Reply

Bulte and Frank raise interesting questions about the positioning of USPIO in diseases in which macrophages are involved.

First, the cells that are displayed in the articles are clearly not endothelial cells, because they are located behind the basal membrane, which is the anatomic frontier used to differentiate endothelial cells and so-called pericytes from nonendothelial cells. They also have the typical form of macrophages; ie, they are ameboid, their cytoplasms are full of vesicles, and they have a large nucleus. Furthermore, the histologic examination with ED I antibody macrophage markers overlaid the sites of the abnormal signal on images.

Second, the images were performed 24 hours after IV injection, which is five times that of the blood half-life of the USPIO in rat. We agree with Bulte and Frank that the iron particles may have crossed the BBB during this window, but no particles were detectable in the extracellular space at the 24-hour electron microscopy examination. As we pointed out in the articles, an active cellular uptake of USPIO is necessary for the particles still to be present in the parenchyma after five blood half-lifes. We also agree with Bulte and Frank that endothelial cells capture USPIO, although CNS endothelial cells have less capabilities of phagocytosis than other endothelial cells. It, however, is not the predominant phenomenon in models in which macrophages are involved (1–5). Macrophages are known to have the highest USPIO uptake, far beyond any other cells (6). This is well demonstrated for MR lymphography, where normal lymph node macrophages are filled by iron particles 24 hours after IV injection (7). In EAE, 24 hours after IV injection of USPIO, the major uptake is realized either by activated microglia (intrinsic macrophages) or by blood monocytes or both that have transported the USPIO (Trojan horse) in the site of inflammation through the BBB (1).

Third, it is very well known and established that macrophages are involved in all steps of the development of MS lesions (between 1997 and 1999, more than 140 scientific published articles were related to macrophages in MS). Macrophages accompany the lymphocytes during inflammation, they remove the myetin debris (lipid-laden macrophages), and they are present during the process of myelin repair (8). In such conditions, USPIO imaging may work on MS patients to display the lesions in which macrophages are active.

Fourth, in references 3 and 4 cited by Bulte and Frank, the animal models are traumatic (freezing) and osmotic models of BBB rupture, models in which macrophages are lacking. Furthermore, in those experiments, imaging and histologic analysis were performed within 60 minutes after USPIO injection, which is very different from the 24 hours that we have proposed. Interestingly, in those experiments, USPIO were located on the basal membrane. Thus, this gives credit to the fact that macrophages are required to make the USPIO enter the brain parenchyma after several hours. In reference 5, quoted by Bulte and Frank, a mouse EAE model was tested and Xu et al noted that USPIO could be detected on Prussian blue iron stain in the vascular endothelium, the perivascular space, and in macrophages within perivascular cuffs and areas of inflammation and demyelination. Bulte and Frank are coauthors of the article. It is interesting to note that the MR studies were performed 6 hours after IV injection. In one rat only, they have performed an MR study 24 hours after injection, and they did not observe abnormalities. I would recommend to perform MR and histologic studies 24 hours after USPIO injection in more animals. With this scanning delay, I would not be surprised that USPIO will be found in macrophages and at much lower levels in the vascular endothelium. The scanning delay is critical for performing macrophage activity imaging (3).

USPIO will be available for human use soon, and it will be of a great interest to investigate the potential of this technique in diseases with active macrophages.