JOURNAL OF CLINICAL MICROBIOLOGY, Apr. 1977, p. 501-502
Copyright ©D 1977
American Society for Microbiology
Vol. 5, No. 4
Printed in U.S.A.
Comparison of Cream of Rice Agar and Horse Serum for
Differentiating Germ Tubes of Candida albicans from
Filaments of Candida tropicalis
NETTIE M. WARWOOD* AND DONNA J. BLAZEVIC
Department ofLaboratory Medicine and Pathology, University ofMinnesota, Minneapolis, Minnesota 55455
Received for publication 26 July 1976
Simple cream of rice agar was superior to horse serum for the demonstration
ofgerm tubes by Candida albicans and in the differentiation ofpseudohyphae of
Candida tropicalis from germ tubes at 37°C. Mycelium and chlamydospores
were also produced on this medium.
The rapid formation of germ tubes by Can-
dida albicans and, more rarely, by Candida
stellatoidea is a useful test in the rapid identifi-
cation of these organisms. In general, some
type of serum or fluid medium has been used
(5). Beheshti et al. (1) have reported the use ofa
solid medium for demonstration of germ tubes
within a 3-h incubation period; subsequently,
mycelial and chlamydospore formation could
also be noted on the same agar. This medium,
RIOT agar, is cream ofrice agar to which oxgall
and Tween 80 are added. Our laboratory has
been using a simpler cream ofrice agar without
additives (2) for production ofpseudomycelium,
mycelium, and chlamydospores, and we de-
cided to evaluate its usefulness in determining
germ tube formation by Candida. Detection of
germ tubes on the same medium used for chla-
mydospore production would eliminate one step
in the identification of these yeasts.
A total of 111 stock cultures (51 isolates ofC.
albicans, 33 of Candida tropicalis, 9 of Can-
dida parapsilosis, 8 of Candida krusei, 4 of
Candida guilliermondii, and 2 of C. stellato-
idea) and 10 yeasts, recently isolated from clini-
cal specimens and subsequently identified as C.
albicans, were tested in both horse serum and
on the cream of rice.
Cream of rice agar was prepared by the
method of Taschdjian (4) as modified by Dun-
can and Floeder (2). After autoclaving, the me-
dium was poured into 60- by 15-mm petri dishes
and allowed to harden. The plates were inocu-
lated by making a light, slightly cloudy suspen-
sion of the yeast in 0.5 ml of sterile saline; a
loopful ofthe suspension was streaked for isola-
tion onto the cream of rice plate; one stab was
made with the loop into the area of heaviest
inoculum; and the area of stab was covered
with a 22-mm square cover slip. The plates
were inverted and incubated for 2 h. Isolates of
C. albicans, all of which were inoculated in
duplicate, were incubated at 35 and 37°C. All
others were incubated only at 37°C. After incu-
bation, covers were removed and plates were
observed for germ tubes under the low and high
dry objectives with decreased light, and then all
were incubated overnight at 30°C.
Tubes of 0.5 ml of horse serum were inocu-
lated at the same time as the cream of rice
plate. The cultures were incubated in a 37°C
heating block (Temp-Blok module heater, Sci-
entific Products, Inc., Evanston, Ill.) for 2 h,
mounted on a slide, placed under a cover slip,
and observed for germ tubes under low power
with decreased light.
All of the cultures of C. albicans produced
germ tubes in 2 h on the cream ofrice medium,
irrespective of the temperature of incubation,
but a 2-degree elevation in temperature made a
difference in the quantity and length of germ
tubes produced. Although germ tubes were pro-
duced at 35°C, those cultures incubated at 37°C
had more numerous and longer germ tubes.
Sixty of the 61 isolates (98%) produced germ
tubes in horse serum. All but five produced
chlamydospores within 24 h, but three of these
were positive by 48 h.
Twenty-eight (85%) of the 33 isolates of C.
were unequivocally negative
germ tubes in the horse serum. The other five,
formed filaments morphologically
similar to germ tubes. All 33 were negative for
germ tubes on the cream of rice plates, al-
though several produced pseudohyphae in 2 h
that were easily distinguished from germ tubes
by their broad bases and definite pinching at
the point of attachment to the yeast cells (3).
None produced chlamydospores.
No other cultures of Candida, including the
two of C. stellatoidea, produced germ tubes in
either medium or formed chlamydospores on
the cream of rice.
The cream of rice method correlated 100%
J. CLIN. MICROBIOL.
with the stock cultures of C. albicans and the
10 patient isolates ofthis organism (all ofwhich
were identified by physiological
compared to a 95% correlation using the horse
serum, which is not statistically significant.
The significant difference was the ability to
recognize the beginning pseudohyphae of C.
tropicalis on cream of rice compared to horse
serum. From our own observations, C. tropi-
produces beginning pseudohyphae
horse serum that are difficult to distinguish
from germ tubes. Had identification rested
solely with the results using the serum, these
cultures could have been misidentified as C.
Due to the ease of differentiating beginning
pseudohyphae ofC. tropicalis from germ tubes,
and because ofeconomy, we feel that the cream
of rice plate method for detecting germ tube
production is superior to serum. We did not
compare our medium with the more complex
rice agar of Beheshti et al. (1). The two media
tested were equally effective for germ tube for-
mation; however, the cream of rice agar offers
the obvious advantage of greater simplicity of
formulation. Neither presumes to differentiate
between C. albicans and the occasional isolate
ofC. stellatoidea capable offorming germ tubes
and/or chlamydospores, of which we had none
in the present study. To differentiate between
the two, however, only a sucrose assimilation
test would have to be done.
1. Beheshti, F., A. G. Smith, and G. W. Krause. 1975.
Germ tube and chlamydospore formation by Candida
albicans on a new medium. J. Clin. Microbiol. 2:345-
2. Duncan, J., and J. Floeder. 1963. A comparison of me-
dia for the production of chlamydospores by Candida
albicans. Am. J. Med. Technol. 29:199-206.
3. Mackenzie, D. W. R. 1964. Morphogenesis of Candida
albicans in uivo. Sabouraudia 3:225-232.
4. Taschdjian, C. L. 1957. Routine identification of Can-
dida albicans: current methods and a new medium.
5. Taschdjian, C. L., J. J. Burchall, and P. J. Kozinn.
1960. Rapid identification ofCandida albicans by fila-
mentation on serum and serum substitutes. Am. J.
Dis. Child. 99:102-105.