M. W. MILLER AND E. M. MRAK

SUMMARY

A species of the bacterial genus Flavobacterium iscapable of synthesizing a polysaccharide or mixture ofpolysaccharides when grown in a medium of a-coniden-drin or vanillic acid as a sole carbon source. The poly-saccharide hydrolysate consists of D-galactose, D-glu-cose, D-mannose, D-ribose and L-rhamnose as deter-mined by paper chromatography.

This organism has the unique characteristic of grow-ing luxuriantly at the expense of the compounds a-coni-dendrin and vanillic acid, while little if any growth oc-curs when the conventional type carbon sources areemployed.

REFERENCES

BRAUNS, F. E. 1945 The occurrence of conidendrin inwestern hemlock. J. Org. Chem. 10, 216-218.

KONETZKA, W. A., PELCZAR, M. J., AND GOTTLIEB, S. 1952The biological degradation of lignin III. Bacterial de-gradation of alpha-conidendrin. J. Bacteriol. 68, 771-778.

PARTRIDGE, S. M. 1948 I. General description and applica-tion to the qualitative analysis of sugars in apple juice,egg white, and foetal blood of sheep. Biochem. J. 42,238-250.

PARTRIDGE, S. M. 1949 Aniline hydrogen phthalate as aspraying reagent for chromatography of sugars. Nature,164, 443.

PEARL, I. A. 1945 Conidendrin from western hemlock sulfitewaste liquor. J. Org. Chem., 10, 219-221.

WILSON, A. T., AND BRUNO, P. 1950 The sterilization ofbacterial media and other fluids with ethylene oxide. J.Exp. Med., 91, 449-458.

Yeasts Associated with Dried-Fruit Beetles in Figs

M. W. MILLER AND E. M. MRAK

Department of Food Technology, University of California, Davis, California

Received for publication March 13, 1953

One of the most common insects infesting figs is thedried-fruit beetle, Carpophilus hemipterus (Linn.). Itsrole as a vector of yeasts and bacteria responsible forcertain diseases of figs has been recognized for decades.Nevertheless, the ecological relationships between theinsects and the yeasts associated with them are inade-quately known. There is a difference in opinion as towhether the dried-fruit beetle feeds on fig fruit tissue orthe fungi (particularly yeasts) occurring in the fruit, orboth. Phillips, Smith and Smith (1925) and Caldis(1927) indicated the beetles feed on fruit tissue whereasSmith and Hansen (1931) and Hansen (1951), expressedthe view that the beetles are mycophagous and enterthe fig in search of fungus foods. Regardless of the reasonfor the insects entering the fig, the result of their en-trance is direct or indirect damage to the fruit. The pres-ence of beetles or their excreta or the microbial damagefor which they are indirectly responsible make the figsinedible.Mrak, et al., (1942) reviewed the literature concerned

with the microbiology of the fig fruit. They also reportedtheir own studies concerning the occurrence of varioustypes of yeasts in soured figs. One hundred and fifteenyeasts were isolated from the interior of infected freshfigs prior to drying. Sixty-four of the yeasts formedascospores, whereas 51 were imperfect forms. Most ofthe isolates were species of Saccharomyces or Candida,although a few cultures of Pichia, Hanseniaspora, Klo-eckera, and Torulopsis and single cultures of Hansenulaand Debaryomyces were also obtained.

There is no available information concerning theassociation of yeasts with the dried-fruit beetle that canbe interpreted in terms of presently accepted taxonomy.Furthermore there are no data to substantiate the viewthat the dried-fruit beetle responds chemotactically toyeast. The purpose of this investigation was to obtaininformation conceriing these points.

EXPERIMENTAL METHODS

Beetle Collection

Beetles were collected between August 14 and Septem-ber 15, 1950, from two varieties of figs (Adriatic andCalimyrna) in three orchards, two of which were locatednear Fresno and one near Planada, California. Collec-tions were made between mid-morning and mid-after-noon since the beetles were more active and more apt tofeed when the fruit temperature was above 20 C.The figs yielding beetles had undergone some drying

and nearly all showed some sign of souring. Beetles werenot found in immature or very dry figs, or in figs suffi-ciently cracked to permit exposure of the inside juicyflesh.

Beetles were collected directly from the fruit byholding the open end of a sterile vial (1 x 3 inches) overthe "eye" of the fig while it was gently tapped andsqueezed. This induced beetles to emerge from the in-fested fruit and drop into the vial. Figs were also tornopen to enable removal of beetles from deeper portionsof the cavity.

174

on Novem

ber 6, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

YEASTS AND DRIED-FRUIT BEETLES IN FIGS

The tubed beetles were either washed and dissectedimmediately or cooled rapidly to 5 C and after severalhours were washed and dissected within 25 minutes.This procedure was essential since the beetles werefound to digest yeasts very rapidly at room tempera-ture.

Dissection of Beetles and Isolation of Yeasts

Each beetle was removed from the vial with sterileforceps and placed in several drops of sterile water onthe suirface of a wort-propionate-agar plate (5° Brix wortwith 0.25 per cent sodium propionate). The washedbeetle was removed and residual water streaked evenlyover the plate. The beetle was then dipped in 95 percent ethanol and mounted, dorsal side up, on a sterileparaffine block. The head was severed and the anal seg-ments of the exoskeleton broken without severing thegut. The gut was then removed with a very fine needlehaving a hooked end, and placed in a few drops of sterilewater on a second wort-propionate-agar plate. It wasthen pulled apart and the contents streaked on theplate. Pure cultures of yeast were obtained by threesuccessive platings.

Identification was accomplished by use of the pro-cedures of Lodder and Kreger-van Rij (1952).

Digestion of Yeasts by Beetles

Very few yeasts were obtained from the intestinaltracts of the beetles during the early stage of the inves-tigation when exploratory tests were being conductedfor the purpose of developing experimental procedures.This was attributed to the rapid rate of digestion ofyeasts by the beetle, although there was no evidence tosubstantiate this view, aside from the observations ofShihata et al., (1952) on Drosophila. It was consideredessential, therefore, to obtain data on the rate of diges-tion of yeasts by C. hemipterus. Beetles which had beendeprived of food for 36 hours were permitted to feed on48-cultures of Candida krusei or Hanseniaspora valby-ensis. The beetles started to feed within a few minutesafter being placed in the vial with food. After feedingfor about two hours the beetles stopped eating andstarted moving about the vial. At this time they wereremoved and held at 25 C for 0, 2, 4, 8, 16,24, or 36 hoursbefore the intestinal tracts were removed and brokenopen in 1 ml of sterile water for subsequent serial dilu-tions and plating for counts. Results of the counts fromthe dissection of 42 beetles are given in table 1. Thenumber of cells observed immediately after feeding ismuch lower than reported for Drosophila by Shihata et.al. (1951). Drosophila, however, has a crop capable ofdistention so that it sometimes nearly fills the abdomi-nal cavity whereas Carpophilus hemipterus does nothave such a structure. Shihata and Mrak observed thatthe rate of digestion was much slower at lower tempera-tures.

Similar observations were made on the dried-fruit

beetle. For these reasons it was considered essential todissect the beetles very soon after collection or to storethem at a low temperature (5 C) until dissection.

Yeasts Isolated from Beetles

Seventy-five cultures of yeasts were isolated from thegut and the exterior surface of the beetles, 32 of whichwere sporulating and 43 non-sporulating forms. Most ofthe isolates were identified as species of Hanseniasporaand Candida rather than Saccharomyces and Candida asfound by Mrak, et al., (1942) on fresh figs. Several iso-

TABLE 1. Rate of digestion of yeasts by Carpophilus hemipterusat 25 C

AVERAGE NO. VIABLE CELLS IN INSECTTIME AFTER FEEDING __

Candida krusei Hanseniaspora valbyensis

hr

0 15,000 15,3002 300 3004 75 1008 40 5016 20 2024 2 336 0 0

TABLE 2. Number and species of yeasts isolated from driedfruit beetles

FROM IN- FROM SUR-TAXONOMIC DESIGNATION TESTINAL FACE OF TOTAL

TRACT BEETLE

Candida krusei.9 16 25Saccharomyces rose. 1 1Hanseniaspora valbyensi .17 13 30Kloeckera apiculata.1 2 3Pichia membranaefaciens .1 1Rhodotorula mucilaginosa 1 1Torulopsis albida .1 1Torulopsis carpophila nov. sp 3 5 8Torulopsis lactis-condeni.1 3 4Torulopsis stellata . . 1 1

Total number of isolates.33 42 75Sporulating isolates .19 13 32Non-sporulating isolates.14 29 43

lates of Torulopsis and Kloeckera and single cultures ofSaccharomyces, Pichia, and Rhodotorula were also iden-tified (see table 2).

Sporulating yeasts. The 30 isolates of H. valbyensiswere similar to the culture (termed Hanseniaspora mel-ligeri) originally isolated from dates by Melliger (Lod-der, 1932) and subsequently from California figs, datesand prunes by Mrak, et al., (1942). The single culturesof Saccharomyces rosei and Pichia membranaefaciens aresimilar to the described species which was first isolatedfrom soil and is presently characterized as a food spoil-age organism capable of tolerating adverse conditionsfor growth.

Non-sporulating yeasts. It is of interest that over half

175

on Novem

ber 6, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

M. W. MILLER AND E. M. MRAK

of these yeasts were identified as Candida krusei, an

organism that has been isolated in California from figsand dates a number of times. Three cultures identifiedas Kloeckera apiculata are probably imperfect forms ofHanseniaspora. Aside from one culture of Rhodotorulamucilaginosa, all other isolates were members of thegenus Torulopsis. Eight of these are included in a new

species (Torulopsis carpophila), although they resembleTorulopsis holmii in some respects. They differ by fer-menting the entire raffinose molecule rather than only13 of it, and by forming clusters of cells as well as singleor pairs of cells in liquid wort. Other differences includethe formation of a thin pellicle and certain slant culturecharacteristics. These differences are considered suffi-cient to justify the description of a new species.

TABLE 3. T'emperature range for growth of yeasts isolated

ORGANISM

Candidakrusei .......

Hanseniasporavalbyensis....

Kloeckeraapiculata ....

Torulopsisalbida .......

T. carpophila,nov. sp......

T. lactis-con-densi ........

Saccharomycesrose.........

Pichia mem-

branae-faciens ......

Rhodotorulamucilagin-osa..........

T. stellata.....

TEMPERATURE (C)

Minimum Maximum

0 2.5 5

+ +

+

8.5

+

34 37 39.51 44 46 48

+ +

Torulopsis carpophila sp. nova

Growth in liquid wort after 24 hours, cells single, inpairs or in small clusters. Cell shape, globose to ellips-oidal. After 4 days and 3 weeks cell characteristics same.Cell size, width (1.5-4.2,u) and length, (2.4-S.1,u). After3 days in liquid wort a very thin incomplete dull filmand heavy ring.Growth on wort-agar after 4 days, cells mostly single

and in pairs, globose to ellipsoidal. Range of cell size(1.5-4.8,) x (2.1-5.11u). Wort slant- culture nearlywhite, surface smooth, slightly glossy, texture soft andpasty, border entire, cross-section slightly convex witha flat narrow border.Ferments glucose, galactose, sucrose, raffinose

strongly and melibiose weakly. Maltose and lactose notfermented. Assimilates glucose, galactose, maltose, and

sucrose. Does not assimilate lactose. Utilizes peptone,asparagin, ammonium sulphate, and urea but not ni-trate, in the presence of vitamins.

Alcohol utilized in the presence of vitaminis, with theformation of a white sediment and a very thin, smooth,dull pellicle. Without vitamins, the pellicle is completeand sediment formation is meager. No esters formed inalcohol medium or liquid wort. Weak to moderate acidformation on Custer's chalk agar. Fat not split. Arbutinsplit readily.

Temperature and growth. Since the activity of thedried-fruit beetle was greater at the higher tempera-tures occurring between mid-morning and mid-after-noon in the localities where the beetles were collected,it was deemed advisable to determine the temperaturerange for growth of the yeasts isolated. This was doneby inoculating 10 per cent wort-agar slants and incu-bating at various temperature. The results obtainedfrom representative isolates are summarized in table 3.Most of the yeasts isolated from the dried-fruit beetle

were able to growr at rather high temperatures for yeast.This is especially true for C. krusei, an organism whichhas not been reported as being able to grow at 46 C.Most yeasts grow best at temperatures below 37 C. Atlow temperatures only T. albida was unusual in that itgrew at 0 C, a characteristic in line with that of the typespecies.

Sugar tolerance. The fig substrates from which theyeasts were isolated ranged in sugar concentration fromabout 40 per cent (relatively moist) to 55 per cent andsometimes 60 per cent (commercially dry). Since theyeasts came from beetles in figs which had shrivelledand undergone some drying, the question arose concern-ing the ability of the yeast to grow under these variousconditions. In order to obtain information relative tosugar tolerance of the organisms isolated, they wereinoculated into fermentation tubes containing 150 Brixliquid wort to which cerelose was added giving concen-trations from 350 to 600 Brix. The majority of the iso-lated organisms produced gas in tubes containing the500 or 550 Brix medium but only two (Saccharomycesrosei and Torulopsis lactis-condensi) showed growth inthe 600 medium.

Transfers of cultures of Saccharomyces isolated byMrak, et al., (1942) were tested for growth in a 450 Brixmedium and it was found to be practically absent. Theyalso noted that all isolates of Saccharomyces as well as anumber of other cultures failed to grow in 400 Brixsyrup. Thus it is quite possible that the influence ofconcentration, resulting from maturation and drying ofthe fig may account for the difference in types of organ-isms isolated in the present study and those found byMrak, et al.

Attractivity tests. Yeasts commonly occurring in figshave been found in large numbers in the intestinal tractsof most of the beetles collected from figs. This might betaken as evidence to indicate that beetles are attracted

176

on Novem

ber 6, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

YEASTS AND DRIED-FRUIT BEETLES IN FIGS

by the yeasts present in the fig. This is still an openquestion, however, for it can be argued that yeastsare ingested by accident and that they play no role inattracting the beetle to the fruit. Such reasoning givesrise to the thought that the effect of yeast and fig tissuecombined might hold the greatest attractivity for thebeetle.

In an effort to obtain information on questions suchas that indicated above, a series of attractivity testswere made by use of a variety of substrates. These con-

TABLE 4. Attractivity tests with dried-fruit beetles

SUB- ORGANISM OR COMBINATION OFSTRATE ORGANSMS

23

4

1920212218910151617185

67

11121314

Yeast A (Candida krusei)Yeast B (Hanseniaspora val-

byensis)Yeast C (Torulopsis carpo-

phila nov. sp.)Yeast A and BYeast A and CYeast B and CYeast A, B, and CNoneYeast AYeast BYeast CYeast A and BYeast A and CYeast B and CYeast A, B, and CYeast A and AcetobacterYeast B and AcetobacterYeast C and AcetobacterYeast A and B and AcetobacterYeast A and C and AcetobacterYeast B and C and AcetobacterYeast A, B, and C and Aceto-

bacter

SUBSTRATE

AgarAgar

Agar

AgarAgarAgar

AgarFig TissueFig TissueFig TissueFig TissueFig TissueFig TissueFig TissueFig TissueFig TissueFig TissueFig TissueFig TissueFig TissueFig TissueFig Tissue

Total .... 146

Average .... 6.63

sisted of all possible combinations of the three predomi-nant types of yeasts isolated from the dried-fruit beetles(C. krusei, H. valbyensis, and T. carpophila) and a cul-ture of Acetobacter (table 4).

Active cultures of yeasts were inoculated onto agarslants and pieces of Calimyrna figs containing about 65per cent moisture. These were incubated at 25 C for 24hours before use in attractivity tests. In the Acetobactertests, transfers were made from active cultures to 24-hour yeast-fig cultures which in turn were incubatedanother 24 hours before use. The combination of Aceto-bacter and yeast was used with the view of imitating figsouring that occurs quite commonly in the orchard.

Beetles were deprived of -food for 36 hours prior tosubjecting them to the attractivity tests.Twenty-two substrates were placed in small vials (1 x

13k1 inches) which were arranged at random on the cir-

cumference of a circle 18 inches in diameter. The entiretest was conducted in a large pan with steep sides toretain the beetles in the test area. The 22 vials wereplaced on their sides with each opening located towardthe center. Precautions were taken to eliminate thepathway of the previous beetle by placing a paper disk(12-inch diameter) in the center of the circle before re-leasing the next beetle. Light intensity was kept lowand evenly distributed over the pan. The room in whichthe tests were conducted was free of other food odorsthat might have distracted the beetles. Attractivity wasindicated by the vial entered and in which the beetle fed.The results were interpreted statistically.An examination of table 4 shows that fig tissue alone

and yeast C alone were least attractive to the beetles.Single yeast cultures on agar also had a low attractivity.The combination of yeasts A and B on agar slants at-tracted larger numbers of beetles; however, other com-binations of yeasts on agar were much less attractive.When single yeasts were grown on fig tissue the attrac-tivity increased to a point. Growing more than one yeaston fig tissue failed to increase the attractivity signifi-cantly in contrast to the effect of combining yeasts onagar slants.

Statistical analysis of the data indicated that there isno significant interaction among the five factors con-sidered: Yeast A, B, and C, Acetobacter, and fermentedfig tissue. Furthermore, the beetles were indifferent tothe factors of Acetobacter and fermented fig tissue, butshowed a preference for yeast A, B, fig tissue, and yeastC in this order. Under the assumption of no interaction,there was a significant preference for yeast A. Yeast Aproduced a stronger ester odor than did B or C. Thefactors of sourness due to Acetobacter and fermentationcaused by yeasts on fig tissue were shown by statisticalanalysis to have no influence on the attractivity of thefig to the beetle.

ACKNOWLEDGMENTSThe authors gratefully acknowledge the kind assist-

ance received from Dr. H. J. Phaff, Dr. Robert Warnerand Mr. C. D. Fisher during the investigation, and fromDr. E. L. Scott and Dr. I. J. Abrams on statistical ad-vice and evaluation with respect to the attractivity tests.

SUMMARY

Seventy-five yeasts were isolated from the gut andexterior of the dried-fruit beetle (Carpophilus hemipte-rus, (Linn.)). Most of the yeasts isolated were Candidakrusei (25) and Hanseniaspora valbyensis (30). Otheryeast isolates were Kloeckera apiculata (3), Pichia mem-branaefaciens (1), Rhodotorula mucilaginosa (1), Saccha-romyces rosei (1), Torulopsis albida (1), Torulopsis car-pophila, nov sp (8), Torulopsis lactis-condensi (4), andTorulopsis stellata '(1).One new species, Torulopsis carpophila, sp nova, is

described.

177

on Novem

ber 6, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from

ROBERTA S. HARTMAN ET AL.

The temperature range of growth was 5-39.5 C. Can-dida krusei was able to grow at 46 C. Torulopsis albidawas able to grow at 0 C.The sugar tolerance was 50-55O Brix syrup. Saccha-

romyces ro8ei and Torulopsis lactis-condensi were able totolerate 600 Brix syrup.

Attractivity tests with dried-fruit beetles indicatedthat they responded most to Candida krusei, then Han-seniaspora valbyensis, fig tissue, and Torulop8is carpo-phila, nov sp in that order.The rate of digestion of yeasts by the insects was

rapid, most of the cells being destroyed at 25 C in 1-4hours after feeding.

REFERENCESCALDIS, 1P. D. 1927 Etiology and transmission of endo-

sepsis (internal rot) of the fruit of the fig. Hilgardia 2,287-328.

HANSEN, H. N. 1951 Personal communications.LODDER, J. 1932 Uber einige durch dass "Centraal Bureau

voor Schimmel-cultures" neuerworbene sporogene Hefen-arten. Zentr. Bakt. Parasitenk., II, 86, 227-253.

LODDER, J. AND KREGER-VAN RIJ, N. J. W. 1952 The yeaat8-a taxonomic study. North-Holland Publishing Company,Amsterdam.

MRAK, E. M., PHAFF, H. J., VAUGHN, R. H. AND HANSEN,H. N. 1942 Yeasts occurring in souring figs. J. Bac-terial. 44, 441-450.

PHILLIPS, E. H., SMITH, E. H., AND SMITH, R. E. 1925 Figsmut. Calif. Agr. Exp. Sta. Bull., 387, 1-38.

SHIHATA, A. M. El-TABEY AWAD, AND MRAK, E. M. 1951The fate of yeast in the digestive tract of Dro8ophila.American Naturalist, 85, 381-383.

SHIHATA, A. M. EL-TABEY AWAD, AND MRAK, E. M. 1952Intestinal yeast floras of successive populations of Dro-sophila. Evolution 6, 325-332.

SMITH, R. E., AND HANSEN, H. N. 1931 Fruit spoilapediseases of figs. Calif. Exp. Sta. Bull. 506, 1-84.

An Electrophoretic Method for the Assay of Bacterial VariantsROBERTA S. HARTMAN, H. T. EIGELSBACH, J. B. BATEMAN, WERNER BRAUN, RUTH HERRING AND R. D. RODGERS

Camtp Detrick, Frederick, Maryland

Received for publication March 19, 1953

Investigations by Moyer (1936b), Joffe and Mudd(1934), Stearns and Roepke (1941), and others haveshown striking differences in electrophoretic mobilitybetween the variant cells of (certain species of bacteria,especially betwveen mutants differing in antigenic char-acteristics and colonial morphology. Therefore it seemedthat at least for some species it should be possible todevelop an electrophoretic method for assaying the de-gree of heterogeneity within populations containingvariants. One method commonly employed to detectvariants differing in antigenicity, virulence, colony mor-phology, and other associated characteristics (Braun,1947) consists of streaking cell suspensions on platescontaining transparent agar medium and subsequentlyinispecting colonial morphology with the help ofobliquely tranismitted light (Eigelsbach et al., 1951).The percentage of each variant type can then be esti-mated by counting the frequency of a given colonialtype among at least 100 colonies inspected. The hope ofbeing able to secure by the electrophoretic method theadvantage of a rapid assay, in contrast to the delay in-herent in the method based upon plating and inspectionof colonial types, led us to the investigation reportedhere.The first series of experiments was desigined not only

to compare the electrophoretic method with the platingmethod described above, but also to provide ani estimate

of the consistency of measurement of each of thesemethods. The measurements were made on mixturescontaining a substantial fraction (at least 20 per cent)of the minor component of a pair of variants. The organ-isms used were Pasteurella tularensis (Bacterium tula-rense), strain 38, variants 38 SI (smooth) and 38 NS,(nonsmooth) (Eigelsbach et at., 1951). Neither strainproduces clumps in liquid culture.The second series was designed to compare the two

methods when the minor fraction comprised 1 to 20 percent of the mixture, and to determine thus whether theappearance and subsequent establishment of the non-smooth variant in an initially smooth broth culturemight be followed by the electrophoretic method. It wasdone with P. tutarensis variants isolated from the Schustrain, namely Schu S3 (smooth) and Schu NS3 (non-smooth). The latter has a pronounced tendency toclump in saline or broth (Eigelsbach et al., 1951).

PRINCIPLES OF THE ELECTROPHORETIC ASSAYIn electrophoresis by the microscope method, parti-

cles suspended in a conducting medium are observed asthey move under the influence of an electric field. Thevelocity is calculated from the observed time (called the"excursion time") taken by the particle to traverse agiven distance, as measured in the eyepiece reticle. Theelectrophoretic velocities of individual cells in a ho-

178

on Novem

ber 6, 2020 by guesthttp://aem

.asm.org/

Dow

nloaded from