evidence from which this concept evolved. Stein
and Stein localized radioautographic grains over
subcellular structures after the intravascular injection
of [SH]glycerol and palmitic acid. Although
the interpretation was made initially that grains
were localized over the rough endoplasmic reticulum
by 2 and 5 min, the published images show
grains only in the vicinity of the ends of parallel arrays
of RER or of single RER cisternae, and no
grains were clearly associated with flattened cisternae
containing ribosomes (49). Although the
grain size was too large for precise localization,
these investigators have recently stated that
grains were seen often in the regions of transition
between the rough and smooth ER (50). Thus,
contrary to previous interpretations, the radioautographic
studies suggest to us that the ribosome-
studded, flattened cisternae of the rough
endoplasmic reticulum do not synthesize triglycerides.
The presence of label clearly localized in
regions of transition between smooth and rough
endoplasmic reticulum, particularly at the earliest
times of sampling of liver (2-5 min after injection),
suggests that triglyceride synthesis occurs
either in this transitional compartment or in
nearby smooth ER cisternae (and is then transportedto the transitional zone) or in both sites.
Glaumann et al. found no clear precursorproduct
relationship between the rates of labeling
of triglyceride obtained from rough- and
smooth-surfaced microsome fractions at the earliest
times of sampling, 3-5 min (21). Triglyceride
synthetase activity is distributed about equally
between rough and smooth microsome fractions
isolated from rat livers (54). Although these observations
have been interpreted to indicate that
triglyceride is synthesized in both compartments in
intact ceils, it is possible that mechanical rupture
of membranes of the intact endoplasmic reticulum
does not successfully separate the critical subcompartments.
Smooth-surfaced ends of rough ER
may form vesicles either with or without ribosomes
attached so that the enzymes, apoproteins and
lipids contained therein are distributed between
smooth and rough microsomes. In both the radioautographic
and cell fractionation studies, there
was adequate time for triglyceride synthesis to
have occurred in the smooth ER and for triglyceride-
rich particles to be transported to the transitionalcompartment. We conclude that none of the
techniques used to date (radioautography, cell
fractionation or immunoelectron microscopy)presently provides adequate resolution to establish
the sequence of events in the assembly of VLDL.
However, we believe that the pathway outlined in
Fig. 14 provides a reasonable explanation for our
own observations and that this pathway is not
excluded by data obtained with other methods.
Previous studies suggested that the Goigi apparatus
of the liver concentrates nascent VLDL into