Multiplex real-time PCR using SYBR Green: Unspecific intercalating dye to detect antimicrobial resistance genes of Streptococcus pneumoniae in cerebrospinal fluid.

Meningitis caused by Streptococcus pneumoniae is still a disease of great impact on Public health, which requires immediate diagnosis and treatment. However, the culture of clinical specimens is often negative and antibiotic susceptibility testing (AST) must be performed with isolated strains. Multiplex real-time polymerase chain reaction (qPCR) has high sensitivity and specificity, produces faster results to identify the pathogen, and it can also be an important tool to identify resistance antibiotic genes earlier than AST, especially in the absence of an isolated strain. This study developed a multiplex qPCR assay, using SYBR Green as a nonspecific dye, to detect antibiotic resistance genes to predict pneumococcal susceptibility/resistance in cerebrospinal fluid (CSF) samples from meningitis patients. From 2017 to 2020, CSF samples were cultured and analyzed by qPCR to detect the main three bacteria causing meningitis. Isolated and reference strains were applied in SYBR Green qPCR multiplex to detect pbp2b, ermB, and mef genes, and the results were compared with the AST. Pneumococcal-positive CSF samples (lytA-positive gene) without isolated strains were also tested to evaluate the antimicrobial susceptibility profile in the region from 2014 to 2020. From the received 873 CSF samples; 263 were cultivated, 149 were lytA-positive in the qPCR, and 25 produced viable isolated pneumococci strains, which were evaluated by AST. Melting temperature for each gene and the acceptance criteria were determined (pbp2b: 78.24-79.86; ermB: 80.88-82.56; mef: 74.85-76.34 ºC). A total of 48/51 strains presented a genetic profile in agreement with the AST results. Resistant strains to erythromycin and clindamycin were ermB-positive, and two were also mef-positive, indicating both resistance mechanisms were present. In the retrospective study of the genetic profile of resistance, 82 lytA-positive CSF samples plus 4 strains were applied in the SYBR Green qPCR multiplex: 51% of samples presented the wild genotype (pbp2b positive and ermB/mef negative); 15% were negative for all the three evaluated, indicating pneumococci resistant to penicillin; and 17% represented the multidrug-resistant pneumococci (pbp2b negative and ermB positive or pbp2b negative and ermB and mef positive). Therefore, SYBR Green qPCR multiplex proved to be a reliable tool to identify resistance genes in S. pneumoniae and would be less expensive than multiplex qPCR using specific probes. This could be easily introduced into the routine of diagnostic laboratories and provide a strong presumption of pneumococcal resistance, especially in the absence of isolated strains.

Introduction

Meningitis is an inflammation of the defensive membranes covering the brain and spinal cord. Bacterial meningitis is a disease of great impact on Public health and can be caused mainly by Streptococcus pneumoniae. Meningitis due to S. pneumoniae occurs worldwide most commonly in infants and elderly, with an estimated incidence rate of 17 cases per 100,000 population in children less than five years of age [1]. From 2007 to 2016, in Brazil, 207,494 cases of meningitis were identified; of these, 102,249 were bacterial meningitis, of which 10,678 were related to S. pneumoniae (10.4%), and with a higher incidence in the southeastern region of Brazil (50.1%) [2]. Bacterial meningitis requires immediate diagnosis and treatment, due to the potential morbidity and high lethality. However, the cultures of clinical specimens and antibiotic susceptibility testing (AST) are slow and often negative. Modern techniques, such as the multiplex real-time polymerase chain reaction (qPCR) technique, have been applied to identify the main species causing meningitis and it was introduced by Corless et al. [3]. qPCR has high sensitivity and specificity, and produces faster results because it does not require the previous growth of the microorganism. Multiplex qPCR of the cerebrospinal fluid (CSF) proved to be a valuable method for improving the rapidity and accuracy of a diagnosis of bacterial meningitis, even in cases with CSF culture-negative results [4,5].

Besides the precise diagnostic, it is important that the information about the antibiotic susceptibility profile of the strain that is causing meningitis. This can lead to a good prognosis for the patient, as well as expand epidemiological information for Public health surveillance. For this purpose, nowadays, it is also necessary the culture of the clinical specimens to perform AST and the evaluation of strain phenotype. Notwithstanding, the issue of often culture-negative results is the same. It is known that resistance to beta-lactam antibiotics in pneumococci is due to alterations in the amino acid sequence of the PBP2x, PBP2b, and PBP1a penicillin-binding proteins [6] driven by selection with the excessive use of antibiotics. Real-time PCR is highly accurate in diagnosing meningitis caused by S. pneumoniae [4,5,7] and it also can be applied as a tool to detect antimicrobial resistance genes faster than the identification of phenotype by AST, especially in culture-negative samples.

Therefore, this study aims to develop a multiplex qPCR assay, using SYBR-Green as a nonspecific dye, to detect genes that confer resistance to S. pneumoniae in CSF samples from patients with pneumococcal bacterial meningitis. This technique would be less expensive than multiplex qPCR using specific probes and could detect resistance genes even in the absence of isolated strains.

Material and methods

Human samples

Cerebrospinal fluid (CSF) samples were collected from suspected cases of bacterial meningitis from 2014 to 2020 in health units in six municipalities (Diadema, Mauá, Santo André, São Bernardo do Campo, São Caetano do Sul, Ribeirão Pires) of the State of São Paulo, southeastern Brazil. The CSF was delivered to the Adolfo Lutz Institute, Regional Laboratory of Santo André for epidemiological surveillance, as part of the diagnostic routine of our laboratory. In the period of 2017 to 2020, the raw samples were also inoculated on a brain and heart infusion (BHI) agar plate with 5% horse blood chocolate and incubated for 24 to 48 hours at 37°C, using an anaerobic atmosphere [8] in the attempt to isolate strains causing meningitis. Isolated strains were identified by microbiological classic techniques. The genetic materials from all samples were extracted by silica column kits, according to the manufacturer’s instructions, or by heating, as previously standardized [9,10]. After extraction, all samples were submitted to the multiplex qPCR developed by Adolfo Lutz Institute of São Paulo [4 and modified by 11], to identify the main three bacteria causing meningitis: N. meningitidis (ctrA gene), S. pneumoniae (lytA gene) and H. influenzae (hpd gene), as part of the diagnostic routine of our laboratory.

The Institutional Ethical Committee has approved this work by CAAE nº 80295817.7.0000.0059 and waived the informed consent from patients for this study and forthcoming publication since clinical specimens reported here were collected for diagnostical purposes and data were analyzed anonymously.

SYBR Green qPCR multiplex

Multiplex qPCR technique was performed using the non-specific intercalating dye SYBR Green to detect different genes found in pneumococcus associated with resistance to antibiotics in the same reaction: pbp2b (penicillin-binding protein 2b gene) indicating a wild type strain and thus its sensitivity to beta-lactams; ermB (erythromycin ribosomal methylase B gene) which confers resistance to macrolides and lincosamides; and mef (macrolide efflux gene) which confers resistance to macrolides. Primers forward and reverse, except probes, from qPCR protocol described by the Centers for Disease Control and Prevention (CDC) [12] were applied, with one modification: primer mef reverse was GGTGTGAAAAGCTGTTCCAA. This alteration was necessary to obtain three fragments with different melting temperatures (Tm). Different primer concentrations were tested to combine the six primers in the same reaction. It was applied PowerUp™ SYBR™ Green Master Mix (Thermo Fisher Scientific, USA), according to the manufacturer’s instructions, using the standard cycling mode for primers with Tm < 60°C, followed by the dissociation curve. It was used the LightCycler® 480 II (Roche, Switzerland) with the following configuration of ‘detection formats’: excitation filter 483 nm, emission filter 533 nm, melt factor 1.2, quant factor 4, and max integration time 1 sec. The thermal cycling was performed at 50°C for 2 minutes for UDG activation; 95°C for 2 minutes for Dual-Lock™ DNA polymerase; and then 40 cycles of 95°C for 15 seconds, 58°C for 15 seconds, and 72°C for 1 minute. Dissociation curve conditions were in three steps: (1) 95°C for 15 seconds with 1.6°C/second of ramp rate; (2) 60°C for 1 minute with 1.6°C/second of ramp rate and (3) 95°C with 0.14°C/second of ramp rate. The acquisition mode was continuous in this third step, with four acquisitions per °C. The qPCR products were differentiated by the Tm obtained in the dissociation curve. Different samples were tested with the protocol standardized in this study: DNA from pneumococcus strain purified using silica column kits (for example PureLink™ Genomic DNA Mini Kit—Thermo Fisher Scientific, USA or QIAamp® DNA Mini Kit—Qiagen, Germany); and DNA from CSF samples, which were previously applied in the qPCR for bacterial meningitis, as described in the previous subsection and categorized as lytA-positive CSF samples, and were purified using silica column kits or extracted by heating (100 °C for 5 minutes) in a dry block.

Analysis

Amplified fragments (amplicons) were evaluated in silico about their probable Tm. It was applied the uMELT Quartz—Melting Curve Predictions Software [13], set for ‘Blake & Delcourt (1998)’ thermodynamics to estimate the temperature where helicity is around 50%.

In the melting curve analysis, the Tm was determined empirically for all genes in each sample. Firstly, all the Tm data were collected and the mean of each replicate was defined when the SYBR Green qPCR multiplex condition standardized here was applied. Then, it was calculated the mean for each condition and the interquartile range (IQR 25–75). The normal distributions of the Tm values for each gene were tested by the Kolmogorov-Smirnov test (KS normality test), D’Agostino & Pearson omnibus normality test, and Shapiro-Wilk normality test, to determine whether they were parametric or not. The results obtained with DNA purified from strains and DNA purified from CSF samples by silica column or heating were compared to the hypothetical value calculated by uMELT by one-sample t-test (parametric) or Wilcoxon signed-rank test (nonparametric). Then, the Tm from all groups for each gene were compared to determine the statistical difference by one-way analysis of variance (ANOVA) or unpaired t-test. The analyses were performed in the GraphPad Prism version 5.0, and P-value < 0.05 was considered to represent statistical significance. Then, the acceptance criteria for a positive and a negative sample were generated. The Tm range for acceptance of unknown samples was chosen by the lowest and the highest percentile for each gene. And the minimum fluorescence was empirically observed.

Comparison of techniques

SYBR Green qPCR multiplex results using pneumococcus DNA from the strains isolated in this study and reference strains, which were obtained from the Center for Interdisciplinary Procedures of Microorganisms Collection of the Adolfo Lutz Institute of São Paulo, affiliated to the World Federation Culture Collections (WFCC), were compared with the AST results performed by Adolfo Lutz Institute of São Paulo, according to CLSI protocols [14]. Briefly, strains were tested by disk diffusion methodology or minimum inhibitory concentration (MIC), depending on the tested antibiotic. CSF samples that were culture-positive for pneumococcus were also submitted to the SYBR Green qPCR multiplex to compare with the DNA purified from the strain. Moreover, lytA-positive culture-negative CSF samples and lytA-positive not cultivated CSF samples were also tested by SYBR Green qPCR multiplex to evaluate the antimicrobial susceptibility profile of pneumococcus strains that caused meningitis in the region from 2014 to 2020.

Results

Characterization of analyzed samples

In the period from January/2014 to September/2020, 873 CSF samples from patients suspected of bacterial meningitis were received, and 149 were lytA-positive in the qPCR applied in the routine of the laboratory, which means they were positive for Streptococcus pneumoniae. From 2017 to2020, 263 CSF samples were received with enough volume and therefore were cultivated, and 26 samples were culture-positive for pneumococci. Of these, 25 isolated S. pneumoniae strains were submitted to the AST, because one strain was not viable in further culture. Moreover, 26 pneumococcus reference strains, which had the antibiotic resistance profile previously characterized at the Adolfo Lutz Institute of São Paulo, were added in the analyses of this study to promote better statistic calculation, thus a total of 51 strains was applied in this study. From all 149 lytA-positive CSF samples, 82 (55.0%) had enough volume to proceed with the SYBR Green qPCR multiplex for antimicrobial susceptibility genes identification. Twenty-one of 82 (25.6%) CSF samples had the antibiotic resistance profile known, due to the isolated strain. Therefore, four isolated strains did not have more CSF samples for further analysis and the results of these 21 CSF samples were compared with the results obtained from purified DNA of isolated strains. On the contrary, 61/82 had the antibiotic resistance profile unknown, which could not be compared to an isolated strain, but it was used for retrospective study in the region.

Standardization of the SYBR Green qPCR multiplex

Firstly, to differentiate the amplicons that would be formed, the Tm for each gene was evaluated in silico. Hence, it would be possible to create a triplex protocol. The calculated Tm were 79.5 °C for pbp2b, 82 °C for ermB, and 77 °C for mef genes. It was observed that the prediction varied according to the software and the setup applied. Some online tools and the software applied here with other settings have calculated Tm far outside from the experimentally observed. Then the assay was performed to check the real Tm for each gene. Secondly, different annealing temperatures (54, 56, 58, and 60°C) were tested using the same master mix. The best temperature observed for all tested samples was 58°C. Also, after extensive tests, the final concentration of 0.3 μM was standardized to each primer when DNA purified from isolated strains was applied and 0.5 μM to each primer when DNA was obtained from CSF samples. These concentrations and annealing temperature were extremely important to avoid false positives and false negatives.

Statistical analyses of the SYBR Green qPCR multiplex

After the assays were performed, the melting curves were checked and the variances of Tm results were analyzed. For each gene, there were three evaluated groups: values of Tm after qPCR using purified DNA from cultivated strain; from CSF sample using silica column; or heating. The Tm values distributions were considered parametric for only three groups, as presented in Table 1. Additionally, each group mean was compared to the hypothetical Tm calculated by uMELT Software. Most of them were statistically different from the hypothetical Tm (Table 1). Furthermore, the groups for DNA purified by silica column or by heating from CSF samples were compared to check the possibility to unite as one group (CSF samples) per gene. It was applied unpaired t-test (Student’s) and a significant difference was observed. For this reason, these results remained separated for further analyses. However, the mef group presented a small number of samples; consequently, the values of Tm were unified in one group of CSF samples. The acceptance criteria for both types of samples (strain and CSF) were established: it was considered a positive sample when the SYBR Green qPCR multiplex presented an amplification curve that cross the threshold line, thus presenting the cycle threshold (CT) positive. Also, the dissociation curve must be analyzed to confer if the sample presented a Tm peak within the acceptance Tm range for any gene (Table 1). Moreover, this peak (-d/dT Fluorescence at 465–510) must have a height superior to one when DNA purified from strains was applied, and any height value when DNA purified from CSF sample was applied, since the Tm was within of the suggested here.

Table 1

Statistic analyses of melting temperature (Tm) values obtained in the dissociation curve after qPCR using DNA purified from cultivated strain and purified from cerebrospinal fluid (CSF) sample.

	pbp2b			ermB			mef
	Strain	CSF (silica)	CSF (heating)	Strain	CSF (silica)	CSF (heating)	Strain	CSF (silica and heating)
Number of values	39	11	8	11	4	3	3	3
Minimum	78.22	77.98	79.39	81.28	80.81	82.07	75.19	74.85
Maximum	79.03	78.83	79.90	81.51	81.19	82.56	75.23	76.34
Mean ± SD	78.74 ±0.13	78.42 ±0.23	79.69 ±0.21	81.36 ±0.07	81.06 ±0.17	82.27 ±0.26	75.21 ±0.02	75.38 ±0.83
IQR 25–75	78.67–78.81	78.24–78.53	79.44–79.86	81.30–81.40	80.88–81.18	82.07–82.56	75.19–75.23	74.85–76.34
Normality test	No	Passed	Passed	Passed	Too small a	Too small a	Too small a	Too small a
Coefficient of variation	0.16%	0.29%	0.26%	0.09%	0.21%	0.32%	0.02%	0.97%
One sample t test (P-value)	< 0.0001	< 0.0001	0.0387	< 0.0001	0.0016	0.2194*	< 0.0001	0.0779*
Acceptance criteria	78.24–79.86			80.88–82.56			74.85–76.34

Strain: DNA purified from isolated strains; CSF silica: DNA purified from CSF samples using silica column; CSF heating: DNA purified from CSF samples by heating; number of values: Number of different samples tested for each condition; Minimum: The smallest value of Tm obtained; Maximum: The largest value of Tm obtained; Mean ± SD: Mean and the standard deviation; IQR 25–75: The interquartile range of the 25th and 75th percentile; Normality test: Results obtained from AP test and SW test to check normal distribution. Passed implies P-value > 0.05, which means that the data are consistent with a Gaussian distribution; Coefficient of variation: The standard deviation divided by the mean; One sample t test: P-value obtained from comparison of the mean to the hypothetical value estimated by uMELT; Acceptance criteria: The acceptance Tm range suggested in this study.

a considered normal distribution by the KS normality test.

* No significant difference to the hypothetical value was observed.

Finally, the ANOVA test was applied, and the mean of Tm for pbp2b and ermB genes groups were statistically different (P < 0.0001). Tukey’s Multiple Comparison Test was performed to compare all pairs of means. Fig 1 presents the mean of Tm for all replicates for each strain or CSF sample, and all genes separately. It also presents the mean of all observed values with the standard deviation. It was presented as well the Tukey test P-value. For mef gene, as the groups presented a small number of samples and the results were unified in one group of CSF samples, unpaired t-test was applied. The two groups were considered not statistically different, P = 0.7389 (Fig 1).

Fig 1

Mean of the Melting Temperature (Tm) values for all replicates for each DNA extracted from strain or CSF sample by silica column or by heating.

Each column represents the data obtained for one gene which is identified at the top. Short horizontal bars indicate the mean; Vertical bars show the standard deviation; Long horizontal bars represent the significance of differences calculated by the Tukey’s Multiple Comparison Test and shown by asterisks: **P < 0.01 and ***P < 0.001. For the mef gene, it was calculated by unpaired t-test: ns means not significant (P > 0.05).

Genotypes versus antimicrobial sensitivity profiles

Of 51 pneumococci strains, 39 were positive for the pbp2b gene, which means they have the wild genotype and thus should be susceptible to penicillin (Table 2). Also, 11 strains were positive for the ermB gene, which means they are expected to be erythromycin and clindamycin resistant, and only 3 were positive for the mef gene, which means they are expected to be erythromycin resistant. A total of 48/51 strains presented a genetic profile in agreement with the AST results. Only three were discordant as presented in Table 2, called strains 357, 404, and 444. It was observed that the phenotype of 11 strains was resistant to penicillin (MIC ≥ 0.125 μg/mL) and all of them did not present the pbp2b gene in the SYBR Green qPCR multiplex standardized in this study; however, one from 40 strains susceptible to penicillin was always negative for pbp2b gene in the standardized protocol (strain 404), and another did not present amplified pbp2b gene according to the acceptance criteria (strain 357) (Table 2). Curiously, the CSF sample of which strain 357 was isolated presented the SYBR Green qPCR multiplex results in concordance with the AST result. All of the 10 strains resistant to erythromycin and clindamycin were ermB-positive. Two from 10 were also mef-positive, indicating both resistance mechanisms were present in these strains. On the other hand, one strain, called 444, also presented positive results for both ermB and mef genes, but it was susceptible to clindamycin and resistant to erythromycin. This was the third strain in disagreement with the AST results. For all the three strains in divergence, the AST was repeated to confirm the antibiotic resistance phenotype.

Table 2

Antimicrobial sensitivity profile and genotypes of clinical isolates using SYBR Green qPCR multiplex.

No. of isolates	Susceptibility profile			Presence of genes
	Penicillin	Clindamycin	Erythromycin	pbp2b	ermB	mef
34	Susceptible	Susceptible	Susceptible	34a	-	-
6	Susceptible	Resistant	Resistant	5b	6	-
6	Resistant	Susceptible	Susceptible	-	-	-
4	Resistant	Resistant	Resistant	-	4	2*
1	Resistant	Susceptible	Resistant	-	1c	1*
51				39	11	3

qPCR with the discordant result:

a, for strain 357, the pbp2b gene is not amplified according to the acceptance criteria, and it presents susceptibility to penicillin;

b, for strain 404, the pbp2b gene is not amplified, but it is susceptible to penicillin;

c, for strain 444, ermB gene is positive, but the strain is susceptible to clindamycin.

This strain is also erythromycin-resistant and has both genes (ermB and mef).

* Total of three strains presented both ermB and mef genes.

Fig 2 presents an example of melting peaks for each gene combination presented in the strains evaluated in this study, using the SYBR Green qPCR multiplex protocol developed here. Five different combinations of resistance genes were detected, as presented in Table 2, and shown graphically in Fig 2. Additionally, when analyzing CSF samples, 21 samples had enough material to be tested by the SYBR Green qPCR multiplex and 19 were compatible with the AST results applied to the isolated strains from these CSF samples. Thus, only two were discordant. One CSF sample did not present the same result as its discordant strain, as described in this subsection.

Fig 2

Fluorescence melting peaks obtained by plotting the negative derivative of fluorescence over temperature (-dF/dT) versus temperature (T) in the SYBR Green channel (465–510 nm) for the SYBR Green qPCR multiplex.

Each gene combination observed in this study is represented here: Combination 1—susceptible to three classes of antibiotics, only pbp2b gene was positive; 2—susceptible to penicillin and resistant to clindamycin and erythromycin, pbp2b and ermB genes were positive; 3—resistant to penicillin and susceptible to clindamycin and erythromycin, all three genes were negative; 4—resistant to three classes of antibiotics, with multiples mechanisms of resistance, pbp2b was negative, ermB and mef genes were positive; 5—resistant to penicillin, clindamycin and erythromycin, ermB gene was positive. The peaks were identified with the name of the genes that they represent.

Retrospective study of the genetic profile of resistance

With the established protocol in this study, the genetic profile of resistance genes presented in lytA-positive CSF samples from 2014 to 2020 was evaluated, which did not have pneumococci isolated strains, as a retrospective study of the region. Fig 3 presents the data from the SYBR Green qPCR multiplex performed with all 86 samples received in the period and which had enough material for the analyses per detected genotype. From all of the data, 61 were CSF samples with antibiotic resistance profile unknown due to the absence of the isolated strain; 21 were CSF samples with antibiotic resistance profile known as a result of the AST performed with the pneumococcus isolated strain; and 4 were also results from SYBR Green qPCR multiplex, but obtained from strains samples since there was no CSF sample for further analyses. Thus 82 CSF samples and 4 strains samples. It is possible to observe (Fig 3) that the genotype pbp2b positive and ermB and mef negative was the majority for the total of samples, and per year, except 2017 and 2020. This would be a wild genotype, with a pneumococcus strain sensitive to penicillin, macrolides, and lincosamides, and represents 51% of the samples. The second genotype observed (15%) was negative for all three evaluated genes, which would mean a pneumococcus strain resistant to penicillin, but sensitive to macrolides and lincosamides. It is worthy to note, that these samples which did not present any amplification, were confirmed as the presence of genetic material by a performance of the single qPCR for the lytA gene also using SYBR Green, in order to check DNA’s integrity after years of storage. Multidrug-resistant (MDR) pneumococci strains, defined as resistant to at least one agent in three or more antimicrobial classes [15] may be here represented by the genotypes: pbp2b negative and ermB positive; pbp2b negative and ermB and mef positive, which would mean strains resistant to penicillin, macrolides, and lincosamides. Both groups represented a total of 15 samples, 17% of all.

Fig 3

Genetic profile of resistance genes detected in all strains or CSF samples received from 2014 to 2020: A—Distribution in percent of all 86 samples per genotype; B—Distribution in percent per year of all 86 samples.

The total of results per genotype is indicated inside of each bar; Genotypes: P+E-M- means pbp2b positive, ermB and mef negative; P+E+M- means pbp2b and ermB positive and mef negative; P-E-M- means pbp2b, ermB and mef negative; P-E+M+ means pbp2b negative, ermB and mef positive; P-E+M- means pbp2b and mef negative, and ermB positive; P-E-M+ means pbp2b and ermB negative, and mef positive; P+E-M+ means pbp2b and mef positive, and ermB negative.

Discussion

Multiplex qPCR technique was performed in this study using intercalating dye SYBR Green. Since this dye is nonspecific for DNA products, any amplification will lead to an increased fluorescent readout. Therefore, besides cycle threshold (CT) evaluation, melting temperature (Tm) must be also confirmed if the target gene was amplified. For this reason, firstly, it was evaluated the hypothetical Tm for each gene. As observed in other studies, which compared different melting temperature calculation methods [16-18], diverse outputs were attained here. However, the uMELT Software [13] proved to be the most reliable tool. Moreover, when the tests were performed, it was noted that there are differences in the Tm according to the material applied. When the groups were compared, different sources of the material, DNA purified from CSF samples or from strains recovered from the same CSF sample, presented Tm slightly dissimilar which was statistically significant. This difference indicates that the experimental Tm obtained in the dissociation curve after qPCR is very sensitive to variation with the source of the genetic material. This is the first time, as far as the authors are aware, that the same genes from different sources present different Tm empirically. Thus, an interval must be adopted to be considered positive according to the origin of the genetic material. Interestedly, despite the Tm being considered different between groups, when analyzed in agarose gel, it was not observed difference in the size (bases pairs—bp). However, it is known that ordinary agarose gel has a poor resolution and is not able to distinguish few bp as by polyacrylamide gels or sieving agarose gel, which can differentiate about 10 bp or by capillary electrophoresis [19] or else by conformation sensitive gel electrophoresis, which can distinguish single-base mutation [20].

With the information obtained about Tm, it was elaborated the criteria of acceptance for the SYBR Green qPCR multiplex with the three genes studied here. These criteria were suggested when using the same parameters: master mix, real-time PCR equipment, and software of analysis, to avoid false-positive results. Genetic material from strains seemed to be easier to be amplified, producing higher fluorescence, whereas DNA from CSF samples always presented lower peaks of fluorescence, probably because they have less bacterial material. Therefore, any fluorescence for CSF samples was accepted, since the sample presented CT and an expected Tm. And for DNA purified from strain, it is suggested that the fluorescence peak must be higher than one. These criteria for acceptance were designed for the platform applied in this study (Roche Lightcycler 480 II), which is widely used in laboratories around the world. Thus, all setting details were exposed here to guarantee the reproducibility of the assay. However, the portability of the assay to other platforms is possible with some adjustments. For example, most of the real-time PCR equipment is automatic and does not have specific settings to configure, only choosing the SYBR Green detection. The statistical difference observed for Tm using different sources of DNA (strain, CSF from column, and heating) was not expected, but it will happen independently of the PCR platform because it is strongly believed that the reason for this observation was the sample quality: the presence of contaminants and human DNA. A necessary adaptation that should be considered is the fluorescence of the Tm peak for positive and negative samples. Different platforms present different fluorescence units. For example, the Roche system presents -d/dT Fluorescence at 465–510 nm, and the Tm peak must have a height superior to one for strain and any height for CSF sample; whereas QuantStudio™ 5 Real-Time PCR System (Applied Biosystems™, Thermo Fisher Scientific) presents different fluorescence unit in the melt curve plot (derivative -Rn), and the order of magnitude is in the thousands. Moreover, despite the required adaptation of this methodology, it is still worthy due to the low cost of this assay. The SYBR Green qPCR multiplex is supposed to cost US$ 15–20 as observed in the literature [21,22]. Here, it was estimated < 1 dollar considering only master mix and primers; whereas qPCR multiplex using three different probes, it was estimated the cost of US$ 8–15, considering commercial kits with master mix, primers, and in this case probes, available for molecular diagnostic of other infectious diseases. The time to run this in-house assay is longer than commercial PCR kits; however, its cost is considerably lower.

When comparing the SYBR Green qPCR multiplex with the antibiotic susceptibility testing applied to the strain, it was verified that only three were discordant. Two of them were negative to the pbp2b gene, but the strains were susceptible to penicillin. Probably, these findings were observed due to point mutations where the primers are annealed, preventing the annealing and gene amplification, but not preventing the production of the penicillin-binding protein making them sensitive to the penicillin in the AST. Similarly, negative detection of the pbp2b gene in susceptible strains by qPCR using specific probes was observed in another study [12]. In addition, another study analyzed the pbp gene by PCR–RFLP and found 24 different composite pattern profiles for three resistance genes: pbp1a, pbp2b, and pbp2x. For penicillin-susceptible strains, they found only one profile for pbp1a and pbp2x genes while pbp2b presented three different profiles. They mentioned it is not clear how some penicillin-susceptible S. pneumoniae isolates can harbor mutations in pbp2b without becoming non-susceptible to the penicillin [23]. Additionally, it was observed here that the MIC values for penicillin have corresponded to the presence or absence of the pbp2b gene: high fluorescence for the expected Tm, thus considered positive, for MICs ≤ 0.06 μg/mL; and no fluorescence for the expected Tm, consequently negative for MICs > 0.125 μg/mL). Except for the two strains mentioned above. Resistant strains with low MIC for penicillin (= 0.125 μg/mL) sometimes presented a peak for the expected Tm, however, with low fluorescence and it did not present CT.

The third strain in disagreement with the AST presented both ermB and mef genes, but it was susceptible to clindamycin and resistant to erythromycin. Therefore, it would be expected to be negative for the ermB gene and positive for the mef gene. It was hypothesized that this strain has the ermB gene, but due to some internal mutations or a deletion/addition that can produce an out-of-frame reading gene, the protein is not in the right conformation, and therefore this resistance mechanism does not work. To the best of our knowledge, there is no documentation about a strain double-positive (ermB and mef genes) with a phenotype clindamycin-susceptible. It was observed erythromycin susceptibility and clindamycin resistance profile in 3 strains of Streptococcus agalactiae portraying erm and lnuB genes by Moroi et al. [24], and they suggested a loss-of-function of ErmB protein, which could be what happened with our strain, and it keeps the resistance to erythromycin due to the mef gene. Although macrolides and lincosamides are not used to treat meningitis, the genetic characterization of antimicrobial resistance of pneumococcal isolates, which can be performed by different molecular marker methods with the isolated strain, is crucial to monitoring the epidemiology of the pneumococcal disease. The methodology developed here showed to be a reliable tool, especially when there is no pneumococcus growth, which contributes to the pneumococcal surveillance.

After standardization of the SYBR Green qPCR multiplex, it was applied to all lytA-positive CSF with enough volume, which could produce a profile of antibiotic resistance of pneumococci strains that caused meningitis in the region from 2014 to 2020. It was possible to observe the prevalence of the wild-type strain (51%), which means strains susceptible to all antibiotics corresponding to the evaluated genes. Furthermore, MDR pneumococci had great representativeness, a total of 17% of all samples, with the genotype pbp2b negative and ermB positive or pbp2b negative and mef, ermB genes positive. However, it indicates that MDR pneumococcal invasive strains were less common in the region of this study, as observed in other works. For example, 33% of MDR strains were identified by Sharew et al. [15]; 56.3% were identified by Ma et al. [25]; and the nationwide surveillance following 10-valent pneumococcal conjugate vaccine introduction in Brazil detected 25% multidrug-resistance in 2017–2019 [26]. All these works performed MIC test for a great variability of antibiotics as the AST, which could explain the higher percents. Another finding was that 10 samples (11%) presented both ermB and mef genes, which included CSF with unknown AST and isolated strains, 70% of these were 19A serotype. In the 2000s, reports about pneumococci with both resistance genes were occasional, but the concurrence of the genes increased considerably worldwide [27], indicating the expansion of multi-resistant clones [28]. The amount of double-positive pneumococcus 19A serotype represented 8% of the total samples analyzed here. In Brazil, it was observed an increase of 7% to 16.4% in the incidence of invasive pneumococcal disease caused by serotype 19A, after the 10-valent pneumococcal conjugate vaccine introduction, especially in 2016–2017 [29]. The molecular detection of both ermB and mef genes in pneumococcal invasive strains was higher in comparison with other studies in Brazil (3 in 159 strains– 1.9%) during 2010–2012 [30] and 7% during 2012–2013 [31]. Probably, because these studies evaluated the genes’ presence in isolated strains. Here, we evaluated CSF samples and isolated strains, which increased the chances of detection. And lastly, it should be mentioned the decrease in the number of samples in 2020 occurred probably due to the COVID-19 pandemic. It was supposed that the use of masks and social distance contributed to the drop in meningitis cases.

Conclusions

Despite the differences observed in Tm when applying the same gene from different sources, the SYBR Green qPCR multiplex proved to be a reliable and inexpensive tool to identify resistance genes in Streptococcus pneumoniae. This could be easily introduced into the routine of diagnostic laboratories and provide a strong presumption of pneumococcal resistance, even in the absence of isolated strain, which would improve and strengthen the surveillance of pneumococcal diseases.

Transfer Alert

This paper was transferred from another journal. As a result, its full editorial history (including decision letters, peer reviews and author responses) may not be present.

22 Mar 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Jose Melo-Cristino, M.D., Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating in your Funding Statement:

(This work was supported by The São Paulo Research Foundation, FAPESP grant 2017/03022-6 (IBC) and 2018/22718-4 (MBS).

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.)

Please provide an amended statement that declares *all* the funding or sources of support (whether external or internal to your organization) received during this study, as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now.  Please also include the statement “There was no additional external funding received for this study.” in your updated Funding Statement.

Please include your amended Funding Statement within your cover letter. We will change the online submission form on your behalf.

3. We noted in your submission details that a portion of your manuscript may have been presented or published elsewhere.

(PLOS Biology (PBIOLOGY-D-21-03354).

They suggested that I could transfer this manuscript to PLOS ONE, and I agreed)

Please clarify whether this [conference proceeding or publication] was peer-reviewed and formally published. If this work was previously peer-reviewed and published, in the cover letter please provide the reason that this work does not constitute dual publication and should be included in the current manuscript.

4. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

5. We note that you have included the phrase “data not shown” in your manuscript. Unfortunately, this does not meet our data sharing requirements. PLOS does not permit references to inaccessible data. We require that authors provide all relevant data within the paper, Supporting Information files, or in an acceptable, public repository. Please add a citation to support this phrase or upload the data that corresponds with these findings to a stable repository (such as Figshare or Dryad) and provide and URLs, DOIs, or accession numbers that may be used to access these data. Or, if the data are not a core part of the research being presented in your study, we ask that you remove the phrase that refers to these data.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors present an adaptation of a previously proposed assay, but now based on melting curve analysis, for the identification of penicillin, macrolide and lincosamide resistant pneumococci in culture negative CSF specimens. The authors advance no justification for adapting the existing assay. The paper then presents a lengthy description of the statistical analyses performed and of the method development which can be greatly shortened and simplified. It is unclear what can be expected from 4/51 strains used to validate the assay since they do not seem to have been tested for their susceptibility to clindamycin or clindamycin and erythromycin. Sensitivity and specificity seem to be high (only 3 strains out of 47 had inconsistent results with phenotypic assay). A major undiscussed point is the portability of the assay to a different platform than the one used by the authors, which can limit its applicability. The authors then analyzed 86 samples received at their center between 2014-2020 (but only 61 were culture negative). The paper should be shortened to focus on the method and its appropriateness to determine pneumococcal susceptibility in CSF samples. Although overall the English is acceptable the paper would still benefit from a revision of English use.

Major points

1) The authors should clearly state the benefits of adapting the existing probe-based assay to a melting-curve based assay. If the motivation is economical, a paragraph stating the price differences should be presented and a convincing argument presented to support this.

2) The detailed description of the statistical tests employed should be shortened and only the relevant sections included. The main message of the paper is the method and its validation, and the paper should focus on this. The entire section on the choice of Tm prediction software (and the related sections in the discussion) should be removed. It is nonsensical to choose a prediction software and method based on its agreement with the empiric determined Tm, as is stated in lines 379-380.

3) The paper should clearly state what were the validation conditions used. If this was against the PCR on the DNA extracted from all the strains or, which would me much more interesting, against the DNA extracted from the CSF from which some of the strains were isolated. The discussion of results with different PCR conditions than the ones used (ex: lines 291-297) should be deleted.

4) The susceptibility of all isolates to all targeted antimicrobials must be known (table 2).

5) Given the variations in melting curves seen when DNA was prepared from different sources, what would be the robustness of the assay when run in different platforms? Although the Roche Lightcycler 480 II is quite a popular platform it is by no means the only one. A few sentences discussing the potential portability of the assay to other platforms would be important for a wider interest in the paper.

6) The purpose of the retrospective analysis and how it could impact on the clinical management of the patients or how the information could help in the surveillance of pneumococcal infections in this area must be clarified or this section removed. The fact that macrolides and lincosamides are not used in the treatment of pneumococcal meningitis raises the question of the usefulness of determining the potential susceptibility to these antimicrobials, a point that should be addressed.

7) The discussion should be shortened and focused on the main message of the paper without repetition of what was stated in the results section

Other points

8) Line 121. The authors changed the reverse primer amplifying the mef gene from the previously published assay. Why was this done? A sentence should be included to explain this option.

9) English use: line 116 “and thus it is sensitivity” should be “and thus its sensitivity”; line 164 “Comparison of technics” should be “Comparison of techniques”; line 181 “Characterization of applied samples” should be “Characterization of analyzed samples”; line 291 and elsewhere “acceptation criteria” should be “acceptance criteria”

10) Lines 222-224. The central limit theorem is not applicable to 39 data points which is what is available for the pbp2 gene (table 1). This sentence must be removed.

11) Table 1. The P values have comas instead of points separating the decimals. This table repeat the data presented in figure 1. I would suggest offering the figure as supplementary material.

Reviewer #2: The paper describes validation of a previously published Taqman multiplex real-time PCR assay for detecting 3 antimicrobial resistance genes, pbp2b, erm B and mef and instead using primers with SYBR green intercalating dye. The authors describe their approach for validation and then apply the assay to several culture-negative samples previously identified as lytA positive indicating presence of S. pneumoniae.

Overall, I found the paper quite difficult to read in its current format and could use some review for correct English sentence structure and grammar.

Specific comments

Line 21-22 and 59. Not necessary to have capital letters for Public Health?

Line 71. It is not clear what you mean by qPCR “provides more complete results than the standard techniques”?

Line 164. “technics” should be “techniques”

Line 206-208. I am curious why you would need to have 2 different primer concentrations when you use isolate vs CSF DNA? Can you explain and would you be concerned with performance of the assay with varying amounts of DNA in your sample? Did you do a serial dilution series to determine LOD?

Line 267-269 and Table 1. It Is not clear why mef has only 2 groups for comparison? Why was CSF silica and CSF heating combined? Was it due to sample size?

Line 292-297. Do you think that this approach using SYBR green without a probe is accurate enough given that you mention false positive and negative results with changes in temperature and primer concentrations. I did not see any data provided showing effects of sample concentration of DNA on the Tm and positive vs negative results. Can you explain in Fig 1 the large differences between Tm for CSF (silica) vs CSF (heating)? Is this related to DNA concentrations?

Line 387-388. You state “Thus, an interval must be adopted to be 388 considered positive according to the origin of the genetic material”. I have my doubts on how accurate this method would be if say for example you used a pleural fluid or blood sample? Would you be confident that this assay would accurately determine penicillin, erythromycin or clindamycin susceptibility in these other specimen types give the variability of the assay?

Lines 399-402. Again, here you indicate that differences such as changes in PCR instrumentation would affect results? I don’t see how this method could be useful to a larger community? Perhaps validated in your own lab for testing but seems problematic in that every lab would need to revalidate the assays in their own hands.

Line 444-445. Did you look for inducible clindamycin resistance by testing this strain with D-zone test?

Line 473-475. I would be very cautious using lytA as a SYBR green assay without the probe. I would be concerned about false positives from oral streptococci.

Line 490-491. In your concluding sentences you make reference to qPCR for 3 main bacteria? I don’t see any of this data in the paper and not really sure of the relevance to the data you are showing in this paper.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

17 May 2022

On behalf of the authors, I would like to thank you both reviewers for spending time in the correction of our manuscript and for all the comments which were important contributions to improving the manuscript quality. The text has been accordingly corrected and reviewed, and missing information was added. All changes in the text have been highlighted.

Moreover, the manuscript was reviewed by an English native teacher, with many years of experience in professional editing of scientific articles. We believe that with the valuable aid of the reviewers the manuscript has been significantly improved, and would be now accepted for publication.

Reviewer #1:

1) The authors should clearly state the benefits of adapting the existing probe-based assay to a melting-curve based assay. If the motivation is economical, a paragraph stating the price differences should be presented and a convincing argument presented to support this.

The economic importance of this new assay was stated in the discussion.

2) The detailed description of the statistical tests employed should be shortened and only the relevant sections included. The main message of the paper is the method and its validation, and the paper should focus on this. The entire section on the choice of Tm prediction software (and the related sections in the discussion) should be removed. It is nonsensical to choose a prediction software and method based on its agreement with the empiric determined Tm, as is stated in lines 379-380.

According to the statistical test applied, we observed that the differences in Tm were or not statistically significant. Therefore, we thought it would be better to explain the reason for our choices and, for example, why the same samples as CSF with two different types of extraction were kept separately in different groups. However, we agreed this is not the focus of our manuscript. Thus, statistical analyses were shortened as you suggested.

The section about prediction software was replaced and modified. What we meant was not clear, so we hope that this modification solved the problem. We believe it is important to keep the message for readers that these software pieces produce extremely different results when compared with the empirically determined Tm, and the software applied in this study proved to be the most reliable because it came closest to the real values. This is extremely important in the design of primers to be applied in this kind of methodology (SYBR Green qPCR) to obtain a PCR fragment with size and Tm as expected.

3) The paper should clearly state what were the validation conditions used. If this was against the PCR on the DNA extracted from all the strains or, which would me much more interesting, against the DNA extracted from the CSF from which some of the strains were isolated. The discussion of results with different PCR conditions than the ones used (ex: lines 291-297) should be deleted.

We agreed with the reviewer and removed results with different PCR conditions. The conditions under which the method was standardized were described in the “Standardization of the SYBR Green qPCR multiplex” section and they were validated with DNA from strain and for CSF samples. The sentence with the criteria for acceptance validated in this study was slightly adapted to be clearer.

4) The susceptibility of all isolates to all targeted antimicrobials must be known (table 2).

Antibiotic susceptibility tests were performed to complete the missing data from 4 standard strains and table 2 and the text were modified accordingly.

5) Given the variations in melting curves seen when DNA was prepared from different sources, what would be the robustness of the assay when run in different platforms? Although the Roche Lightcycler 480 II is quite a popular platform it is by no means the only one. A few sentences discussing the potential portability of the assay to other platforms would be important for a wider interest in the paper.

The statistical difference observed for Tm of fragments obtained from different sources of DNA (strain, CSF from column, and heating) was not expected. But it does not mean the assay is not applicable on different platforms. This phenomenon will happen no matter the platform since we observed it in other studies from our laboratory. Probably, the criteria for acceptance should be adapted according to the platform. A statement about this was added in the discussion section.

6) The purpose of the retrospective analysis and how it could impact on the clinical management of the patients or how the information could help in the surveillance of pneumococcal infections in this area must be clarified or this section removed. The fact that macrolides and lincosamides are not used in the treatment of pneumococcal meningitis raises the question of the usefulness of determining the potential susceptibility to these antimicrobials, a point that should be addressed.

The retrospective analysis can be useful in the surveillance of pneumococcal infections in the region, and it is shown that the methodology applied here can help to monitor the epidemiology of the pneumococcal disease, even in the absence of isolated strain. A sentence addressing this issue was included in the discussion.

7) The discussion should be shortened and focused on the main message of the paper without repetition of what was stated in the results section

In the discussion, we intended to explain all findings in our study, besides the main message of the paper which is the determination of resistance genes in samples without isolated strains, with a low-cost methodology. The discussion was changed accordingly in the revised manuscript.

Other points

8) Line 121. The authors changed the reverse primer amplifying the mef gene from the previously published assay. Why was this done? A sentence should be included to explain this option.

This change was necessary to generate an amplicon with a different Tm. Using the same primers from the previous study generates a fragment of mef with the same Tm of pbp2 fragment. This explanation was added in the material and methods section.

9) English use: line 116 “and thus it is sensitivity” should be “and thus its sensitivity”; line 164 “Comparison of technics” should be “Comparison of techniques”; line 181 “Characterization of applied samples” should be “Characterization of analyzed samples”; line 291 and elsewhere “acceptation criteria” should be “acceptance criteria”

They were changed in our revised manuscript. Also, as mentioned before, the manuscript was reviewed now by an English native teacher and experienced in scientific texts.

10) Lines 222-224. The central limit theorem is not applicable to 39 data points which is what is available for the pbp2 gene (table 1). This sentence must be removed.

It was changed in our revised manuscript.

11) Table 1. The P values have comas instead of points separating the decimals. This table repeat the data presented in figure 1. I would suggest offering the figure as supplementary material.

It was changed in our revised manuscript. However, the information presented in figure 1 and table 1 is different. The p-value in the table presents the comparison with the Tm predicted by the software. Figure 1 presents the comparison of the empirically determined Tm from all samples of the same gene (ANOVA test).

Reviewer #2:

Specific comments

Line 21-22 and 59. Not necessary to have capital letters for Public Health?

We kindly request to differ from the reviewer, but according to the English teacher who revised our manuscript, "Public health" is correct, as over here “Public" works as an adjective for the noun.

Line 71. It is not clear what you mean by qPCR “provides more complete results than the standard techniques”?

This part was removed from the text.

Line 164. “technics” should be “techniques”

It was changed in our revised manuscript.

Line 206-208. I am curious why you would need to have 2 different primer concentrations when you use isolate vs CSF DNA? Can you explain and would you be concerned with performance of the assay with varying amounts of DNA in your sample? Did you do a serial dilution series to determine LOD?

We believe that these two different primer concentrations were necessary due to DNA template concentration. DNA extracted from pure culture is massively more concentrated than in the CSF sample. Thus, it is necessary to have high concentrations to get a positive result when using DNA extracted from the CSF sample, whereas when these higher concentrations were applied to DNA from culture, false-positive results can be produced. In this study, we extensively tested this assay and we applied genetic material from CSF samples with different qualities, which means higher and lower bacterial DNA concentrations. Therefore, we are sure that with this higher primer concentration, it will be able to detect lower bacterial DNA concentration. Also, when using CSF samples, false-positive results will not be obtained since the amount of bacterial DNA is never equivalent to DNA from culture. We did not determine LOD because it is not possible to know the exact amount of bacterial DNA present in the CSF sample. This is a mixture of human and bacterial DNA. Therefore, measurement by optical density, for example, would be compromised. Also, serial dilution of DNA from culture would produce samples with known concentration; however, it does not correspond to the total amount of DNA present in the CSF sample, due to this mixture of DNA.

Line 267-269 and Table 1. It Is not clear why mef has only 2 groups for comparison? Why was CSF silica and CSF heating combined? Was it due to sample size?

This was previously explained in the first version of the manuscript in lines 233-235. These two groups were combined in one due to the small number of samples; three CSF samples were extracted by silica column and one of them was also extracted by heating. We could not perform the statistical analyzes using a group with only one sample. This gene was not found in abundance in the samples which were checked by the AST, only three samples. Now, the information is reinforced in the text.

Line 292-297. Do you think that this approach using SYBR green without a probe is accurate enough given that you mention false positive and negative results with changes in temperature and primer concentrations. I did not see any data provided showing effects of sample concentration of DNA on the Tm and positive vs negative results. Can you explain in Fig 1 the large differences between Tm for CSF (silica) vs CSF (heating)? Is this related to DNA concentrations?

We think the qPCR using SYBR green without a probe is accurate and reliable but the dissociation curve must be analyzed and Tm must be in the range established here. There is no effect of sample concentration on the positive or negative result, as observed with DNA extracted from culture is much higher than CSF sample which the same strain was isolated and they produce the same result. Surely, if you have a CSF sample with so much lower concentration, you would not have a positive result, like any other assay, due to the limit of detection. However, with certainty, from this sample would not be possible to isolate a strain due to the low bacterial concentration. What changed the outcome here was the primer concentration, and using the specified quantity of primers according to the sample (culture or CSF) should have no problems. The variation in Tm was observed with different types of extraction (silica column and heating). Unfortunately, we could not identify why this phenomenon happens. We believe this is not due to DNA concentration, since a CSF sample before the extraction with silica column and heating is the same, thus the same amount of DNA. It is hypothesized the quality of the sample could affect the Tm, for example, DNA from culture is pure, without contaminant; DNA from CSF sample purified by column has a mixture of bacterial and human DNA; DNA from CSF sample extracted by heating has many contaminants present from the fluid.

Line 387-388. You state “Thus, an interval must be adopted to be 388 considered positive according to the origin of the genetic material”. I have my doubts on how accurate this method would be if say for example you used a pleural fluid or blood sample? Would you be confident that this assay would accurately determine penicillin, erythromycin or clindamycin susceptibility in these other specimen types give the variability of the assay?

We believe that genetic material from different sources such as pleural fluid or blood samples would produce similar results, with Tm in the range established here. However, this was not performed in this study. As with any assay validation, further studies would be necessary to test with these kinds of samples.

Lines 399-402. Again, here you indicate that differences such as changes in PCR instrumentation would affect results? I don’t see how this method could be useful to a larger community? Perhaps validated in your own lab for testing but seems problematic in that every lab would need to revalidate the assays in their own hands.

This sentence was misunderstood, for this reason, it was removed. Actually, changes in PCR instrumentation do not affect the results, positive or negative. What should be changed are the criteria for acceptance and the interpretation. Because each piece of equipment has itself way to present the final result and also a specific configuration.

As with any diagnostic kit, it is validated with specific equipment and reagents by the manufacturer. Commercial kits for real-time PCR can be applied in different instruments; however, the manufacturer only guarantees if applied with the condition established and validated by them. In our study, we established the condition for two types of samples (strain and CSF), with only one reagent (PowerUp™ SYBR™ Green Master Mix - Thermo Fisher Scientific) in one piece of equipment (LightCycler® 480 II - Roche). But it does not mean the assay cannot be applied on other platforms. The use of other PCR equipment is absolutely possible. However, the criteria for acceptance must be adapted. For example, the Roche system presents -d/dT fluorescence at 465-510 nm, and the Tm peak must have a height superior to one for strain and any height for CSF sample; whereas QuantStudio™ 5 Real-Time PCR System (Applied Biosystems™, Thermo Fisher Scientific) presents different fluorescence unit in the melt curve plot (derivative -Rn), and the order of magnitude is in the thousands.

A new sentence in the discussion was included to address this issue.

Line 444-445. Did you look for inducible clindamycin resistance by testing this strain with D-zone test?

It is part of the laboratory routine of Adolfo Lutz Institute in São Paulo to evaluate the inducible clindamycin resistance by the D-zone test, but none was positive. The strains studied here showed constitutive resistance to clindamycin.

Line 473-475. I would be very cautious using lytA as a SYBR green assay without the probe. I would be concerned about false positives from oral streptococci.

The lytA gene is the standard for PCR detection of Streptococcus pneumoniae in CSF samples. At the moment of diagnosis, these samples were tested by qPCR with a specific probe for pneumococcus detection using the lytA gene. In our study, we performed with SYBR green, only to check if the genetic material was undamaged after years in freezer storage. We removed this information from the discussion and mentioned it in the results. Also, I don’t believe that oral streptococci may be present in the CSF sample.

Line 490-491. In your concluding sentences you make reference to qPCR for 3 main bacteria? I don’t see any of this data in the paper and not really sure of the relevance to the data you are showing in this paper.

We agreed and the last sentence was modified.

Submitted filename: Response to Reviewers final.pdf

Click here for additional data file.

30 May 2022

Multiplex real-time PCR using SYBR Green: unspecific intercalating dye to detect antimicrobial resistance genes of Streptococcus pneumoniae in cerebrospinal fluid

PONE-D-22-02044R1

Dear Dr. Campos,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Jose Melo-Cristino, M.D., Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors have addressed most of the comments raised by the reviewers and provided necessary information in the text.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

6 Jun 2022

PONE-D-22-02044R1

Multiplex real-time PCR using SYBR Green: unspecific intercalating dye to detect antimicrobial resistance genes of Streptococcus pneumoniae in cerebrospinal fluid

Dear Dr. Campos:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Prof. Jose Melo-Cristino

Academic Editor

PLOS ONE