Regarding “Imaging Tracks Fibrinogen Activity in MS Brain Inflammation” (Jan. 17; http://bit.ly/1137hCu) that features our study on the role of fibrin in neuroinflammation, we write to respond to some questions and concerns raised by one of the commentators. [For the abstract of the original study, see http://1.usa.gov/Wt40YL.]
In response to the comment that anticoagulation “would not be expected to abrogate binding of leaked fibrinogen to CD11b/CD18 on microglia,” we must point out that anticoagulant treatment is well known to inhibit fibrin formation, and we detected no fibrin in the spinal cord after anticoagulant treatment (Figure 5b). This observation is consistent with earlier studies by our team (Adams RA, et al. J Exp Med 2007; 204:571-582) and others (Paul J, et al. J Exp Med 2007;204:1999-2008). Therefore, anticoagulation abrogates fibrin and evidently CD11b activation.
The commentator noted that “limiting binding of fibrinogen to CD11b/CD18 on microglia by genetic manipulation decreased microglial cluster formation, but the data about its effect on the course of EAE [experimental autoimmune encephalitis] were not provided.” Our paper referenced the 2007 paper by RA Adams in the Journal of Experimental Medicine, describing EAE suppression in mice with genetic manipulation of fibrinogen that inhibits its binding to CD11b, and stated that “studies in the Fibγ390-396A mice ameliorate responses of both cell populations [our study] and clinical EAE severity.” Figure 1c shows data on the statistical significance of pre-onset EAE; one-way analysis of variance (ANOVA) of pre-onset and all other EAE time-points showed significant cluster formation. We detected no clusters in the control group, which is noted in the graph as undetectable.
Finally, in response to the comment that our paper lacked sufficient experimental detail, we must note here that we described our experimental protocols in as much detail as possible within the journal's 1500-word limit. The commentator also noted omission of the NINDS recommendation for preclinical animal studies, and the potential bias of microscopy. We did not cite the NINDS recommendation as it was published after our paper was submitted on Aug 20, 2012. However, we strongly agree with those guidelines and are happy to describe below our criteria for randomization, blinding, and data handling.
We performed correlated histological analyses in the exact same spinal cord fragments imaged in vivo by immunohistochemistry and electron microscopy (EM) (see Supplemental Figure 1). Three independent observers, blinded to the mouse genotypes and treatment, scored the EAE experiments in our laboratory. Correlated histology was performed in all mice that were imaged in vivo and the only criteria for exclusion were accidental injury during surgery or EAE severity to moribund state that required sacrificing the mice as per the Institutional Animal Care and Use Committee protocol. Immunohistochemistry and EM were performed in two different research facilities at Gladstone and the University of California, San Diego, respectively.
To further analyze the EM data, images were also evaluated at the University of Vienna. For randomization, immunohistochemistry was performed on random sections from the exact same spinal cord fragments that were imaged in vivo and quantification was performed blinded by two individuals. Cell clusters were identified by increased green fluorescent protein signal density over neighboring (non-clustered) regions always drawn within the same images. Fibrinogen co-localization was quantified on randomized images in an automated fashion, using statistical software as described in the Methods section of the study. Three individuals performed fibrinogen injections, and quantification was done in a blinded manner.
We appreciate that our paper was featured in Neurology Today and the opportunity to describe our robust experimental design, interpretations, and conclusions.
Katerina Akassoglou, PhD
Senior Investigator, Gladstone Institutes
Professor of Neurology
University of California, San Francisco
San Francisco, CA