Observations on Early Stages of Phagocytosis of Colloidal Particles by 



Hepatic Phagocytes of the Mouse^ 



1, 2 



H. F. Parks and A. D. Chiquoine 



Department of Anatomy, University of Rochester School of Medicine and Dentistry, Rochester, N.Y., and the 

 Laboratory for Biological Ultrastriictiirc Research Department of Anatomy, Karolinska Institiitet, Stockholm, 



and Department of Biology, Princeton University, Princeton, N.J. 



When colloidal suspensions of gold or mercury 

 chloride are injected intravenously in the mouse, 

 microscopically visible inclusions of the colloidal 

 particles very rapidly become visible in hepatic 

 phagocytes. Optically homogeneous aggregates from 

 I // in diameter down to the limits of microscopic 

 visibility are seen in animals killed as early as 15 

 seconds after injection. Larger spherical inclusions 

 about 1 micron in diameter appear after about 2 

 minutes and are very numerous by 10 minutes 

 following injection (1). They have been observed 

 previously with the electron microscope and seen 

 to be membrane-enclosed structures (3). The question 

 of how inclusions of phagocytized colloidal particles 

 are formed is presently being investigated, and mor- 

 phological observations on hepatic phagocytes of 

 animals killed during the first 2.5 minutes following 

 injection are reported here. 



Hepatic tissue was taken from 50 white mice weighing 

 15 to 20 grams. Twenty-five were killed at intervals of 

 15 seconds to 2.5 minutes following intravenous injection 

 of colloidal gold or mercury chloride. The others received 

 no experimental treatment. Small strips of liver were 

 removed in a cold room and placed in buffered (pH 7.2) 

 isotonic I "o osmium tetroxide solution, where they were 

 cut into blocks one mm or less in thickness with a stainless 

 steel razor blade. After 15 minutes to 4 hours (usually 

 one hour) in fixative, the tissue was rinsed 10 minutes 

 to I hour in Tyrode's solution, dehydrated in a graded 

 series of alcohols, and embedded in methacrylate (10 or 

 20 parts methyl to 90 or 80 parts butyl). The polymeriza- 

 tion was catalyzed by 0.4 "^o powdered benzoyl peroxide. 

 A considerable part of this material was discarded be- 

 cause of improper polymerization of methacrylate in the 

 center of the blocks of tissue. The sections were cut on a 

 Sjostrand ultramicrotome and analyzed in an RCA 

 EMU 2c electron microscope. 



(1) At 15 seconds following injection colloidal 

 particles were seen in several relationships to endo- 

 thelial phagocytes: 



Particles attached to cell surface. Some cells appar- 

 ently have a "sticky" membrane that immobilizes 

 particles coming in contact with it. These cells were 

 seen with their surface largely covered with particles 

 while neighboring cells showed few or no attached 

 particles (fig. I ). Their surfaces were sometimes very 

 irregular, characterized by numerous small pseudo- 

 pod-like processes that were also covered with 



^ Partially supported by the National Science Foundation, 

 Washington, D.C., and the Whitehall Fund. 



- An earlier stage of this work was reported in an abstract 

 in the Anatomical Record, V, 124 (1956 Meetings of the 

 American Association of Anatomists). 



particles. Adherence to a cell membrane is of course 

 a necessary condition to phagocytosis, and the rela- 

 tionship here described is considered an early stage 

 of the phagocytosis process. 



Particles located in depressions of the cell metn- 

 brane. Some sections were seen in which the whole 

 luminal surface of a cell did not show an affinity 

 for particles, but particles were found associated 

 with semicircular (in profile) depressions of the cell 

 membrane. The fact that particles were in relation 

 to the depressed portion but not to other parts of 

 the cell membrane suggests that the depression repre- 

 sents an invagination formed in response to the 

 presence of particles on the membrane. A depression 

 appearing semicircular in profile could be either a 

 groove or a spherical invagination. The latter possi- 

 bility seems likely in this case because circular profiles 

 of membrane-enclosed groups of particles were 

 usually seen in the cytoplasm in the vicinity of the 

 turned-in portion of membrane; these were sections 

 of spherical inclusions or cross-sections of spherical 

 (or cylindrical) invaginations from the surface. This 

 general picture suggests that some phagocytosis 

 takes place by a localized spherical or cylindrical 

 invagination of cell membrane that pinches off the 

 surface membrane to form a spherical inclusion. A 

 miniature example of this type of relationship is 

 seen in fig. 1 , B. 



Particles in intracellular clefts. In some sections 

 shallow to deep infoldings of cell membrane were 

 seen enclosing narrow cleft-like intracellular spaces 

 (cf. fig. 2). Some contained a large number of 

 particles; others only a few. It cannot be stated 

 whether such clefts pre-existed or were formed by 

 invagination of the cell membrane in response to 

 adherent colloidal particles. 



Particles in intercellular ( ?) spaces. In many sec- 

 tions a cleft-like space between separate masses of 

 cytoplasm was seen communicating both with the 

 lumen of the sinusoid and the perisinusoidal space. 

 Whether such a cleft was intra- or intercellular in a 

 given instance could not be ascertained from a single 

 section; it might be a cleft between two cytoplasmic 

 processes of the same cell or between two separate 

 cells. At any rate, particles were seen both in these 

 clefts and in the perisinusoidal space. 



(2) At later stages of the phagocytosis process up to 

 2.5 minutes, few changes of a qualitative nature 

 were seen. Phagocytes were more conspicuous be- 



