3.15.
FREEZE-SUBSTITUTION
79
could be connected to the specimen chamber with Teflon tubing. Alternatively, they might be fitted with nipples of a size to fit standard hypodermic needles. If the specimen chamber had Teflon sidewalls, entrance
and exit ports could be made by sticking hypodermic needles through
these. The needles themselves were equipped with stopcocks. Thus, different reagent bottles could be connected in turn, breaking no more
than one tiny seal for a brief interval of time. Rebhun used glass serum
bottles as specimen chambers and fitted them with stoppers pierced by
two hypodermic needles for introducing and withdrawing fluid (one of
the needles served as a vent).
Fernandez-Moran has emphasized to this writer that he believes ice
crystal "seeding" should be avoided at low temperatures, and he reported
extraordinary precautions taken to avoid this with all solutions that were
used. Mainly the problem is to avoid water crystals that result from trace
quantities of water in even the best commercially available, so-called anhydrous reagents. He, therefore, advocated removing all traces of water
by exposing all reagents to Linde 4A molecular sieves (Chapter 4.2). After
this, particulate matter was removed by filtering the anhydrous compounds through sintered bacteriological ultrafilters. In some instances it
also proved necessary to have vacuum redistilled the reagent before this.
Reagents prepared in this way subsequently were kept under vacuum to
avoid gas absorption. Before chilling, the reagents to be used were put
in sealed chambers of a design such that they could be connected to the
specimen chamber at will, and transfers made without breaking the seal.
Thus, water vapor was never allowed to condense upon and contaminate
the reagents or specimen.
The time required for freeze-substitution obviously depends upon such
a number of variables that only guide lines can be indicated. Bullivant
(1960), who studied small pieces of pancreatic tissue, allowed 2 weeks
for substitution at —75° C, using three changes of methanol as the solvent.
Rebhun (1961), studying clam oocytes, waited for 3 days to 2 weeks with
acetone as the solvent at — 80°C. The presumption is that these time intervals were determined empirically, and they may have been excessive. Bartl
(1962) found that it took about 2 weeks to substitute small tissue blocks
and cell suspensions with glycol methacrylate at —60° or —40° C. In this
instance tissue floated until the substitution was nearly complete.
Fernandez-Moran suggests that a simple "model system" be incorporated in the specimen chamber which can serve to give a clue as to when
substitution is complete. Basically this is no more than a short length
FREEZE-SUBSTITUTION
79
could be connected to the specimen chamber with Teflon tubing. Alternatively, they might be fitted with nipples of a size to fit standard hypodermic needles. If the specimen chamber had Teflon sidewalls, entrance
and exit ports could be made by sticking hypodermic needles through
these. The needles themselves were equipped with stopcocks. Thus, different reagent bottles could be connected in turn, breaking no more
than one tiny seal for a brief interval of time. Rebhun used glass serum
bottles as specimen chambers and fitted them with stoppers pierced by
two hypodermic needles for introducing and withdrawing fluid (one of
the needles served as a vent).
Fernandez-Moran has emphasized to this writer that he believes ice
crystal "seeding" should be avoided at low temperatures, and he reported
extraordinary precautions taken to avoid this with all solutions that were
used. Mainly the problem is to avoid water crystals that result from trace
quantities of water in even the best commercially available, so-called anhydrous reagents. He, therefore, advocated removing all traces of water
by exposing all reagents to Linde 4A molecular sieves (Chapter 4.2). After
this, particulate matter was removed by filtering the anhydrous compounds through sintered bacteriological ultrafilters. In some instances it
also proved necessary to have vacuum redistilled the reagent before this.
Reagents prepared in this way subsequently were kept under vacuum to
avoid gas absorption. Before chilling, the reagents to be used were put
in sealed chambers of a design such that they could be connected to the
specimen chamber at will, and transfers made without breaking the seal.
Thus, water vapor was never allowed to condense upon and contaminate
the reagents or specimen.
The time required for freeze-substitution obviously depends upon such
a number of variables that only guide lines can be indicated. Bullivant
(1960), who studied small pieces of pancreatic tissue, allowed 2 weeks
for substitution at —75° C, using three changes of methanol as the solvent.
Rebhun (1961), studying clam oocytes, waited for 3 days to 2 weeks with
acetone as the solvent at — 80°C. The presumption is that these time intervals were determined empirically, and they may have been excessive. Bartl
(1962) found that it took about 2 weeks to substitute small tissue blocks
and cell suspensions with glycol methacrylate at —60° or —40° C. In this
instance tissue floated until the substitution was nearly complete.
Fernandez-Moran suggests that a simple "model system" be incorporated in the specimen chamber which can serve to give a clue as to when
substitution is complete. Basically this is no more than a short length