets generated and analysed during the current study are available in the BankIt2943844: PV443260 - PV443407.

Declarations

Ethics approval and consent to participate

This research was approved by the director of Centre Hospitalier Regional d’Oyem (CHRO) and Centre Hospitalier Universitaire Amissa Bongo de Franceville (CHUAB). Each participant was an explanation of the study’s impact on the monitoring of strain that can be acquired during hospital stays and the procedure for collecting samples. We obtained oral informed consent from each participant and for children, informed consent has been obtained from their parents before their participation. Environmental samples (surfaces, split, and equipment) were collected under institutional authorization.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Clinical trial

Non-applicable.