Gitelman and Bartter syndromes (GS/BS), two rare genetic tubulopathies characterized by hypokalemia and metabolic alkalosis, high angiotensin II levels, yet exhibit normotension or hypotension and are associated with protection from cardiovascular and renal remodeling, are reported to have increased ACE2 levels, which could make GS/BS patients more susceptible to SARS-CoV-2 infection [1]. We surveyed our cohort of GS/BS patients to ascertain their level of coronavirus disease 2019 (COVID-19) infections, finding virtually no cases of COVID-19 among the 128 GS/BS patients interviewed, which was statistically significant vs. the adjusted prevalence of COVID-19 in the northern Italian population [1,2].

Wu et al. [3] showed that, at the same level of ACE2 expression, there was an increased binding of the regional binding domain of SARS-CoV-2 to blood group A compared to group O, providing a mechanism whereby blood group A individuals are more susceptible to SARS-CoV-2 infection. This finding and our surveys in GS/BS patients suggesting protection against SARS-CoV-2 infection [1,2], led us to examine the blood group distribution of our GS/BS patients to help explain this protection.

Eighty GS/BS patients (GS, 72 and BS, 8)—53 males and 27 females, aged 16 to 72 years—genetically and biochemically characterized, were considered. Patients’ blood groups were obtained from patients’ health records, and clinical information regarding their COVID-19 status was collected through telephone interviews [1,2]. Eight patients (GS, 6 and BS 2; five males and three females) were confirmed to have had COVID-19: four were asymptomatic and four had very mild symptoms. All participants were informed about the nature of the study, gave their consent, and were not exposed to any risk by the irreversible anonymization of data.

The distribution of GS/BS blood groups was compared with Italian population data using a chi-square goodness-of-fit test. The sum of standardized residuals was used to identify categories contributing most to any observed deviation. A two-tailed significance threshold of p < 0.05 was applied. Statistical analyses were performed using R version 4.5.0 (R Foundation for Statistical Computing).

Blood group distribution in our 80 GS/BS patients (group O, 51; group A, 21; group AB, 4; and group B, 4) was significantly different compared with the Italian population (p < 0.001) (Table 1).

We tried to identify which of the O, A, B, and AB groups drove the overall significant results, where any residual over 2 or under –2 is noteworthy. The standardized residuals of GS/BS’s blood groups showed that group O (3.36) was overrepresented and group A (–2.39) underrepresented, while neither group B (–1.62) nor group AB (0) contributed to the distribution difference (Table 1). Thus, although the GS/BS blood group distribution differs from that of the Italian population, the fact that GS/BS have a reduced proportion—rather than an absence—of group A makes it unlikely that blood group distribution alone can account for their resistance to COVID-19 [2].

Many studies focused on the role of factors either linked to SARS-CoV-2, such as the S protein, or related to the host cell, such as ACE2. The blood group distribution of patients with SARS-CoV-2 infection was another ground of study, and blood group A was identified as providing more susceptibility to infection [3].

During the pandemic, our GS/BS patients—despite having increased ACE2 expression, which potentially confers greater susceptibility to infection [1]—were surprisingly protected from SARS-CoV-2 infection, as only a very small number tested positive for COVID-19 and were either asymptomatic or had mild symptoms. This virtual absence of COVID-19 was statistically significant compared with the adjusted prevalence in the Italian population [2].

This study shows that the blood group distribution in GS/BS patients is statistically different from that of the Italian population. However, based on the standardized residuals, GS/BS patients have an excess of group O and a deficit of group A. As group A appears to confer increased susceptibility, it is notable that, despite the presence of 21 group A GS/BS patients, only three were infected—one asymptomatic and two with very mild symptoms. Conversely, the reduced proportion of other blood groups compared to group O was not significant enough to suggest a protective effect of group O, as none of those patients were infected. Our findings therefore suggest that the protection of GS/BS patients from COVID-19 is unlikely to be explained by their blood group distribution.

We further characterized the basis for GS/BS patients’ resistance to COVID-19, speculating that the protection might arise from their metabolic alkalosis, which could induce an intracellular environment unfavorable to SARS-CoV-2 infection [4,5]. The entry and replication–transcription mechanism of SARS-CoV-2 is based, in fact, on binding to glycosylated ACE2, whose glycosylation depends on the acidic pH environment of the trans-Golgi network and post-Golgi pathways in the endosomes, as well as on proteases—including cathepsin L—whose activity also depends on acidic endosomal pH. The increased intracellular organelle pH induced by metabolic alkalosis may interfere with ACE2 glycosylation, S protein-mediated binding, SARS-CoV-2 entry, and the replication–transcription mechanism [4,5]. We found that GS/BS patients had significantly higher non-glycosylated ACE2 levels and lower cathepsin L activity than healthy subjects, and that cathepsin L activity was inversely correlated with blood bicarbonate, providing a mechanistic explanation for the near absence of COVID-19 in GS/BS patients [4,5]. Interestingly, although children express more ACE2 than adults, they generally experience milder COVID-19 symptoms, raising the question of whether ACE2 glycosylation may change with age [6]. Further evidence of the importance of glycosylation in COVID-19 is that the in vivo properties of a soluble ACE2–immunoglobulin 1 decoy are heavily influenced by how its protein is glycosylated. Moreover, sialylation of the decoy’s native glycans increased its half-life, and this optimized decoy was therapeutically efficacious [7]. In addition, glycosylation at the N343 site of the SARS-CoV-2 S protein—a potential hot spot for mutations in the glycan shield—plays a significant role in the structure and stability of the S protein [8].

In summary, our earlier surveys of GS/BS patients found virtually no cases of COVID-19. The current study ruled out differences in blood group distribution among GS/BS patients as a protective factor, while their protection may instead arise from metabolic alkalosis–induced increases in intracellular pH, which alter ACE2 glycosylation and also interfere with cellular viral processing. The altered ACE2 glycosylation observed as a protective factor in GS/BS patients, together with the key role of glycosylation in SARS-CoV-2 infection, supports the development of glycosylation-based drug candidates.