PREreview of Quantitative multi-organ proteomics of fatal COVID-19 uncovers tissue-specific effects beyond inflammation

In its current state, the assessment of differing proteomes between normal individuals and those experiencing different pathologies has been limited to analyzing the global changes. In this study, the authors utilized mass spectrometry-based proteomics to determine the abundance of protein levels at specific organ sites between COVID-19 samples and a control cohort. They determined that representative markers of panels for coagulation, platelets, plasma and red blood cells strongly influenced the assessed protein abundance in a number of other tissues, contributing to at least 50% of the intensity of the signal. The paper collectively refers to this as circulatory-mediated proteome masking. These markers became used to distinguish organ-specific effects from systemic effects. 

With these adjustments incorporated into their analysis, the researchers identified a number of findings. In the lungs, they identified an upregulation of nicotinamide N-methyltransferase (NNMT), fibroblast growth factor receptor substrate 3 (FRS3), and the negative regulator of collagen production Reticulocalbin-3 (RCN3), among many others. Additionally, they noted significant differences between the lung proteomes of COVID-19 patients and patients of other lung pathologies, with a minimum fold-change 1.5 at a q-value of 0.05. The paper also analyzed the phosphorylation pathways and identified significant differences between the phosphoproteomes of their control groups as compared with COVID-19 patients. In the lymph nodes, they noted an upregulation of CDK2, RecQ-like DNA helicase BLM, RAD9A, ORC ¾, Nucleoporin Nup37, and other proteins associated with cell cycle regulation. The brain indicated upregulation of a number of neurotransmitters, including HOMER1, RYR2, GNAL and many others. A comprehensive list of all upregulations the researchers found throughout each organ-system may be found in the paper itself. 

The limitations of this paper arise in various aspects of their methods and systems that I will discuss below. 

1. The paper should be more clear in how it defines ‘organ-specific’ changes. The proteomes of many organ systems are influenced by hormones and neurotransmitters produced in the brain and many glands. If the levels of these proteins are impacted by COVID-19, it will impact the levels of proteins in other organ systems. It is vital to be clear as to whether these changes are accounted for when identifying organ-specific changes. The nature of this is important in understanding the accuracy of their data as ‘organ-specific.’ A recent paper chronicling the development of a comprehensive proteome (PMID: 32916130), discussed an underdetection of membrane proteins and secreted proteins. If hormone levels are being underdetected in their MS-based model, that may influence their results. It may be important to address this.

2. The paper is detailed in its discussion of differences in the phosphoproteome they found between COVID-19 patients and their control group. However, I believe that the authors should also clarify whether their MS-based model can differentiate between active and inactive forms of different proteins beyond phosphorylation. Additionally, they could note whether their model can identify proteins that are bound to other proteins. For example, thiiodothyronine (T3 hormone) can exist as a normal version which is active and a reverse version which is inactive. Additionally, it may be bound by a protein known as thyroxine-binding globulin. A clear picture of protein levels that account for these factors is important to create an accurate understanding of which proteins are upregulated.

3. The paper provides a strong discussion comparing the proteomes of COVID-19 patients and patients with other lung conditions. However, they noted that these samples were derived from autopsies at the University Medical Center Augsburg from April to May 2020 (PMID: 32437497). They followed guidelines set by previous research that developed these samples (PMID: 32437497), which stated that autopsies happen within 24 hours of the patient’s time of death. These tissues were preserved using a formalin-fixed, paraffin-embedded (FFPE) method. Meanwhile, the researchers obtained their control samples from currently living healthy donors and patients with lung conditions. However, it is important to note that protein levels can shift in a few hours due to changes in external stimulus. Simple actions such as breathing can alter the proteome of the body. The samples taken from the post-mortem COVID-19 patients have an absence of these external stimuli, which is inconsistent with the control cohort. This could be another possible explanation for certain alterations in the proteome that may not necessarily be attributable to COVID-19. It is important for the researchers to address this possibility within their paper. 

Rating:

Reliable. The main study claims are generally justified by the data and analytic methods used. There are some reservations that may need to be addressed. The study produced results and conclusions that are novel and provide new insight into the proteomics involved in COVID-19. The study’s main claims should be considered actionable with limitations.

Competing interests

The author declares that they have no competing interests.