Differential Gene Expression in the Upper Respiratory Tract following Acute COVID-19 Infection in Ambulatory Patients That Develop [LC], 2024, Biondi+

[SNT Gatchaman]

Differential Gene Expression in the Upper Respiratory Tract following Acute COVID-19 Infection in Ambulatory Patients That Develop Long COVID
Biondi, Mia J.; Addo, Mary; Zahoor, Muhammad Atif; Salvant, Elsa; Yip, Paul; Barber, Bethany; Smookler, David; Wasif, Sumaiyah; Gaete, Kayla; Kandel, Christopher; Feld, Jordan J.; Tsui, Hubert; Kozak, Robert A.

BACKGROUND
Post-acute sequelae of COVID-19, or long COVID, is a condition characterized by persistent COVID-19 symptoms. As long COVID is defined by clinical criteria after an elapsed period, an opportunity for early intervention may aid in future prophylactic approaches; however, at present, the pathobiological mechanisms are multifactorial. By analyzing early virally infected upper respiratory tract tissue prior to eventual clinical diagnosis, it may be possible to identify biomarkers of altered immune response to facilitate future studies and interventions.

METHODS
This is a sub-group analysis of samples collected from those with confirmed COVID-19. RNA extraction from nasopharyngeal/mid-turbinate samples, sequencing, and bioinformatic analysis were performed to analyze long COVID and non-long COVID cohorts at day 14 post infection. Differences in mean viral load at various timepoints were analyzed as well as serological data.

RESULTS
We identified 26 upregulated genes in patients experiencing long COVID. Dysregulated pathways including complement and fibrinolysis pathways and IL-7 upregulation. Additionally, genes involved in neurotransmission were dysregulated, and the long COVID group had a significantly higher viral load and slower viral clearance.

CONCLUSIONS
Uncovering early gene pathway abnormalities associated with eventual long COVID diagnosis may aid in early identification. We show that, post acute infection, in situ pathogenic deviations in viral response are associated with patients destined to meet consensus long COVID diagnosis that is entirely dependent on clinical factors. These results identify an important biological temporal window in the natural history of COVID-19 infection and long COVID pathogenesis amenable to testing from standard-of-care upper respiratory tract specimens.

Link | PDF (Pathogens) [Open Access]

 

[SNT Gatchaman]

Here, we investigated the host gene signatures present from the upper respiratory mucosa at the end of the acute phase of infection between individuals who subsequently developed LC, as currently defined, and those who did not. By associating the host transcriptional and immunological response in the upper respiratory tract with the subsequent development of long COVID, as determined by an established scoring system [12], we highlighted unique gene expression patterns in these individuals.

[12] is Development of a Definition of Postacute Sequelae of SARS-CoV-2 Infection (2023, JAMA)

Viral RNA persistence in the nasal swabs from LC and non-LC patients was evaluated. In the non-LC cohort, consisting of 18 patients, 81% of patients were RNA-positive at enrollment, 33% at day 7, and 6% at day 14. In the LC group of 9 patients, 75% of patients were RNA-positive at enrollment, 77% at day 7, and 22% at day 14. Lower viral loads were observed in the non-LC group throughout the course of the study.

The early host responses of LC and non-LC patients were compared from samples collected at enrollment and day 14 to identify uniquely dysregulated genes during the acute phase of infection. To accurately identify which individuals were classified as long COVID patients, differential gene expression was analyzed for cohorts defined using two different LC definitions: the World Health Organization (WHO) LC definition and criteria and the LC scoring criteria developed by Thaweethai et al. [12].

No significantly differentially expressed genes were identified at the enrollment time point. Using the Thaweethai classification system, gene expression patterns were compared from samples at the end of acute illness and twenty-six genes were found to be upregulated. When patient samples were analyzed based on the WHO criteria, fewer differentially expressed genes were identified), suggesting that the Thaweethai classification produced a larger analysis background.

 

[SNT Gatchaman]

GeneCards —

CE2C
KLK4
KRT79
LINC00664
RPS6KA1
YWHAE
NBPF26_1
LIMS4
APCDD1L-DT
SMDT1
ITGA2B
SPPL2C_1
KCNH5
LACRT
LOH12CR2
CPB2
LOC107984156
RPS25_1
SNAP25
SDR9C7
C8G
LINC02389
СРА2
CST4
TMEM78
LCE2B

 

[SNT Gatchaman]

Common enriched genes in the LC cohort between both scoring methods included YWHAE, KLK4, LCE2C, KRT79, and RPS6KA1. [...] YWHAE encodes the 14-3-3 epsilon regulator protein and is a target of SARS-CoV-2. The gene was enriched for the negative regulation of dendritic cell apoptotic processes and the negative regulation of synaptic vesicle exocytosis.

Given the upregulation of the gene, this may relate to LC patients being more susceptible to secondary bacterial infections post COVID [23]. We noted that 77% (n = 7/9) of LC patients experienced brain fog and were therefore compared to the 18 non-LC patients in a separate analysis in which the top thirty significant DEGs were identified. This was done to determine if brain fog is indicative of differences in gene expression. Although most patients in the LC group experienced this symptom, it may not be a key factor as no significant changes were found here or through functional analysis.

 

[SNT Gatchaman]

Pathway analysis was performed on patient samples classified using the criteria developed by Thaweethai et al. [12]. Enriched pathways identified with EnrichR (p < 0.05) included the classical complement pathway and the fibrinolysis pathway.

Another upregulated gene, CPB2, plays a role in inhibiting fibrinolysis, and CPB2 and C8G are involved in the regulation of complement cascade.

In our cohort, we noted an increased expression of the enzyme carboxypeptidase U (CPB2), which may account for the presence of dense clots seen in LC patients as the enzyme is a potent inhibitor of fibrinolysis [26]. Additionally, CPB2 plays a role in the activation of the complement pathway. The upregulation of the complement proteins, CPB2 and C8G, indicates a thrombo-inflammatory response and the over-activation of the immune response.

[26] is Carboxypeptidase U CPU, carboxypeptidase B2, activated thrombin‐activatable fibrinolysis inhibitor inhibition stimulates the fibrinolytic rate in different in vitro models (2018, Journal of Thrombosis and Haemostasis)

 

[SNT Gatchaman]

There was a clear trend in viral load, indicating that LC patients have a greater mean viral load over time. In addition, time to viral clearance was significantly prolonged in LC patients. [...] none of the individuals in the LC group had detectable SARS-CoV-2 in the samples taken during the convalescent phase.

Serological data from LC and non-LC patients displayed a similar level of anti-N seropositivity at the day > 90 timepoint. No significant trends were found for the antiS serological data, and mean concentrations in both LC and non-LC groups remained relatively equal at both timepoints. Overall, this suggests that serology may be of little additional benefit as a predictor of long COVID. These data do not account for serological avidity, which may provide better insight into the differences in immune responses between LC and non-LC patients.

 

[Hutan]

A total of 45 patients were selected for analysis, among whom we identified 34 non-LC and 11 LC patients. All individuals were positive for SARS-CoV-2 by RT-PCR in their samples collected at the initial time point. From this, 18 non-LC and 9 LC participants were included for RNAseq analysis due to the availability of residual clinical material.

A hope-inducing study, but it was small and the groups weren't well matched
e.g. LC 77% female; controls 50% female

Data on symptoms were collected at day 90 or later:
6/9 of the LC group reported post-exertional malaise (as did 2/18 of the controls)
8/9 of the LC group reported alterations in smell and taste (as did 4/18 of the controls)

 

[Hutan]

This is Figure 2a



Genes are the rows, columns are different participants, with the Long Covid participants identified by the red bar at the top (versus black of the controls).

What I want to show you is the variation. Take the gene at the top, NBPF26, really only one participant was expressing that gene differently. YWHAE is mentioned in the text, but at most 6 participants have a higher expression of that gene, and 2 of them are controls.

Advocates often say 'look at all the abnormal results, there is so much evidence of pathology' and might point to a study like this. And, in there somewhere, there may well be the clue, something that really is different. But, there's a lot of noise.

They have grouped the genes together according to the similarity of their expression across the participants (on the left). I would have preferred to see the patients grouped together according to their gene expression.

 

[Hutan]

There was a clear trend in viral load, indicating that LC patients have a greater mean viral load over time.

I don't think we can assume that this tells us anything about ME/CFS-like Long Covid. With 8/9 of the Long Covid participants having issues with taste and smell, it might just be indicating that high viral loads in nasal swabs is associated with taste and smell disturbances.

We need more studies like this, but bigger, and with "Long Covid" criteria broken down into subgroups that make sense. We already know that alterations to taste and smell don't seem to have much to do with the ME/CFS-like pattern of symptoms. Persisting cough doesn't seem to either, it probably can be part of a symptom collection to do with easily identifiable lung tissue damage. If controls are reporting PEM, then it is possible that people haven't understood what PEM is.

 

[Eleanor]

In the cohort description it says "Individuals recruited were symptomatic COVID-19 patients aged 18 or older, who had at least one risk factor for severe disease."