Serum-IgG responses to SARS-CoV-2 after mild and severe COVID-19 infection and analysis of IgG non-responders.

BACKGROUND: To accurately interpret COVID-19 seroprevalence surveys, knowledge of serum-IgG responses to SARS-CoV-2 with a better understanding of patients who do not seroconvert, is imperative. This study aimed to describe serum-IgG responses to SARS-CoV-2 in a cohort of patients with both severe and mild COVID-19, including extended studies of patients who remained seronegative more than 90 days post symptom onset.
METHODS: SARS-CoV-2-specific IgG antibody levels were quantified using two clinically validated and widely used commercial serological assays (Architect, Abbott Laboratories and iFlash 1800, YHLO), detecting antibodies against the spike and nucleocapsid proteins.
RESULTS: Forty-seven patients (mean age 49 years, 38% female) were included. All (15/15) patients with severe symptoms and 29/32 (90.6%) patients with mild symptoms of COVID-19 developed SARS-CoV-2-specific IgG antibodies in serum. Time to seroconversion was significantly shorter (median 11 vs. 22 days, P = 0.04) in patients with severe compared to mild symptoms. Of the three patients without detectable IgG-responses after >90 days, all had detectable virus-neutralizing antibodies and in two, spike-protein receptor binding domain-specific IgG was detected with an in-house assay. Antibody titers were preserved during follow-up and all patients who seroconverted, irrespective of the severity of symptoms, still had detectable IgG levels >75 days post symptom onset.
CONCLUSIONS: Patients with severe COVID-19 both seroconvert earlier and develop higher concentrations of SARS-CoV-2-specific IgG than patients with mild symptoms. Of those patients who not develop detectable IgG antibodies, all have detectable virus-neutralizing antibodies, suggesting immunity. Our results showing that not all COVID-19 patients develop detectable IgG using two validated commercial clinical methods, even over time, are vital for the interpretation of COVID-19 seroprevalence surveys.

Introduction

The coronavirus disease 2019 (COVID-19) pandemic continues, causing considerable morbidity and mortality worldwide. The severity of COVID-19 ranges from asymptomatic to fatal pneumonitis, with mildly symptomatic patients accounting for approximately 80% of all cases according to current understanding [1]. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative agent of COVID-19, gains entry to human cells by binding the angiotensin-converting enzyme 2 (ACE2) receptor with the receptor-binding domain (RBD) of its spike (S) protein [2]. Thus, antibodies targeting the S-protein may effectively neutralize the virus [3]. Seroprevalence studies usually measure SARS-CoV-2 S- and nucleocapsid (N-) protein specific IgG antibodies; whether these antibodies correlate with protective immunity is however still unknown.

The serological responses to other beta-coronaviruses vary. Whilst all patients infected by SARS-CoV were found to develop IgG antibodies [4], some patients with mild symptoms of MERS-CoV failed to develop detectable levels of IgG [5]. Several short follow-up studies of mostly hospitalized patients have reported the development of IgG in serum against SARS-CoV-2 in the majority of patients [3, 6–10]. When patients remain seronegative, it is often concluded that seroconversion would likely occur later. Serological findings over a longer period than 30 days post symptom onset (PSO) and in patients with non-severe disease remain limited and conflicting. For example, SARS-CoV-2-specific IgG levels in patients have both been found to remain stable approximately 82 days PSO [11] and to wain 2–3 months after infection [7, 12].

Further investigation of patients who fail to produce detectable levels of IgG is lacking and antibody responses in patients with mild symptoms are also poorly described. Here, we investigated serum-IgG (S-IgG) responses to SARS-CoV-2 in a cohort of patients with both severe and mild COVID-19, profiling the patients who remained seronegative.

Materials and methods

Patients and sample collection

A cohort of 47 patients were recruited between February 25th and March 25th 2020, at the Department of Infectious Diseases, Sahlgrenska University Hospital, Gothenburg, Sweden. The study protocol was approved by the Swedish Ethical Review Authority (Registration number 2020–01771) and patients were included after written informed consent. Disease severity was divided into severe and mild: severe cases were defined as those requiring invasive mechanical ventilation or high-flow nasal oxygen, and mild cases as not requiring oxygen nor in-patient hospital care [13]. Blood samples were collected during hospitalization and/or during follow-ups. Seroconversion was defined as detectable levels of SARS-CoV-2-specific IgG antibodies in serum.

Real-Time Polymerase Chain Reaction (RT-PCR) assay

All patients had been diagnosed with SARS-CoV-2 with RT-PCR from the upper respiratory tract (pooled nasopharyngeal and throat swabs) during acute phase of the infection. Nucleic acid was extracted from clinical samples in a MagNA Pure 96 instrument using the Total Nucleic Acid isolation kit (Roche). RT-PCR targeting the RdRP region was performed in a QuantStudio 6 instrument (Applied Biosystems, Foster City, CA) using the primers and probe described [14]. Cycle threshold (Ct) values <38 were regarded as positive.

Protein expression

The SARS-CoV2 spike protein RBD (amino acids 319–541) was produced using an expression vector obtained through BEI Resources, NIAID, NIH, which is vector pCAGGS containing the SARS-CoV-2, Wuhan-Hu-1 spike glycoprotein gene RBD with C-terminal Hexa-Histidine tag (NR-52309). 293F cells (Cat nr R79007, ThermoFisher Scientific) were cultured in Freestyle 293 medium at 37°C in 5% CO2 in Optimum GrowthTM flasks (Thomson instrument company) at 130 rpm in a Multitron 4 incubator (Infors) and transfected at 2xE6 cells/ml using FectoPro transfection reagent (Polyplus transfection). Protein-containing culture supernatant (1L) was harvested after 90h, filtered using Polydisc AS 0.45 μm (Whatman) and loaded onto a 5 mL HisExcel column (GE healthcare). The column was washed with 20 mM sodium phosphate, 0.5M NaCl and 30 mM imidazole before elution of the protein using the same buffer but with 300 mM imidazole. Pooled fractions were concentrated using 10 kDa Vivaspin concentrators (MWCO 10 kDa, Sartorius), passed over a HiPrep 26/10 desalting column (GE Healthcare) in phosphate-buffered saline and finally concentrated again.

Detection of SARS-CoV-2 specific serum antibodies

Serum-IgG antibodies against SARS-CoV-2 were analyzed using two commercially available serological assays: the qualitative Architect chemiluminescent microparticle immunoassay (Abbott Laboratories, USA), measuring IgG against SARS-CoV-2 N-protein, and the quantitative iFlash 1800 chemiluminescent immunoassay (YHLO, China), which measures IgG against both SARS-CoV-2 S- and N-proteins. All samples were analyzed using both assays. IgG concentrations were obtained using the iFlash 1800 assay, and ≥10 AU/ml were defined as positive. The time of seroconversion was defined as the time-point at which the first positive serum-IgG result was observed.

The RBD ELISA was performed as previously described [15] with some modification. Plates were coated with 25μL per well at 4μg/mL SARS-CoV2 RBD in 1X PBS, incubated overnight at 4°C. Blocking was done with 100μL per well of 2% sterile filtered BSA in 1X PBS at RT for 1hr. Plates were washed with 1X PBS + 0.05% Tween using a 405 LS Washer, Biotek and incubated with 25μL sera (not heat-inactivated) per well for 1hr, with a starting dilution of 1:50 followed by 2-fold serial dilution. Secondary antibodies were diluted in 1X PBS + 0.05% Tween all 1:6000 and 25μL added to each well at RT for 1hr. Antibodies used: Goat Anti-Human IgG Fc-HRP (SouthernBiotech, cat. No. 2048–05), Goat Anti-Human IgA-HRP (SouthernBiotech, cat. No. 2050–05), Goat Anti-Human IgM-HRP (SouthernBiotech, cat. No. 2020–05). After washing, signal was developed with addition of 1-step Ultra TMB-ELISA (ThermoFisher, cat. No. 34029) and reaction was stopped after 5 min by addition of 2M H2SO4. OD for plates was measured at 450nm. AUC was calculated in GraphPad Prism 8 (GraphPad Software). The assay was verified by using 10 negative samples, collected prior to COVID-19 outbreak and 10 positive samples from patients hospitalized for COVID-19 infection. Seropositivity was defined as AUC value over mean+2SD of negative samples.

Total concentrations of Ig A, G and M were determined using commercially available reagents on the Alinity platform (Abbott Laboratories).

Flow cytometry

Blood was collected in EDTA Vacutainer tubes and was analyzed using Multitest 6-Color TBNK reagent (337166) or anti-CD45-PerCP (345809)/anti-CD14-APC (345787) in Trucount tubes (340334) according to normal procedures using FACSCanto II flow cytometers (all from BD Biosciences, San Jose, CA).

SARS-CoV-2 Neutralizing antibody assay

Neutralizing antibodies (NAb) were determined after inactivation of the complement in serum for 30 minutes at 56°C, by incubating 25μL of 2-fold dilutions (1/2–1/264) of each serum in maintenance medium (MM) with 25 μL of 100TCID50 of SARS-CoV-2 (DE strain, isolated from sample collected February 25, 2020) in duplicate for two hours at 37°C. Thereafter the serum/virus mixture was added to confluent Vero cells (ATCC CCL-81) in 96 well microtiter plates with 175 μL MM and incubated at 37°C in a CO2 incubator. Ten-fold serial dilutions of the virus, 10–1,000 TCID50, were added in duplicate to separate wells as an infection control. The plates were examined daily, using an inverted microscope, and complete cytopathic effect (CPE) was usually complete in the virus only control-wells after three days. The presence of any CPE of the cells was then recorded in the wells and the titer of the sera was calculated as previously described [16]. Sera with antibody titers ≥4 (1/4 dilution) were considered neutralizing, demonstrating neutralization in both 1/2 and 1/4 dilutions, confirming the presence of antibodies with capacity to block infection. Sera (n = 17) from patients and blood donors from before December 2019 were used as a negative control and no neutralization was detected in any of these samples.

Statistical analysis

Time to event analysis with interval censoring was used to compare time to seroconversion between groups [17]. Differences between groups were analyzed using Welch’s t-test and longitudinal changes with paired t-test. P<0.05 was considered statistically significant. Statistical analyses were performed using GraphPad Prism 8 (GraphPad Software, Inc) and R 4.0.0.

Results

SARS-CoV-2-specific serum-IgG antibodies in severe and mild COVID-19

Forty-seven patients provided a total of 156 serum samples (mean 3.3 per patient, range 1–7), 5–117 days PSO. Of the 47 patients, 15/47 (32%) had severe and 32/47 (68%) had mild COVID-19 (Table 1). The patients with severe symptoms were older (mean age 58) and all male, compared to patients with mild symptoms (mean age 45, 56% female).

Table 1

Demographic and clinical characteristics of 47 patients with COVID-19.

	Mild n = 32	Severe n = 15	Total n = 47
Characteristics
Age, mean (range)	45 (19–71)	58 (46–81)	49 (19–81)
Female, no. (%)	18 (56)	0 (0)	18 (38)
Max level of care
Outpatient, no. (%)	32 (100)a	0 (0)	32 (68)
Hospitalized, no. (%)	0 (0)	15 (100)	15 (32)
ICU, no. (%)	0 (0)	12 (80)	12 (26)
Comorbidities
Hypertension, no. (%)	1 (3)	5 (33)	4 (9)
Diabetes, no. (%)	0 (0)	2 (13)	2 (4)
Heart Disease, no. (%)	1 (3)	2 (13)	3 (6)
Cerebrovascular disease, no. (%)	0 (0)	1 (7)	1 (2)
Asthma, no. (%)	3 (9)	1 (7)	4 (9)

a7 patients in this group were isolated in hospital to prevent viral transmission according to control policy at that time but did not require hospital care. ICU = Intensive Care Unit.

All 15 patients with severe COVID-19 developed SARS-CoV-2-specific IgG antibodies in serum in both of the commercial IgG assays (Architect and iFlash). Of the patients with mild symptoms, 29/32 (90.6%) developed SARS-CoV-2-specific IgG antibodies. Among these, 27/29 developed detectable IgG antibodies in both of the commercial IgG assays (Architect and iFlash) and 2/29 were classified as negative (index <1.4) in Architect, although the index values were clearly above negative samples (index between 0.83–0.95). Follow-up samples from these two patients presented similar reactivity in Architect which could be considered as grey-zone reactivity and as all samples were positive in iFlash, these two patients were interpreted as IgG-positive.

Among the 15 patients with severe symptoms, seroconversion was observed after in median 11 (range 7–20) days PSO (Fig 1A). Among the 29/32 patient with mild symptoms that were considered IgG positive, seroconversion was observed after median 22 (range 14–79) days PSO (Fig 1B). When accounting for the varying sampling times by using interval censoring, time to seroconversion was still significantly shorter in patients with severe symptoms than in those with mild symptoms (P = 0.04). Furthermore, we found significantly higher concentrations of IgG antibodies in patients with severe symptoms (mean 107 AU/ml) than in patients with mild symptoms (mean 65 AU/ml) within 35 days PSO (P = 0.004; Fig 1C). Within both groups, antibody concentrations did not change significantly in patients >75 days, hence the differences between the groups remained (though no longer significantly (P = 0.294, Fig 1C) due to fewer patients analyzed at this time-point).

Fig 1

SARS-CoV-2-specific serum IgG antibody responses in patients with severe and mild COVID-19.

Concentrations of serum IgG (AU/ml) over time in patients with severe (A, red) and mild (B, blue) disease. (C) Maximum concentrations of serum IgG (AU/ml) in early (1–35 days) and late (>75 days) follow-up post symptom onset. Cut-off for positive sample indicated by dotted line. ns P>0.05, ** P<0.01.

Among patients with mild symptoms, 3/32 (9.4%) did not develop detectable IgG antibodies as determined using the commercially available assays during the follow-up period, 91–105 days PSO.

Analysis of patients without detectable IgG-responses

We assessed the three patients without detectable IgG-responses further (Table 2, Fig 2). The lowest observed Ct values in the non-IgG patients were of a similar range as in the patients with mild symptoms who seroconverted (15.9–24.9 vs. 11.9–37.2). Additionally, the non-IgG patients had similar range of duration of symptoms (20–28 days) as the patients with mild symptoms who seroconverted (1–44 days, data from 26/29 patients). Flow cytometric analyses of peripheral blood cells in two of the non-seroconverting patients available for additional testing (days 91 and 105 PSO) revealed no overt signs of immunodeficiency (Table 3). Total concentrations of IgG, IgA and IgM isotypes in serum sampled at the same time-points were also within the normal adult range, indicating no antibody deficiencies (Table 2).

Table 2

Demographic, clinical and laboratory findings in three patients with undetectable levels of serum IgG against SARS-CoV-2 after COVID-19 using commercially available kits.

	Patient 1		Patient 2		Patient 3
Age	46		19		43
Sex	Male		Female		Female
No. symptomatic days	28		29		20
Lowest Ct-valuea	23		24.9		15.9
No. positive viral PCR	6		3		3
Total serum antibody concentrationsb	IgG	8.1	IgG	10	IgG	9.7
	IgA	1.8	IgA	1.2	IgA	0.9
	IgM	0.48	IgM	1.4	IgM	1.5
α-SARS-CoV-2 IgG	Neg		Neg		Neg
Neutralizing ab titerc	12		8		48
α-RBD antibodies	IgG	Pos	IgG	Neg	IgG	Pos
	IgA	Neg	IgA	Neg	IgA	Neg
	IgM	Neg	IgM	Neg	IgM	Neg

Blood samples analyzed for specific antibodies and total serum antibodies collected at day 76 (patient 2) and 91 (patient 1 and 3) post symptom onset.

aCt values <38 are considered positive.

bNormal range: IgM 0.27–2.1, IgG 6.7–15, IgA 0.88–4.5 g/L. RDB = Receptor binding domain.

cTiters ≥4 are considered positive.

Fig 2

SARS-CoV-2 viral load, IgG antibody concentration and symptom duration in three IgG-negative patients >90 days post onset of symptoms.

Ct values over time (blue circles, left y-axis), concentration of SARS-CoV-2-specific serum IgG antibodies over time (red triangles, right y-axis) and number of days of with symptoms (green bar, x-axis). Cut-off for positive viral sample indicated by dotted line. ns P>0.05, * P<0.05, ** P<0.01, *** P<0.001.

Table 3

Flow cytometric analysis of peripheral blood cells in two patients with undetectable levels of serum IgG against SARS-CoV-2 using commercially available kits after COVID-19.

Cell type	Phenotype	Patient 1a	Patient 3	Normal range
T cells	CD3+	1,1	2	0,7–2,1
T helper cells	CD3+4+	0,69	1,19	0,3–1,4
T cytotoxic cells	CD3+8+	0,35	0,66	0,2–0,9
B cells	CD19	0,08	0,25	0,1–0,5
NK cells	CD3-56+	0,12	0,15	0,09–0,6
Classical monocytes	CD14+16-	0,17	0,31	0,26–0,38

aSamples collected at 105 (patient 1) and day 91 (patient 3) post symptom onset.

Neutralizing and RBD-specific antibodies in patients without detectable IgG-responses

To further investigate the humoral immune response against SARS-CoV-2 in the three patients without detectable IgG, neutralizing antibodies, considered the golden standard of anti-viral serological testing, were analyzed in serum samples collected days 78–91 PSO. All 3 patients had detectable NAb activity (Table 2) indicating that these patients had mounted a functional humoral immune response against SARS-CoV-2. To ascertain if the neutralizing activity could be explained by low antibody levels against the RBD undetected by the commercial assays, an in-house ELISA was used (Table 2, Fig 3). Indeed, the two patients with highest levels of neutralization had measurable anti-RBD-IgG in the same samples whilst the patient with low neutralizing activity lacked detectable anti-RBD-IgG (Table 2). Compared to late serum samples (>75 days PSO, n = 20) from patients with detectable anti-SARS-CoV-2-IgG with mild and severe disease, these three patients had lower levels of anti-RBD-IgG, IgA and IgM (Fig 3).

Fig 3

Antibody responses against the receptor-binding domain of SARS-CoV-2.

RBD-specific serum IgG (A), IgA (B) and IgM (C) antibodies in patients with severe symptoms (red, n = 7), mild symptoms and IgG-positive (blue, n = 13), mild symptoms and IgG-negative (green, n = 3) collected 78–91 days post symptom onset. NC = negative controls (black dots, n = 10). Cut-off (mean + 2SD of NC AUC) indicated by dashed lines. ns P>0.05, * P<0.05, ** P<0.01.

Discussion

In this study, we describe IgG antibody responses in 47 patients during and after severe and mild COVID-19. Patients with severe disease seroconverted earlier and had higher maximum concentrations of anti-viral IgG than those with mild disease. Whilst all of the patients with severe symptoms seroconverted, three (9%) of the 32 patients with mild disease failed to produce levels of IgG detectable with commercial assays, even more than 90 days PSO. However, NAbs against SARS-CoV-2 were detected in all of these three patients.

Our results confirm previous findings that clinical severity of disease is associated with higher SARS-CoV-2-specific serum-IgG antibodies [18-21]. Studies comparing time to seroconversion between these groups are still lacking and we show that clinical severity is also associated with significantly earlier seroconversion. Due to the long follow-up period of this study, we were also able to observe that all seroconverted patients with both mild and severe symptoms still have detectable IgG levels after more than 75 days. Long et al reported that 97% of 37 patients with mild COVID-19 had decreased levels of IgG 2–3 months PSO [7], while another study with 34 hospitalized patients with COVID-19 presented increased levels of IgG until 5 weeks PSO, followed by consistent levels up to 7 weeks PSO [22]. Interestingly, our study shows that several patients with both mild and severe symptoms had increased in IgG concentrations over time. Wajnberg et al have shown that S-protein IgG levels could increase up to mean 82 days PSO in patients whose antibody levels were low initially [11]. The discrepancies between the results may at least partly be explained by different target antigens used in antibody detection in the different studies. Of note, all of the patients in the severe group in this study were male. In a study comparing serum-IgG responses between the sexes, no significant differences were found between women and men with severe COVID-19 [23], suggesting that the differences in proportions of males within the groups are unlikely to explain the concentration differences.

In this study, we found that almost 10% of patients with mild COVID-19 did not develop detectable anti-SARS-CoV-2 IgG in serum as evaluated by assays used in clinical practice. Previous studies have similarly failed to detect IgG antibodies in patients with mild disease [24-26], but due to short follow-up (less than 25–50 days) no conclusions regarding the proportion of patients who do not seroconvert have been made in the belief that antibody levels become detectable later in time. We show that despite 90 days or more PSO, not all patients develop detectable levels of IgG in these assays. However, using a virus neutralization assay, considered the golden standard of serology testing, all patients with undetectable IgG using commercial methods had NAbs. Furthermore, when using an in-house RBD ELISA, the two patients with highest levels of NAbs also had detectable anti-RBD-IgG. Several other studies have seen a correlation between the magnitudes of NAbs and IgG against viral S-protein epitopes [10, 27]. These results suggest that 100% of patients had in fact seroconverted, compared to 91% detected when using commercial methods.

While none of the patients with undetectable IgG using the commercial assays had anti-RBD-IgA, interestingly, the patients with detectable IgG also had detectable anti-RBD-IgA >75 days PSO, with no difference in IgA levels between mild and severe cases. Other studies have found serum-IgA after mild COVID-19 to be transient and undetectable after only one month post recovery [18, 28], again highlighting the sensitivity of our in-house RBD-ELISA. It is unsurprising that the validated SARS-CoV-2 antibody assays used in clinical practice miss a proportion of positive samples which other assays may detect. In these situations, the balance between assay sensitivity and specificity must be weighted to reduce the risk of false positives. This study demonstrates the risk of false negatives, in which low serum antibody levels may be detected with more sensitive assays.

We acknowledge several limitations in this study. The number and timing of serum sampling differed between patients; detectable IgG may have occurred before sampling, meaning that the exact timing of seroconversion is uncertain. Also, a larger proportion of patients in the mild group than the severe group have been available for late follow-up sampling.

Conclusions

We show that patients with severe COVID-19 both seroconvert earlier and develop higher concentrations of SARS-CoV-2-specific IgG than patients with mild symptoms. That not all COVID-19 patients develop detectable levels of IgG using two validated commercial methods, even over time, are vital for the interpretation of COVID-19 seroprevalence surveys and estimating the true prevalence in populations.

3 Aug 2020

PONE-D-20-21112

Serum-IgG responses to SARS-CoV-2 after mild and severe COVID-19 infection and analysis of IgG non-responders

PLOS ONE

Dear Dr. Gisslén,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Sep 17 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

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: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Stephen R. Walsh, MDCM

Academic Editor

PLOS ONE

Journal Requirements:

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

1. 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.PLOS ONE requires experimental methods to be described in enough detail to allow suitably skilled investigators to fully replicate and evaluate your study. See https://journals.plos.org/plosone/s/submission-guidelines#loc-materials-and-methods for more information. To comply with PLOS ONE submission guidelines, in your Methods section, please provide a more detailed description of how you produced SARS-CoV2 RBD.

In addition, if methods are well established, authors may cite articles where those protocols are described in detail, but the submission should include sufficient information to be understood independent of these references. Please revise your manuscript so that cited protocols are briefly but sufficiently described.

3. In your Methods section, please provide additional details regarding the 293F and Vero cell lines used in your study. Please include the source from which you obtained the cells, the catalog number if applicable, whether the cell line was verified, and if so, how it was verified. For more information on PLOS ONE's guidelines for research using cell lines, see https://journals.plos.org/plosone/s/submission-guidelines#loc-cell-lines.

4.We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

5.PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

Additional Editor Comments (if provided):

I would like to thank the authors for choosing to submit their manuscript for consideration by PLoS One. We have sought the opinion of three peer reviewers who are experts in the field and they have raised a number of issues. In their opinion, and mine, the manuscript could be considerably improved by addressing each of these issues. If the authors wish to address these comments and suggestions, please respond in a point-by-point fashion to each one.

[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

Reviewer #3: Yes

**********

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

Reviewer #1: Yes

Reviewer #2: I Don't Know

Reviewer #3: 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: No

Reviewer #3: 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: Yes

Reviewer #3: Yes

**********

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 manuscript by Marklund et al describes the analysis of serum IgG responses to SARS-CoV-2 in a cohort of patients with severe (n=15) or mild (n=32) COVID-19 disease. Patients with severe disease were found to seroconvert earlier and had higher IgG concentrations within the first 35 days post-symptom onset compared to patients with mild disease. It was noted that three individuals with mild disease never demonstrated detectable IgG responses using two commercially available serological assays. A virus neutralization assay was used to demonstrate that detectable serum neutralizing antibody activity could be detected in all three of these individuals, and two patients also had detectable IgG responses to RBD using an in-house ELISA assay. The authors conclude that commercially available assays may not have the sensitivity or specificity to detect positive IgG antibody responses in patients with mild disease, which will be important for interpreting results from seroprevalence studies. Overall the finding that commercial kits may have a rate of false negative results is not surprising, as even the authors state in the discussion. Additional data would be helpful in interpreting the antibody responses in these three individuals and comparisons with other individuals in the mild disease cohort that did develop positive serum reactivity. Below are more specific comments for consideration.

1: Materials and Methods. The description of the neutralizing antibody assay is confusing. Serial dilutions of serum are co-incubated with a standardized amount of virus for 2 hours at 37’C, and then the mixture added to Vero target cells, as per standard neutralizing antibody assay protocol. But then it describes adding ten-fold serial dilutions of virus. Is this for separate wells, or the positive control? What are the criteria for determining a serum dilution is “positive” for neutralizing activity? It mentions measuring CPE, but not clear if plaque reduction is being measured? Were normal human serum samples used as negative control to determine background of the assay? Clear details regarding this assay should be provided.

2: The authors state that all serum samples were tested using two commercially available serological assays, but data are only presented for one of the assays. Was there general concordance between the two assays or were discrepancies noted? This should be discussed, and data from the second assay platform should be presented.

3: It is difficult to interpret the data presented in Figure 2 without a similar comparison of individuals with mild disease that did develop IgG responses. It is not clear if there could be differences in Ct values over time or number of days with symptoms between the two categories of mild disease patients that may help characterize the development of the antibody response. Figure 2 would be more informative if the same parameters were graphed for patients with mild disease with detectable IgG responses.

4: Neutralizing antibody titers were only measured at a single late timepoint for the three individuals without detectable IgG. The titers measured are very low, so it would be helpful to have data from negative control serum samples to demonstrate background of the assay and validate that the responses measured are real. It would also be informative to understand the neutralizing antibody responses at earlier timepoints closer to PSO, as perhaps they would be more robust. Likewise, it would be helpful to understand the titers of neutralizing antibodies measured in mild patients that do develop detectable IgG titers (Table 2) for comparison.

5: Figure 1C should include labels to differentiate the two sets of scatter plots (i.e. Days 1-35 and >75 days).

6: Table 2: Annotations for “b” and “c” are mixed up in table legend.

Reviewer #2: Reviewer Summary:

The main claims of the paper are that anti-SARS-COV-2 IgG responses may be detected by a neutralizing antibody assay or sensitive ELISA in instances of mild infection where these responses were not detected in commercially available serology assays. Of interest are the findings that patients with severe symptoms tend to seroconvert faster than those with milder symptoms, and it seems that these antibody responses do not decrease over time (in the first 90 days or so), in contrast to the results of other studies with SARS-CoV-2 and other beta coronaviruses. These findings have implications for the interpretation of serology assays when used in seroprevalence surveys. This paper confirms the existence of the wide range in sensitivity of various serology assays (both commercial and lab-based), and raises an interesting point that serum neutralization may be a more sensitive readout than binding to the SARS-CoV-2 spike protein or the RBD of the spike protein. The manuscript is well organized and written clearly overall, but additional detail could be added to help non-specialists to understand the data and its interpretation. There are many opportunities available to add detail to the methods, results, and discussion sections to clarify to strengthen the interpretation of the data and to provide a resource for others looking to expound upon these results with other SARS-CoV-2 infection cohorts. Overall the manuscript has good potential for publication and these additional details would strengthen the manuscript so that it would be fit for publication.

Specific Areas for Improvement:

General: Timelines of antibody seroconversions are referred to frequently in the manuscript without context – e.g. “11 vs. 22 days”. Does this refer to post symptom onset? This should be clarified throughout.

Abstract: In the last sentence of the abstract, it is stated that “Our results…are vital for…. estimating the true infection prevalence in populations.” Is it true that antibody serology assays are being used to measure both seroprevalence AND infection prevalence? It is my understanding that RT-PCR is being used in most cases to measure infection prevalence.

Introduction: Literature on SARS-CoV-2 serology is constantly increasing and our thinking is changing by the day. It would be worthwhile to add more context to previous literature, including what has emerged in the past few weeks since submission if possible.

Methods: I could not find any supplemental information for this manuscript and thus no detailed protocols. The authors do refer to the Amanat et al paper for their SARS-CoV-2 ELISA and it appears that they made some modifications, but more details and a rationale for changing the conditions would be very helpful to the reader. It would be helpful to add more details for the RBD ELISA, such as incubation times, what positive controls were used, the readout (concentration determined by normalizing to a standard curve or OD) and how the threshold for a positive response was determined. There were also several modifications done compared to the published method by Amanat et al, and it would be helpful to provide a rationale for these differences (such as the increased coating concentration, differences in buffers/diluents, etc). It is also important to state whether the samples were also heat inactivated before the analysis of binding antibodies (either with the commercial assays or the RBD ELISA). Please state what positive controls were used in the RBD ELISA and how the threshold for a positive response was calculated. For the neutralization assay, which strain of SARS-CoV-2 was used, and what were the controls in the assay? How was positivity determine in the neutralization assay?

I do not have extensive statistical expertise to comment on the methods used, but I have raised some questions regarding Figure 3 (below) that could be added to the section on statistical analysis.

Results and Data: Overall the results section tells a nice story and it easy to follow from one figure/table to the next. There are some details that could be added that would help the reader better understand and interpret the results for figure 3.

Data could not be found in a repository. The description of data location is “All files are available from the Zenodo database (DOI:XXX)” so I was unable to find this data.

Table 2: the footnotes for b and c are switched.

Figure 3: What p values are denoted by the * and **? It would also be helpful to have a positivity cutoff line, similar to what you have in Figure 1 as well as an explanation of how the positivity threshold was calculated. What percentage of samples in the Mild and Severe infection categories would be positive? How is AUC calculated and can this be added to the methods/stats section? What is the readout of the RBD ELISA- concentration or OD? Is it possible that the low titers of IgA and IgM are due to competition with high levels of IgG for binding to the RBD?

Discussion: It is intriguing that the authors claim they did not observe a notable decline in antibody responses with the commercial assays. This should be discussed in further detail in the context of current literature on SARS-CoV-2. Would the same results be expected if a longitudinal analysis was conducted using the RBD ELISA or the neutralization assay? Also, on page 12 line 277, the authors refer to the increased sensitivity of their in-house ELISA. Is this more sensitive than the one published by Amanat et al? Did the various modifications make the ELISA more sensitive than the one published by Amanat et al? Or is the statement just meant to compare the RBD ELISA to the commercial serology assays? Clarifying and expanding on these points would further support your claims and be benefit to the field of SARS-CoV-2 serology.

Reviewer #3: The manuscript described a series of covid patients and revealed that patients with mild symptoms developed antibodies with lower titers than those with severe symptoms. The authors also identified three patients with undetectable antibody titers by commercial IgG assays. The study has certain merits but with some issues:

1. In abstract, there is no methods section.

2. For the RBD in-house antibody assay, how to decide the cut-off should be stated in the methods.

3. For neutralizing antibody assay, how to decide the cut-off should be clarified in the methods.

4. In Line 167, a p-value is needed.

5. Missing data from patient2 in Table 3 and Table 3 should be presented as supplemental table.

6. In Line 273-282, the argument about anti-RBD-IgA may not be true. One sample in the M group has a very low IgA (the dot is touching the 0.00 base line in Figure 3). The high sensitivity of in-house assay may come with high false positivity. IgA covid tests are known for their high false positivity.

**********

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

Reviewer #3: 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.

1 Sep 2020

Page and line numbers refer to the Revised manuscript (clean copy).

Journal requirements

1. To comply with PLOS ONE submission guidelines, in your Methods section, please provide a more detailed description of how you produced SARS-CoV2 RBD.

A detailed description of the protein expression has been added the methods section (page 4, lines 97-110).

2. In your Methods section, please provide additional details regarding the 293F and Vero cell lines used in your study. Please include the source from which you obtained the cells, the catalog number if applicable, whether the cell line was verified, and if so, how it was verified.

These details have now been added (page 5, line 100 and page 7, line 150). Both cell lines were bought commercially and have not been further verified.

3. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data.

The DOI is https://doi.org/10.5281/zenodo.3934336, which has been added to the relevant section in Editorial Manager.

4. PLOS requires an ORCID iD for the corresponding author in Editorial Manager.

Magnus Gisslén (the corresponding author) ORCID iD: 0000-0002-2357-1020.

We have not been able to enter it into Editorial Manager due to multiple accounts, but a merge is in process.

Reviewer #1

1: Materials and Methods. The description of the neutralizing antibody assay is confusing. Serial dilutions of serum are co-incubated with a standardized amount of virus for 2 hours at 37’C, and then the mixture added to Vero target cells, as per standard neutralizing antibody assay protocol. But then it describes adding ten-fold serial dilutions of virus. Is this for separate wells, or the positive control? What are the criteria for determining a serum dilution is “positive” for neutralizing activity? It mentions measuring CPE, but not clear if plaque reduction is being measured? Were normal human serum samples used as negative control to determine background of the assay? Clear details regarding this assay should be provided.

The method description for the neutralizing antibody assay has now been adjusted in accordance with the reviewer’s comments (Page 6, lines 146-160).

2: The authors state that all serum samples were tested using two commercially available serological assays, but data are only presented for one of the assays. Was there general concordance between the two assays or were discrepancies noted? This should be discussed, and data from the second assay platform should be presented.

There was general concordance between the two assays, with a discrepancy seen in only two patients. This is now presented clearly in the results section (page 8, lines 188-193). Since the Architect assay is qualitative and not quantitative, we see no additional value in presenting the results from that assay separately, except for the two patients mentioned above.

3: It is difficult to interpret the data presented in Figure 2 without a similar comparison of individuals with mild disease that did develop IgG responses. It is not clear if there could be differences in Ct values over time or number of days with symptoms between the two categories of mild disease patients that may help characterize the development of the antibody response. Figure 2 would be more informative if the same parameters were graphed for patients with mild disease with detectable IgG responses.

It is unfortunately not feasible to graph the same parameters for the 29 patients with mild disease and detectable IgG responses. Instead, we have added details of these patients in the results section (page 10, lines 218-222). We state that the lowest observed Ct-values in the non-IgG patients were of a similar range as in the patients who seroconverted (15.9–24.9 vs. 11.9–37.2), and it has now been clarified that this is referring to the seroconverted patients with mild disease. Further, durations of symptoms in the non-IgG patients and in the majority of the patients with mild symptoms who seroconverted has been added.

4: Neutralizing antibody titers were only measured at a single late timepoint for the three individuals without detectable IgG. The titers measured are very low, so it would be helpful to have data from negative control serum samples to demonstrate background of the assay and validate that the responses measured are real. It would also be informative to understand the neutralizing antibody responses at earlier timepoints closer to PSO, as perhaps they would be more robust. Likewise, it would be helpful to understand the titers of neutralizing antibodies measured in mild patients that do develop detectable IgG titers (Table 2) for comparison.

Data on the NAb titers of the negative controls in this assay has been added to the methods section (page 7, lines 158-160). We agree that it would be informative to investigate the NAb kinetics closer. However, NAb analyses are work-intense and expensive and we do unfortunately not have the resources to do this right now, but funds are being applied for and such an in-depth study is planned.

5: Figure 1C should include labels to differentiate the two sets of scatter plots (i.e. Days 1-35 and >75 days).

The annotation of days 1-35 and >75 days was mistakenly omitted; this has now been rectified.

6: Table 2: Annotations for “b” and “c” are mixed up in table legend.

This is now corrected (page 11, line 246-247).

Reviewer #2

1: General: Timelines of antibody seroconversions are referred to frequently in the manuscript without context – e.g. “11 vs. 22 days”. Does this refer to post symptom onset? This should be clarified throughout.

This does refer to post symptom onset, which has now been clarified throughout.

2: Abstract: In the last sentence of the abstract, it is stated that “Our results…are vital for…. estimating the true infection prevalence in populations.” Is it true that antibody serology assays are being used to measure both seroprevalence AND infection prevalence? It is my understanding that RT-PCR is being used in most cases to measure infection prevalence.

Reference to infection prevalence has been removed from the abstract (page 2, line 48).

3: Introduction: Literature on SARS-CoV-2 serology is constantly increasing and our thinking is changing by the day. It would be worthwhile to add more context to previous literature, including what has emerged in the past few weeks since submission if possible.

We have added a section in the introduction with more recent references reflecting the recent conflicting findings in this field (page 3, lines 69-73).

4: Methods: I could not find any supplemental information for this manuscript and thus no detailed protocols. The authors do refer to the Amanat et al paper for their SARS-CoV-2 ELISA and it appears that they made some modifications, but more details and a rationale for changing the conditions would be very helpful to the reader. It would be helpful to add more details for the RBD ELISA, such as incubation times, what positive controls were used, the readout (concentration determined by normalizing to a standard curve or OD) and how the threshold for a positive response was determined. There were also several modifications done compared to the published method by Amanat et al, and it would be helpful to provide a rationale for these differences (such as the increased coating concentration, differences in buffers/diluents, etc). It is also important to state whether the samples were also heat inactivated before the analysis of binding antibodies (either with the commercial assays or the RBD ELISA). Please state what positive controls were used in the RBD ELISA and how the threshold for a positive response was calculated. For the neutralization assay, which strain of SARS-CoV-2 was used, and what were the controls in the assay? How was positivity determine in the neutralization assay?

We have now updated the method section to give extensive details regarding the RBD ELISA (Page 5, lines 121-135). AUC was calculated from the ELISA curves using GraphPad Prism as in Amanant et al and many other publications.

Most of the differences are in the secondary antibodies used: the secondary anti-human-IgG used by Amanant is no longer produced by ThermoFisher and anti-IgA and IgM were not used by Amanant et al. We have used secondary antibodies from the same supplier (SouthernBiotech). Sera were not heat inactivated, but we did run comparisons and found no differences in titers induced by heat inactivation.

For the neutralization assay, all sera with antibody titers ≥4 (1/4 dilution) were considered neutralizing, demonstrating neutralization in both 1/2 and 1/4 dilutions, confirming the presence of antibodies with capacity to block infection. Sera (n=17) from patients and blood donors from before December 2019 were used as a negative control and no neutralization was detected in any of these samples (page 7, lines 158-160)..

5: Data could not be found in a repository. The description of data location is “All files are available from the Zenodo database (DOI:XXX)” so I was unable to find this data.

See Journal Requirements #3.

6: Table 2: the footnotes for b and c are switched.

This is now corrected.

7: Figure 3:

What p values are denoted by the * and **?

It would also be helpful to have a positivity cutoff line, similar to what you have in Figure 1 as well as an explanation of how the positivity threshold was calculated. What percentage of samples in the Mild and Severe infection categories would be positive? How is AUC calculated and can this be added to the methods/stats section? What is the readout of the RBD ELISA- concentration or OD? Is it possible that the low titers of IgA and IgM are due to competition with high levels of IgG for binding to the RBD?

P-value summaries are now added to the figure legends (page 9, line 211 and page 10, line 232-233).

We have now added a cutoff line in Fig 3 at 2SD over the negative control average AUC.

The readout of RBD ELISA is an absorbance read at several serum dilutions. From the dilution curve, the AUC (area under the curve) was calculated using GraphPad Prism.

It is possible that competition between IgG and IgA/IgM would impact the titer, but we think it is unlikely. If titers were equal and the only difference would be the amount of antibodies, then we should see binding of IgA at a subsaturating IgG dilution, which we do not. Further, at a saturating IgG dilution we see increased IgA binding with decreasing dilution, indicating available binding sites on RBD.

Discussion: It is intriguing that the authors claim they did not observe a notable decline in antibody responses with the commercial assays. This should be discussed in further detail in the context of current literature on SARS-CoV-2. Would the same results be expected if a longitudinal analysis was conducted using the RBD ELISA or the neutralization assay? Also, on page 12 line 277, the authors refer to the increased sensitivity of their in-house ELISA. Is this more sensitive than the one published by Amanat et al? Did the various modifications make the ELISA more sensitive than the one published by Amanat et al? Or is the statement just meant to compare the RBD ELISA to the commercial serology assays?

That is an interesting and important question. There are by now a few papers suggesting sustained levels of IgG antibodies few weeks after infection. We would expect that RBD ELISA and neutralization titers would follow the general trend observed with Architect and iFlash, however we have no data supporting this hypothesis, which will be the topic of future studies. This is now commented upon in the discussion (page 13, lines 308-311).

Reviewer #3

1: In abstract, there is no methods section.

A sentence regarding the method has been added to the abstract (page 2, line 31-33).

2: For the RBD in-house antibody assay, how to decide the cut-off should be stated in the methods.

We have amended the methods and figure 3 legend to indicate how the cut-off was calculated (page 6, line 135).

3: For neutralizing antibody assay, how to decide the cut-off should be clarified in the methods.

We have amended the methods section to indicate how the cut-off was established (page 7, lines 156-160).

4: In Line 167, a p-value is needed.

This has been added (page 9, line 201).

5: Missing data from patient2 in Table 3 and Table 3 should be presented as supplemental table.

As explained (page 10, line 222-224), only two of the three non-seroconverting patients were available for additional testing, thus we have no additional data from patient 2.

6: In Line 273-282, the argument about anti-RBD-IgA may not be true. One sample in the M group has a very low IgA (the dot is touching the 0.00 base line in Figure 3). The high sensitivity of in-house assay may come with high false positivity. IgA covid tests are known for their high false positivity.

The reviewer is right that IgA COVID tests are known for false positive. Of note, our 10 negative controls have all low IgA levels to RBD. It is also correct that several patients in the mild group are negative for IgA. With the cut-off of average AUC of negative +2SD still >50% of patients with mild disease and severe disease were over the cut-off. However, as a group, the IgG- patients had a statistically lower IgA reactivity compared to other groups.

In addition to these responses to reviewer comments, we have discovered an error in the data analysis. One sample from a patient with mild disease had been incorrectly entered as >75 days, thus there is now one person less in the mild group in figures 1C and 3. This has had no impact on the statistical significances.

Submitted filename: Response to reviewers.docx

Click here for additional data file.

5 Oct 2020

PONE-D-20-21112R1

Serum-IgG responses to SARS-CoV-2 after mild and severe COVID-19 infection and analysis of IgG non-responders

PLOS ONE

Dear Dr. Gisslen,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Nov 19 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

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: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Stephen R. Walsh, MDCM

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

Thank you for addressing the reviewers' comments. We believe the manuscript has been improved. Please note that Figure 3 appears to be mislabeled.

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

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: (No Response)

Reviewer #2: All comments have been addressed

Reviewer #3: 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

Reviewer #2: (No Response)

Reviewer #3: Yes

**********

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

Reviewer #1: Yes

Reviewer #2: (No Response)

Reviewer #3: 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

Reviewer #2: (No Response)

Reviewer #3: 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

Reviewer #2: (No Response)

Reviewer #3: 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: Figure 3: The figure legends for IgA and IgM do not match the y-axis labels on the graphs. Are the y-axis labels incorrect or are the graphs between Figures 3B and 3C switched?

Reviewer #2: (No Response)

Reviewer #3: The response is appropriate and the revision has addressed all my concerns . Thank you for all the works!

**********

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

Reviewer #2: No

Reviewer #3: 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.

7 Oct 2020

Dear Editor,

We are pleased that our revision of the manuscript“Serum IgG responses to SARS-CoV-2 after mild and severe COVID-19 infection and analysis of IgG non-responders” was to the reviewers’ satisfaction.

Regarding Reviewer #1, comment #6: The y-axis labels in Figure 3B+C were indeed incorrect and have now been remedied.

9 Oct 2020

Serum-IgG responses to SARS-CoV-2 after mild and severe COVID-19 infection and analysis of IgG non-responders

PONE-D-20-21112R2

Dear Dr. Gisslen,

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,

Stephen R. Walsh, MDCM

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

13 Oct 2020

PONE-D-20-21112R2

Serum-IgG responses to SARS-CoV-2 after mild and severe COVID-19 infection and analysis of IgG non-responders

Dear Dr. Gisslen:

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

Dr. Stephen R. Walsh

Academic Editor

PLOS ONE