LATENT INFECTIONS'

K. F. MEYERFrom the George Williams Hooper Foundation, University of

California, San Francisco, California

In the contemporary literature which deals with the infectiousdiseases, the designation "symptomless infection" is used withincreasing frequency. The question immediately arises: Dorecent observations, experimental and clinical, and new epidemio-logic experiences justify the creation of this term? Is it not likelythat a new word has been coined to describe the well-known stateof "latent infections"? No one will deny that the usual signs ofan infectious malady-the manifest and latent stages-may bereadily bridged by the so called rudimentary or "abortive" formsof the disease. In other words, the symptomatology of an infec-tion oscillates between the frank, clinically typical disease andthe unrecognizable reaction. The persistence of a disease agentfrom the time of its entry into a host or the "symptomless" stateof infection, which may follow clinical recovery, has been de-scribed, since the discovery of the tubercle bacillus by R. Kochin the tissues of apparently healthy individuals, as a "latent in-fection." Furthermore, one was satisfied to characterize certainatypical forms of a disease as "abortive" when it was believedthat the disease-producing virus multiplied only to a moderateextent and consequently induced feeble reactions or no symptomsat all. The recognition of the "abortive" cases has graduallyled to a broader conception of the infectious maladies. Since adisease is generally considered from an etiologic point of view,it is not surprising to note the descriptions of scarlet fever withouta rash and poliomyelitis without paralysis. Whether or not thevariant types of a disease are truly "abortive" forms has certainlynot been determined for a great many infections. In fact, the

'Presidential address prepared for delivery before the Society of AmericanBacteriologists at its thirty-seventh annual meeting, New York, December 27,1935.

109

JOURNAL OF BACTURZOLOGY, VOL. 31, NO. 2

1K. F. MEYER

recent intriguing studies by Paul, Salinger and Trask and othersamply illustrate the difficulties which may be encountered in theinterpretation of the minor illnesses which are observed in asso-ciation with clinical poliomyelitis. Moreover, it must be reservedfor future experimental studies to furnish proof that the attenu-ated clinical course of an infection, which cannot be recognizedwithout the most searching medical examination, is truly theresult of a supposedly diminished multiplication of the diseaseagent. Observations to be discussed later indicate that a numberof additional factors are of equal if not of greater importance.What is the proper concept of a "symptomless infection?"

An observation made by Charles Nicolle in 1912 in connectionwith certain studies on typhus lead to its precise definition byNicolle and Lebailly in 1919. These investigators had notedthat, occasionally, in a series of guinea pigs experimentallyinfected with typhus blood, one or several would fail to show acharacteristic febrile reaction. The blood or organs of suchanimals transferred to other guinea pigs would, however, inducethe typical fever and thus prove the existence of the virus in theanimals which had shown no symptoms. This peculiar behaviorprobably conditioned by a particular constitutional state of theguinea pig was designated by Nicolle as "une infection inap-parente ou silencieuse." In recent years, this phenomenon hasbeen intensively studied and amply confirmed by numerousinvestigators (Doerr, Breinl and others). Thus, Breinl showedthat the simultaneous but separate injection of virus and serumin proper dosage produces an afebrile infection with regularity.Under the influence of the immune serum the virus multipliesand reaches the same concentration in the blood as in the non-treated animals with the only difference that the onset of thevirus propagation may be delayed for a few days. The virusgrowth curve, except for a slight shift in time, is the same in bothanimals, irrespective of the fact that the serum-treated guineapigs remain permanently afebrile. That this type of infectionleaves a permanent immunity was anticipated and experimentallyproven. A further analysis of these interesting observations byBreinl and Singer demonstrated that not only the fever but the

110

LATENT INFECTIONS

other typical symptoms such as the perivascular infiltrations, theblood monocytosis, and the loss in weight were equally absent.Only the polymorphonuclear leucocytes and the lymphocyteswere slightly but temporarily increased for a period of timewhich ordinarily corresponds to the febrile state of the typicallyreacting animal. Obviously, this peculiar type of infection is notabsolutely "symptomless" but it deviates so markedly from whatis generally recognized as the norm that the formulation of a newterm appeared justified. The deciding criterion of the "symp-tomless infection" is a rise and fall of the virus growth curve.Under normal conditions this proliferation of the virus inducesthe well-known symptoms. However, under certain conditions-individual disposition of the animal (Nicolle), influence of im-mune serum (Weil and Breinl), infection with louse-passage virus(Breinl) or shifts in the electrolytes of the body (Breinl andJohn)-the symptom complex fails to materialize though thetypical virus multiplication curve remains unchanged. Theparasite increases in the macroorganism to the characteristicsaturation point (10-6 M.L.D. in the brain) without provoking anoticeable defense reaction, other than a completely "symptom-less" process of immunization. This phenomenon should beconsidered for the time being as a condition decidedly differentfrom that which is generally described as "latent infection."The term "symptomless" or "inapparente infection" should beapplied only to those maladies in which the virus is distributedwith respect to location, time and quantity in a manner char-acteristic for the usual course without, however, causing theappearance of clinical or anatomical signs of disease. As late as1933 Nicolle declares himself opposed to the recognition of thistype of infection as a "latent process." He defines a "latentinfection" as a subacute or chronic state of disease withoutevident symptoms in which the carrier harbors the disease agent.The microbe may have been acquired through a preceding disease;it may regain its infectiousness and thus affect the carrier or mayserve as a source of an infection to other individuals.

Nicolle warns against the error of describing a disease as "inap-parente" should the pathogenic agent escape detection by the

iull

K. F. MEYER

examiner. Non-visibility is not synonymous with "inapparence."Finally, the exponent of the term "inapparente" vigorouslydemands that it be limited to the infectious disease, but not todeficiencies or dystrophies as already proposed by Moriquand,Velu and Balozet.True "symptomless infections" may be expected to develop

when agents similar to the spotted fever virus are involved.According to Nicolle,2 this type of infection is observed in thefollowing diseases and species of animals:

Epidemic European spotted fever in man, rats and miceEndemic rat typhus in guinea pigs, monkeys, mice and ratsKala-azar in the spermophiles (Citeus citellus) and in dogsWeil's disease in rodents and guinea pigsTreponema pallidum infections in mice, rabbits, possibly manVariola in the rabbit, dog and catYellow fever in man and monkeysPoliomyelitisHerpes and encephalitis in manLymphogranulomatosis in the monkeyMalaria in man and probably in AnophelesSpirochaeta recurrentis in the gondi and other animals (Ctenodactylua

gondi)Infectious anemia of horses, rodents and ticksDourineFowl plagueBrucella infections, not infrequently in man, constantly in goats

This brief compilation merely suggests that the world may beflooded by a sea of disease agents which, as a rule, produce merelya "symptomless infection." In order to emphasize the signifi-cance of the problems involved, additional examples are herewithpresented.

In 1929 Blanc, Caminopetros and Manoussakis injected ahuman being intravenously with 3.5 cc. of serum collected from apatient with dengue fever. He remained clinically well but theblood removed from him on the third day after the administration

2 Destin des Maladies Infectieuses, Paris, Felix Alcan, 1934, p. 119; Introduc-tion A la carri~re de la m6decine exp6rimentale, Paris, 1932, pp. 85-91.

112

LATENT INFECTIONS

of the serum and reinoculated in an amount of 7 cc. into a thirdperson proved to be infectious. A typical dengue infection wasproduced. Subsequently, the person with the "inapparenteinfection" was found to possess a specific immunity towards thedengue virus. Nine cubic centimeters of whole virulent blood,which proved virulent in a control experiment induced neithermanifest symptoms nor did the blood of the person reacquireinfectious properties. The same investigators found in 2 out of6 experiments that guinea pigs injected with dengue blood mayyield no febrile reactions while their blood transferred to humanbeings may prove infectious. Transfers to guinea pigs were notsuccessful, doubtless on account of the very marked variabilityof their individual susceptibility. Similarly, experiments con-ducted by Blanc, Caminopetros, Dumas and Saenz on monkeys(Macacus cynomolgus and Cercopithecus callitrichus) have shownthat these species of animals pass through "inapparente infec-tions" when injected with virulent dengue blood. They areneither visibly ill nor is their temperature elevated. However,their blood, avirulent 24 hours after the injection, becomes in-fectious for man on the 5th, remains so until the 8th but loses itspathogenicity on the 12th day. It is of interest that the infec-tiousness can be proven only by transfer experiments to man.Reinoculations on species of monkeys that react quite regularlywith an "inapparente infection" to an injection of dengue bloodinvariably failed. These observations, provided they are con-firmed, would indicate that the extinction of an "inapparenteinfection" in its first generation in certain animal species may bea phenomenon of greatest general biologic significance.Under the influence of a potent immune serum, the course of a

cattle plague or hog cholera infection may be so modified that"symptomless infections" are by no means infrequent. Doubt-less, the combined influence of an inherent natural resistanceenhanced by the immune serum is responsible for these results.That the virus increases in the hog, which has been treated withserum and then injected with virulent blood, may be readilyproven by systematic transfer inoculations to susceptible youngpigs. As an aftermath of this infection, which passes through an

113

K. F. MEYER

afebrile course, a permanent active immunity is produced. Asidefrom the "inapparente infection," a hog cholera virus invasionmay be followed by a typical "latent infection." The virus maypersist for 3 weeks in the blood and up to 103 days in the lymph-nodes (McBryde, David); it may even be discharged irregularlyin the urine for a much longer time.A measles infection under the influence of convalescent serum

may be modified to an unrecognizable condition, absence of feveror Koplik's spots with an indefinite rise in temperature. Yet, it iswell-known that these so called "mitigated forms" of measles arehighly infectious and not infrequently the nodal centers for largeepidemics.In connection with the studies on herpes occasional "inap-

parente infections" have been noted. Levaditi has produced inthe rabbit encephalitides without symptoms; their existence wasplaced in evidence after death through the demonstration of theneuropathologic changes. Other workers have demonstratedthe virus in considerable quantity in the nervous system, irre-spective of the fact that the clinical signs of the brain infectionwere rudimentary or entirely absent. Olitzky and Long notedvirus propagation in the brain of guinea pigs without encephalitisand Remlinger and Lebailly found the herpes virus "non-patho-genic" for the turtle (Testudo mauretanica) although the infectiveagent was demonstrable for a long time by transfer of the braintissue into the susceptible rabbit. Virus increase in the centralnervous system and the pathogenic effect are apparently twoentirely different processes. In fact, no one is in a position to-dayto explain the reason why an increase in the virus is followed inone instance by an acute fatal disease while manifesting itself inanother experiment as a "symptomless infection."

It is the merit of the late Professor Kolle to have applied theterm "symptomless infection" to certain observations which heand his associates, Evers and Prigge, had made in connectionwith the experimental study of syphilis. Rabbits treated withbismuth preparations and subsequently injected with syphiliticmaterial are apparently protected against a luetic infection.Although a chancre formation and the enlargement of the lymph-

114

LATENT INFECTIONS

nodes-the usual criteria of a "take"-are missing, the spiro-chaetes remain viable without producing lesions. Since it hadbeen shown by Pearce and Brown that the popliteal lymph-nodesof rabbits recovered from a syphilitic infection retain the spiro-chaetes, it occurred to Kolle and Evers to extirpate various nodesof the bismuth-treated and clinically unaffected animals, andtransfer them into the testis of normal rabbits. In numerousinstances typical chancres were produced. Thus, it was shownthat the bismuth preparation failed to sterilize and although itprevented the development of local and general syphilitic lesions,it permitted the general distribution of the spirochaetes. Sincethis phenomenon was discovered under more or less artificialconditions, it was at first believed to be of little or no generalsignificance. However, within a short time the field of "inap-parente" spirochaetal infection was broadened. It was notedthat rats and mice may be readily infected by placing pieces ofgummatous tissues under the skin of the back or the testes. Thelocal lesions healed without the formation of a chancre. Aftera lapse of many months the subinoculation of the lymph-nodes,spleen or brain of the infected muridae produced in rabbits typicalchancres. Furthermore, the transfer of organ particles removedfrom mice with "inapparente infections" again created in thisspecies a "symptomless infection." The mice became hosts forthe Treponema pallidum without showing clinical or anatomicalmanifestations. Comparative studies on rabbits leave no doubtthat the distribution and the degree of the spirochaetal increasewithin the tissues correspond in every respect with that estab-lished for the manifest infections. Inflammatory reactions arenot histologically demonstrable in the animals which are notvisibly sick. For the latter the designation "residual braininfection" and for the former "symptomless infection" would becorrect.

In many ways analogous, but not identical, are the observa-tions dealing with the Spirochaeta recurrentis infections of burrow-ing rodents and other animals. Experiments carried out inCalifornia indicate that chipmunks and tamarack squirrels mayharbor spirochaetes in their tissues. Particularly late in the fall

115

K. F. MEYER

these findings are by no means infrequent. Since anatomicallesions are invariably absent and the presence of the spirochaetesis established by subinoculation into mice, the majority opinionof investigators is that these rodent infections are "symptomless."A few carefully controlled tests, however, have shown that thischronic state is preceded by an acute infection which has con-ferred a certain degree of immunity on these carriers. Theserodents serve as a reservoir for the spirochaete and continue tofurnish the parasites which, through the agency of OrnithodorusHermsi, a newly discovered vector are accidentally transmittedto man. With this incomplete evidence at hand one would preferto designate the process in the rodents as a "latent," though insome other animal species and the tick as an "inapparente," oraccording to Nicolle as an "invisible" infection.These and similar reflections now logically demand an analysis

of the term "latent microbismus," "silent or resting infection"introduced during the war by surgeons to characterize the de-layed appearance of tetanus and gas gangrene infections. It iswell-known through the animal experiments of Vaillard andRouget (1892), Tarozzi (1906), Canfora (1908) and Koser andMcClelland (1918) that spores of the tetanus bacillus may remainat the site of the inoculation or may be carried by phagocytosis tovarious parts of the animal body where they persist for longperiods. Although it is as yet unknown whether such persistingspores are free in the tissues or inclosed in phagocytes, the factremains that they may be activated by various chemicals, bac-terial products, etc. and thus induce typical tetanus. For exam-ple, Schneider demonstrated tetanus spores in the liver, spleenand even the blood of rabbits 43 days after their injection withdetoxified spores. Every process, which results in tissue necrosis,spontaneous bacterial infections or artificially induced disease,may lead to an attack of tetanus in these animals which harborthe spores. This peculiar "latency" has been experimentallyestablished for many other toxicogenic anaerobes. In fact, it iswell-known that anaerobic spores introduced into the human bodyby traumatic injuries and gunshot wounds, especially shrapnelsplinters, may remain within the wound or scar tissue without

116

LATENT INFECTIONS

causing the least reaction. These spores, whether deposited withor without a foreign body, are a constant source of danger asnumerous observations collected between 1914 and 1919 havedefinitely shown. An aseptic operation as, for example, the re-moval of shot fragments performed weeks, months or years afterthe original injury may lead either to a local or general tetanus(Courtois, Suffit and Giroux) or a rapidly fatal gas gangreneinfection (Penhallow, 1916, Lesne, 1916, Phocas, 1915, Plant andRoedelius, 1918). There is no doubt that an individual, who, inthe interval between the injury and the subsequent operations,carries the seeds of tetanus or gangrene, is in a state of "latentinfection." However, it is equally obvious that these forms areby no means comparable with the classical "symptomless infec-tion" as illustrated by certain Rickettsia and virus infections.What are the differences? A multiplication of the microbe in thescar may with considerable certainty be excluded. Furthermore,this lack of vegetation of the spores is not due to any activeimmunological effort on the part of the host, but is probablyconditioned by factors which influence the growth phases of themicroorganism. What are the factors which temporarily inhibitand later activate the vegetation of disease agents? Through aseries of brilliantly conceived experiments Fildes (1927) came tothe conclusion that the normal positive oxygen tension of thehealthy tissues is sufficient to account for the lack of germination.This may be reduced in various ways: A slight inflammatoryprocess may cause through capillary thrombosis a hindrance ofoxygen diffusion or the leucocytes may place an increased demandon the oxygen; perhaps the presence of foreign particles induceslocal asphyxia and favorable respiratory requirements for thegermination of the spores. There is no regulated incubation,generalization through the blood stream, no critical disappearanceof the disease agent and no subsequent immunity. It is not thehost per se but certain environmental factors influencing thevegetative phases of the microbe which are of importance.Although Fildes' experiments offer a reasonable explanation foran understanding of the factors which initiate the germinationof the spores, they fail to elucidate the primary inhibition after

117

1K. F. MEYEIR

they have gained entrance to the tissues. Whether phagocytosis(Vaillard and collaborators) or the fortuitous absence of contam-inating microorganisms of the Clostridium Welchii or Clostridiumsporogenes varieties may have something to do with this apparentinhibition deserves further experimental scrutiny. How far theso-called dormancy of spores as observed in the test-tube has abearing on this problem has not as yet been investigated. Inany event it is advisable to define this intriguing state of non-reactive persistence of anaerobic spores in the tissues of man andanimals as "silent or resting microbismus."Promptly the question arises: Are the well-known diseases with

"long incubation times" as, for example, tuberculosis, BruceUainfections, leprosy, fungoidal disease, processes due to cocci,perhaps rabies, psittacosis, etc., analogous to the conditions justreviewed? In no instance has an absolute or complete inhibitionof the growth activities of the implanted microorganism beendemonstrated. Moreover, systematic investigations, as far asthey have been carried out, have always revealed a definitethough occult primary focus. It is supposed that within thisfocus, growth of the disease agent may be very slight and fre-quently retarded or completely arrested by factors which are asshrouded in mystery as the primary inhibitory influences on thespores. The experimental studies by Rous and Jones on theprotection of bacterial cells from the action of destructive sub-stances suggest one of the factors which may be operative. Inany event, all the evidence available marks these conditions as"quiescent infections" in contradistinction from the "silentmicrobismus."

It has already been mentioned that Nicolle presented sufficientevidence to exclude from consideration as "symptomless infec-tions" those conditions in which the persistence of pathogenicmicroorganisms on the surfaces of the mucous membrane oreven in the blood has been recognized. This differentiation hasnot been followed. In fact, under the influence of the newertrends to analyze medical bacteriologic phenomena from a generalbiologic point of view, an interesting terminology has appeared.The word saprophytism is suddenly replaced by symbioses (Rim-

118

LATENT INFECTIONS

pau). Kolle and Prigge in their monograph on "symptomlessinfection" list as "symbionts" (a misnomer for the Greek word"symbiote" = companion or partner) of the mucous membranesor the blood the following disease agents and their hosts.Man. Lamblia intestinalis, pneumococci, anaerobic spore-bearing

bacteria such as Clostridium tetani and Clostridium Welchii; diphtheria,influenza and Ducrey bacillus, fungi and various species of filaria.

Mammals. Pasteurella pestis in rats, fowl chlora bacillus in chickens,various spirochaetes in rodents; Trypanosoma Lewisi in rats; Brucellamelitensis in goats.

Insects. Plasmodia in the Anopheles; trypanosomes in Glossina,yellow fever virus in Aedes Aegypti, pappataci virus in the Phlebotomusflies; spotted fever Rickettsia in lice; Rocky Mountain spotted fever inthe tick, tsutsugamushi virus in mites, etc.

The list could be readily enlarged; one could include the Bar-tonelloses of rats, the anaplasmoses and gonderioses, the findingof plague bacilli in the inguinal glands of natives at Dakar byLeger and Baury, etc.On the other hand, certain corrections and limitations must be

pointed out. The Brucella infections in goats, particularly theelimination of the organism in the urine or its presence in the milkducts, is always associated with definite anatomical lesions and iscertainly not a state comparable with true symbiosis. Manyyears ago Carter had shown that the massive ingestion of game-tocytes by the Anopheles may lead to profound disturbances, evendeath of the mosquito. The flea tolerates an invasion with theRickettsia of endemic typhus while the louse succumbs (Nicolle).More important in this connection, however, is the pertinentinquiry as to whether such a terminology as "symbionts" and itsimplication are truly justified.The presence of a disease incitant on a mucous membrane

merely attests that the microorganism has found conditionssuitable for its vegetative development. However, it may lackthe ability to penetrate and thus be transformed from a potentialto a true disease-producing agent. For the infections withoutsymptoms it has been proven that the agent has the power toinvade, and that the immunity mechanism of the subelinically

119

K. F. MEYER

infected macrobrganism neither neutralizes its action, if any, nordestroys it more readily than in the visibly diseased. AlreadyVaughn recognized such a state of affairs when he applied the termsyssitic to the organisms living commensally on the surfaces of themucous membranes without production of evident disease. Top-ley believes that they possess inherently the power of multiplica-tion on mucous membranes or so called supragliscence. Underthese circumstances, it is quite reasonable to assume that themicroorganism has not as yet adapted itself to the macro6rganism.The residence on the membranes or in the blood is merely thefirst act, the beginning of an interplay which may lead from theinvisible to the visible disease form.

In the light of modern biologic thinking which graduallypermeates the utilitarian concepts of medical bacteriology, it isbelieved, although by no means proven, that the infection with-out symptoms is the sequel and not its precursor. Such reasoningoutlines the problems still to be solved. By taking into considera-tion the newer knowledge on symbiosis, one may perhaps definethe paths along which future inquiries may be directed. Further-more, a discussion of this important and significant relationshipbetween various microorganisms and their hosts may in part helpto clarify such unfamiliar words as Epi- and Endosymbiovsi(Rimpau) as applied to the microbic behavior under consideration.During my early training as a zoologist, the phylogenesis of

parasitism was always presented as an evolution of the parasiticfrom non-parasitic types and it was assumed that the pathogenicascended from the non-pathogenic. With increasing insight itbecame clear that the reverse may also be true. Benign parasitesevolve from harmful ones. Although it is obviously impossibleto observe this phylogenesis, a brief insight into certain familiesof Diptera connected with myiasis offers many convincing obser-vations relative to the correctness of the deductions (Martini).The mode of nutrition of certain species clearly sketches theevolution from true to facultative parasites. The larvae ofmuscae are rarely found in suppurative wounds, while those ofthe Fannia group may live on wound secretions, although theyprefer decomposing plant material. Representatives of the

120

LATENT INFECTIONS

sarcophaginae generally found in feces and dead flesh may invadewounds and destroy even healthy tissues. The genus Wohlfahrtiacauses severe destruction in the healthy tissues of man andanimals which may be the victim of these flies. Lucilia caesarand Calliphora vomitoria now used to cleanse wounds (Baer andothers) unquestionably destroy the dying tissue fragments.Chrysomyia may extend its activities from wounds and puscavities to healthy tissues. In fact, in Australia a great manyCalliphorae deposit their larvae in the wool of sheep. They notonly live on the wool but they attack the skin and cause extensivedestruction. The larvae of Cordylobia, which bore into the skinof dogs and occasionally of man in Senegambia, are primaryparasites. They form the transition stage to the obligatoryparasites belonging to the subfamilies of Cuterebrinae and Hypo-derminae. Their advanced stage of parasitism is indicated by aretrograde formation of the mouth parts, and it is interestingthat many of these forms are living at the expense of the woundsecretions which accumulate within the boil. They avoid in-vasive feeding in the healthy tissues. Polyhagism of certainspecies of Dermatobia is replaced by a specific adaptation of thegenus Hypoderma for certain ungulatae. It appears that the grad-ual and more intimate adaptation to a host and to its mode of livingmore or less improves the existence of the parasitic species. Simul-taneously with this adaptation, the ability to invade other hosts isprogressively lost.

Quite generally, it seems that the less violent action of a para-site on its vertebrate host, the monophagic tendencies and thedifficulty of growing on artificial media are signs of a progressiveand phylogenetically old parasitism. Viewed from this angle, itis obvious that the slowly fatal or chronic illness of the host is ofgreater advantage to the parasite than the rapidly deadly orrapidly cured malady. The highest type of adaptation is recog-nized as a life-long infection of the host (herpes-virus) and, inorder to insure persistence of the parasite, a more or less fullydeveloped pseudoheredity. The final stage is reached with thestate of symbiosis in which the host becomes infected early duringits development and receives no injury through the presence of

121

K. F. MEYER

the "parasite." In fact, the physiology or mode of living of thehost may be so adjusted that it cannot exist without its "sym-bionts." This master-servant relationship has doubtless reachedits highest development in the insects. Of interest in this dis-cussion, however, are the conditions in which the adaptations ofthe parasite to the host are as yet incomplete and the equilibriumbetween host and invader is quite labile.The fields of helminthology, protozoology and botany offer

many examples of harmless and highly adapted parasitism. Un-fortunately, these balanced infections have aroused little generalinterest in the workshops of medical bacteriologists. The micro-biologist, who intentionally administers huge doses of micro6rgan-isms in order to decide the pathogenicity of a disease agent forcertain animals, has doubtless drawn and continues to drawerroneous conclusions. Fortunately, in recent years it has beenappreciated that the massive infections, only too often supple-mented by procedures which paralyze and deaden the resistance,have in no way imitated the natural conditions. Exposureexperiments or the herd studies of Webster, Topley, Greenwoodand Wilson more readily harmonize with the concepts of theparasitologist. This type of experimentation will answer moredefinitely the pertinent question as to which vertebrate or plantis the true host for a microbe or virus. To repeat, it is the animalor plant which assures the parasite a satisfactory permanentexistence and since the physiology of the host harmonizes withthat of the invader, the transmission to a new host of the samespecies is simple and permanently assured. Viewed biologically,an infection would pass phylogenetically through the followingphases: from a free existence, the microbe, through an unequalsymbiosis-parasitism sensutrictu-would reach the balanced andadjusted symbiosis.The term symbiosis as originally used by the botanist, Anton

de Bary (1879), denotes a condition of conjoint life existingbetween different organisms which are benefited in a varyingdegree by the partnership. As Nuttall has pointed out, the termsymbiont, strictly speaking, applies equally to the partners. It has,however, come to be used also in a restricted sense as meaning the

122

LATENT INFECTIONS

microscopic, perhaps invisible member or members of the partner-ship in contradistinction to physically larger partners which inconformity with parasitological usage are conveniently termedthe "hosts." On the other hand, symbiosis must be regarded asa condition of life balancing between two extremes-completeimmunity and deadly infective disease. To be sure, it has beenshown by botanists (N. Bernard, Magrou and others), whosetheories are similar, that the factors governing immunity fromparasites or symbionts are essentially the same for each partner.The struggle between the two partners had led in the course oftime to a form of mutual adaptation. The symbiotic relationshipof orchids and fungi has been experimentally investigated fromthis point of view, and it has been shown by N. Bernard thattuberisation is a sequel or a manifestation of an advanced adapta-tion of the plants to a communal life with fungi. It is worthconsidering that such structures as the mycetocytes, bacteriocytesand mycetomes in the insects are survivals of previous profoundpathologic changes. How far this knowledge has any bearing onthe various symbiotic stages which one observes in man and ani-mals has not as yet been investigated.

It seems advantageous to discuss various other aspects of theproblem of symbiosis as far as they pertain to the subject underdiscussion. Of particular interest are the important studies byOehler, Pringsheim, Goetsch and others on the separation andrecreation of algae symbiosis in paramecia, hydra, rhizopoda, etc.Dependent on whether their habitat is in fresh or in sea water,the endoplasm of these organisms may become occupied byChlorella or Zooxanthella. An analysis of the factors involved inthis type of symbiosis has shown that a separation of the partnersmay be achieved and that each may be cultivated independentlyof the other. However, this separation is no easy task and themany painstaking experiments which have been carried on attestto the striking tenacity with which the algae may adhere to thehost. Various gradations of intimacy may be noted in one andthe same species. Far more important than the separation of theplant from the animal is the causation of the symbiosis. Thereis absolutely no question that the immunity of the Algae against

123

K. F. MEYER

the enzymes of the digestive vacuoles is of prime importance.As Pringsheim has demonstrated, it may be an inherent propertyof the Algae or it may be acquired within a few weeks. CertainAlgae, despite systematic exposure to the invertebrate, remainsusceptible. Equally important in the restoration of the sym-biosis is the regulatory mechanism of the animal partner whichcontrols the growth rate of the invaded Algae. Without thisfactor, the immune plant would readily overwhelm the animal asa parasite. Aside from this mechanism, one encounters as an-other host factor the disposition to accept the Algae as "sym-bionts." Certain species of Hydra freed from Algae, receive theplants and become green but within 10-14 days expel them just likeordinary food particles. In other cases the fresh water polyps areinjured but following a massive and stormy multiplication andexpulsion of clouds of Algae they ultimately recover. This formof elementary parasitism shows a remarkable analogy with theinvasion of man and animals by unicellular plants. Furtherexperiments indicate that the animal partner is, in a manner asyet unknown, profoundly impressed by a preceding symbioticrelationship. This property manifests itself in a greater recep-tivity; thus the customary partnership is rapidly renewed andbalanced. Finally, the operation of the host-symbiont factorsare rendered complex by the formation of physiologic races ortypes, morphologically indistinguishable but non-acceptable tothe actinizoon or protozoon.In connection with the study of a bacterial intracellular

symbiosis in mollusks, the idea was expressed that this conditionhad its origin in true parasitism, probably disease. The samethought probably applies to the many and exceedingly complexsymbioses observed among insects, irrespective of the doubtswhich have been voiced by Buchner, Reichenow, Florence andothers. The comparative pathologist cannot recognize anythingin the mycetocytes and mycetomes other than the survival ofprofound lesions induced by highly parasitic organisms. Suchreasoning casts doubt on the widely expressed idea that theintracellular microorganisms are in a mutualistic relationshipwith their hosts. Until some of these forms have been cultivated

124

LATENT INFECTIONS

-an extensive new field for the bacteriologist acquainted withtissue culture technique-little can be said concerning theirtaxonomic position and their biologic significance. In fact, it isexceedingly doubtful whether a true state of symbiosis has beenestablished. However, in order to avoid further confusion, it isadvisable to retain the term "symbiosis" for the time being.Glaser, who critically compared the symbionts of the insectswith the rickettsiae, correctly concludes that each of these intra-cellular elements originated among the invertebrates in the formof a disease, and that some of them are still in process of adapta-tion to higher animals in which they may produce definite dis-eases. The many examples of "inapparente infections" amplysupport his contention. Furthermore, he proposes a novel con-cept to explain the parasitic tendencies of Rickettsia melophagi.A gradual loss of characters adaptive to the invertebrate may beaccompanied by a progressive modification leading to parasitismin the higher animals, particularly the mammals. His viewsare in harmony with present-day biological concepts and havenumerous analogies as, for example, the evolution of the lifecycle of the malarial plasmodium from a parasitic "symbiotic"state in the Anopheles to an advanced form of parasitism in manor birds. Finally, the biologically minded must agree with himwhen he casts doubt on the common belief that all new diseasesmay arise through sudden permanent modifications or variationsof a disease agent instead of through a slow, orderly, evolutionaryadaptation to the new hosts. It is not the purpose of this analysisto express an opinion relative to the importance of the "sym-bionts" as sources of growth-accessory factors, hormones, etc.That the "symbiosis" has a diversified function throughout theanimal kingdom is fully recognized in all the speculative inter-pretations which have been published. However, Carter re-cently suggested that the phytotoxic secretions of certain coccidstransmitting the spotting of pineapple leaves may be conditionedby the activities of the mycetome and its included symbionts.What bearing this hypothesis may have on the solution of certaininsect borne diseases must remain a matter of conjecture.Many of the facts briefly considered in the foregoing para-

125

126

graphs have prompted a number of bacteriologists to propoundwell sounding hypotheses and far reaching analogy deductions.For example, the commensalism of protozoa, ciliates in particular,in the rumen or stomach and in the coecum of herbivore has beendesignated as an endosymbiosis, which is active in the digestion ofcellulose. This and the "symbiotes" themselves are thought toserve as nitrogen sources (for the horse it is estimated to be aquarter to one-third of the daily protein requirements) for thehost. The implantation of bacteria into the buccal cavity and.the intestinal tubes is compared with similar conditions in theinsects and is known as episymbiosis. It has even been suggestedthat the early seeding of the mucosa of the mouth may be com-parable with the bacterial contamination of the eggs of Lagriahirta while passing the smear glands of the urogenital tract of theinsect. The new-born, passing through the vaginal canal, thusbecomes the recipient of the maternal "symbionts." Additionalexamples of visionary comparisons could be cited but the fewchosen emphasize again the unfortunate tendency to generaliza-tions. From what has been outlined concerning symbiosis, itshould be obvious that not one of the examples listed by Kolleand Prigge conforms to the fundamental definitions of a symbiosis.One could hardly support the idea that any one of the bacteria-pneumococcus, diphtheria bacillus, meningococcus, etc.-is in amutualistic relation with the human host or that the gray rat isbenefited by a partnership with Trypanosoma Lewisi. Since atrue state of symbiosis has not been established, it is only properto describe the organisms present on the mucous membranes orin the blood and tissues without evident disease as commensateand not as symbionts.

It is quite evident that such terms as "inapparente," silentinfections or latent microbismus and commensalism describe avariety of processes. Since not infrequently the terminology hasbeen applied without the necessary degree of care, an unfortunatestate of confusion and contradictions has arisen. It is to thelasting credit of R. Doerr (1931 and 1934) to have clarified theproblem, not only by proposing a tentative system in which thevarious independent processes or phases of an infectious disease

Kr. P. MEYER

LATENT INFECTIONS

previously discussed may be arranged, but by listing some of thefactors responsible for infections without disease. He clearlystates that the principal criterion is the absence of manifestationsof the disease which may be recognized either by direct observa-tion or may find expression in man by a feeling of malaise. Theopposite to manifest is latent; consequently the nomenclaturewill be greatly simplified if the various terms are classed under thecollective designation "latent infection." Since "latency" ismerely a clinical concept, it is quite permissible to use this termeven when anatomical lesions inaccessible to direct observationor immunological and allergic reactions are recorded. Undercertain circumstances one may even go further and describe ahuman being as "latent-infected" when general symptoms areabsent and local reactions so exceptionally mild that the condi-tion is recognized only by bacteriological procedures (for examplea mild pharyngitis with diphtheria bacilli). Equally significantand applicable to a better understanding are the concepts withrespect to the phases of an infectious disease. The state of atyphoid carrier is a latent one and not a manifest chronic infectionirrespective of the fact that the presence of the typhoid bacillusmay reveal itself clinically and anatomically (cholecystitis, focalnephritis). In clinical circles typhoid fever is defined as an acute,three-phased process which terminates with the restoration of thetemperature to normal. With the onset of recovery, the clinicalsyndrome "typhoid fever" is passed and the ensuing carrier stagerepresents an entirely different condition which remains mostlylatent and may only be recognized through a bacteriologicalexamination.

In his latest discussion of the subject Nicolle (1935) now definesthe "inapparente infection" as an infectious disease withoutsymptoms. Its existence may be placed in evidence by direct orindirect laboratory methods, inoculation tests or serologic,immunologic procedures. He says, "L'infection inapparenteperdra une part de son domaine au ben6fice de l'infection Asympt6mes. R6duites, elle demeurera la forme mineure desinfections, celles qui 6chappe a la pratique usuelle du clinicien.L'inapparence n'a de sens que par rapport A la clinique."

127

K. F. MEYER

From what has been said, it is obvious that "latent infections"may be either independent processes or phases of an infectiousdisease. This is clearly set forth in a descriptive system proposedby Doerr (1934) which makes no claim for completeness butmerely outlines the main characteristics, their occurrence andtheir relationship to the clinically manifest processes.

SYSTEM OF LATENT INFECTIONS

(1) Infections latent during their entire existence(a) Cyclical processes. This group covers the "infections

inapparentes" of Nicolle.(b) Acyclical or chronic processes. It is advisable to subdivide

the group into (aa) the "silent microbismus" (tetanus, anaerobes,etc.) and (bb) the "quiescent infections" (tuberculosis, leprosy,etc.) irrespective of the fact that a clear distinction between thetwo forms is sometimes impossible.

(2) Latency as a phase of an infectious disease(a) As a precursor of the malady. In the majority of infectious

diseases, this period of latency may have a remarkably regularand consequently normal value while in others it may be eitherunknown or it may be exceptionally long-certain chronicinfections, tuberculosis, actinomycosis, undulant fever, malaria(primary latency), rabies, etc. Diseases with pronounced tend-ency to latency rarely reveal regular incubation times. Undulantfever offers a great many striking examples in support of thisstatement: A young man has a febrile attack of one day duration;his blood yields a culture of Brucella melitensis but a frank clinicalattack of undulant fever develops but 3 months later. Intyphoid, cerebrospinal fever and diphtheria these so calledincubationary (Nicholls) or precocious carriers may play animportant rile in the epidemiology of these diseases.

(b) As a sequel of an infectious disease. The state of latencyin these so called "shedders" may be temporary and be terminatedbv spontaneous recovery or they may last for years or even duringthe entire life of the host. The word "shedder" has been em-ployed since the proliferating microbes are, as a rule, eliminated

128

LATENT INFECTIONS

in the excreta and secreta. However, a carrier of "malarialgametocytes" is not a "shedder." Furthermore, distinctionshave been attempted by designating as a "germ carrier" ananimal or a human being who contracted the latent stage as anindependent process without a previous visible disease. Thisconfusing nomenclature is properly replaced by the significantdesignations "healthy" and "convalescent" carriers. Certainobservers (Sacquep6e and Doerr) believe that this differentiationis particularly valuable in diphtheria since they found the bacillito persist longer in the convalescent than in those who had norecognizable clinical signs of the disease.

(c) As an intermediary stage. In the course of certain diseaseswith an intermittent or undulatory clinical course, periods oflatency alternate with those of obvious disease. Dependent onthe length of time which elapses between the manifest attacks, itis customary to speak of exacerbations or relapses, provided areinfection may with certainty be excluded. In malaria, relapsingfever and certain septic processes, occasionally in Brucella in-fections, the sequence and duration of the latent and manifestphases follow with striking precision and are well regulatedaccording to time. Malarial fevers offer an opportunity toanalyze some of the factors which guide the latency. The febrileattacks coincide with the schizogony of the merozoites or aga-monts and develop only provided the asexual forms reach a cer-tain threshold which is in part dependent on the type of themalarial parasite (according to Ross, at least 140 quartan, 200 to300 tertian and 600 to 3,000 tropical parasites per 1 cc. must bepresent before a febrile attack is induced). In vaccinationmalaria, not less than 14,000 tertian parasites per cubic centimetercause fever. From the many speculative hypotheses, whichhave been advanced in order to explain the febrile paroxysms as anaftermath of the schizogony, the concept of poison formation bythe interaction of agamonts and cellular detritus on the bloodplasma (Doerr), and the phagocytosis of the parasites by thereticulo-endothelial cells (Taliaferro) deserve serious considera-tion. The investigations of Cannon and Taliaferro on the cellularbasis of resistance to both avian and simian malaria suggestively

129

K. F. MEYER

illuminate the r6le of the macrophage-lymphocyte system in themechanism of latency. These superb studies have shown thatin the course of these infections a certain percentage of the para-sitized red blood cells are phagocytized by the "hemophages" ofKyes and rendered inert. Peculiarly, this function is not con-tinuous but synchronizes with the development of each newgeneration of parasites. The periodicity of malarial fever maywell be intimately connected with this mechanism. Duringlatency against plasmodia, the host acquires some factor orfactors which specifically cause the macrophages to ingest theparasites. Whether this altered reactivity of the reticulocytesis supported by a humoral factor in form of a tropin is not as yetsettled. Similar in many ways is the mechanism in relapsingfever. Autopsies on children (Bykowa, 1926) or animals in-fected with Spirochaeta recurrentis show a marked hypertrophyof the reticulo-endothelial apparatus. An interplay betweenlytic antibodies and the mesenchyme cells guides the periodicityof the fever attacks. Protection of the spirochaetes within thecells and development of antibody-fast variants may lead tolatency. In many septic diseases (undulant fever, variousstreptococci infections, etc.), it has been noted that, as a rule,the causative microbes are less likely to be isolated from theblood during an afebrile than a febrile period. Then again, incertain animal diseases, the phenomenon of "microbial showers,"by no means explained, merely indicates that in all probabilitythe bacteria multiply irregularly. Some clinicians, by analogywith protozoan infections, attribute this behavior to growthcycles. However, the evidence, which gradually accumulates,involves the cellular activity of the host and not of the parasites.The diverse experimental studies to convert a latent infection

into a manifest process and vice versa offer many suggestions rela-tive to the causes of latency. By splenectomy, latent plasmodialinfections of apes (Gonder and Rodenwald) and piroplasmoses(dog, sheep and horses) have been changed into severe, evenfatal infections. The entire Bartonella problem received a greatimpetus from the systematically planned removal of the spleenin rodents. The same principle applied to other latent protozoaninfection has been even more fruitful. A Trypanosoma Lewisi

130

LATENT INFECTIONS

infection of the gray rat runs, as a rule, a benign and clinicallylatent course. Regendanz and Kikuth (1927) showed that insplenectomized rats the multiplication of the trypanosomes wasfrequently not retarded. In fact, the infection was thus trans-formed into a manifest process and frequently terminated fatally.Similarly, the extirpation of the adrenal by Marmorston, Gotter-man and associates was followed by a like result in approximately70 per cent of the experiments.To the disappointment of many a worker, latent bacterial or

virus infections in laboratory animals have been recognizable bythe injection of diverse agents. Carriers of Bacterium leptisep-ticum may be made recognizable by placing silver nitrate or dyesinto the nostrils of the suspected rabbits. The microbian asso-ciations (Microbio de salida or Microbes de sortie of MauriceNicolle) with protozoa in virus diseases belong to this group.With increasing frequency, "latent infections" are converted intovisible diseases by transferring organ material of various animalsto new hosts. One recalls the discovery of the virus III of rab-bits, the encephalitis virus of mice and the protozoon Encephali-tozoon cuniculi. Saprophytic viruses are discovered in cultures(Barnard). Vice versa, as already detailed, manifest infectionson a large scale and quite successfully have been transformedinto latent stages by shifting the field of action from one hostto another species. Although theoretically of greatest interestand subject to extensive investigation, these experimental meta-morphoses contribute little to the understanding of the problemof latency in man.

Concerning the mechanism or the cause or causes of latency,the available knowledge is exceedingly fragmentary. For a fewcases, a satisfactory explanation may be offered. The localiza-tion of the infectious process may condition the latency. A smalltuberculous or fungoidal focus in a lymph-node frequently doesnot provoke clinical signs. The commensalistic existence ofmeningo- and pneumococci on the mucosa of the respiratorytract remains "symptomless" until the invasion of certain organsis followed by the typical clinical syndromes of cerebrospinalfever or lobar pneumonia.

Certain forms of latency are, in part at least, influenced by the

131

1K. F. MEYER

behavior of the disease-producing agent. The mechanism of theresting tetanus or gas gangrene infection has been explained as aphenomenon of growth inhibition which takes place at themoment the microbe is transmitted. In contrast to this primarylack of vegetation of the spores, a secondary retardation in thegrowth of bacteria in encapsulated abscesses, scarred or calcifiedtissues may be equally responsible for the absence of symptoms.A small number of infective agents, in part inhibited, may nevermultiply to a level which incites the clinical reactions. Under thecircumstances, the infection may remain latent either through along incubation time or for variable periods between primaryattacks and relapses. In the considerations which are given tothe carrier problems, one meets not infrequently the hypothesisthat the latency was the result of a microbe of low or even non-pathogenicity. It is easy to assume fluctuations in virulence ofthe causative organism but its relationship to latency can rarelybe proved. When the pathogenicity is associated with a specifictoxin, successful correlations have been established. Thus,occasionally healthy diphtheria carriers may harbor organismsof relatively low toxicity. In fact, injrare instances, non-toxicstrains have been isolated. However, as a whole there is noevidence that a true Corynebacterium diphtheriae loses completelyits ability to produce toxin, and as such becomes an orgamsmparticularly endowed to create carriers. On the other hand, thereare authenticated reports in the field of latent microbian andvirus infections which conclusively prove a persistent and uni-formly high pathogenicity during a very long sojourn of the dis-ease agent in the tissues of the host. Under the influence ofincreasing but incomplete knowledge on the variability of bacte-ria, it is by no means surprising to note that latency is believedto be the result of a transformation from a pathogenic to a non-pathogenic variety. The frequent demonstration of roughvariants in the excreta of typhoid-paratyphoid carriers lendssupport to these hypotheses. In fact, certain groups encouragethe concept that these variants are stages of a continuous adaptiveprocess leading ultimately to a harmless symbiont. Others,making use of the unsolved problems of life cycles, suspect

132

LATENT INFECTIONS

changed facultative invisible forms of the disease agents to befactors responsible for the latency in tuberculosis and syphilis.The effort is thus made to incriminate only the microbes. Despitethe ubiquity of certain bacteria and the disease induced by them,not everybody becomes a carrier or acquires a latent infection.Without more precise information relative to such vague termsas "infectiousness" or pathogenicity, little progress will be made inthe understanding of latency.The mechanism known as "organotropism" by which is selected

the tissues in which the parasites localize themselves and themany factors which control their growth and multiplication havebeen analyzed. Technical difficulties have prevented studieswith plant parasites and viruses, but for certain macroparasites,Trichina spiralis and protozoa (malaria), the selective organlocalization has been proven. Equally encouraging are thepreliminary investigations on the influence of a trypanosome in-fection on the protein, carbohydrate and fat metabolism of certainrodents. It is assumed that the parasites disturb the metabolismand that, in turn, the abnormalities affect the nutrition andmultiplication of the protozoa living in such a milieu. A studyof Pasteurella pestis infections in hibernating rodents with pro-foundly changed metabolism discloses interesting and importantfacts which may help to explain certain causes of latency.

In the course of an infection, the environment to which theparasite is exposed changes continuously. It is certain that thedisease agent is subjected to influences of varying intensitiesduring the incubation time, the onset, the peak and the healingof the infection. Adaptations are suspected and newer studiesattempt to demonstrate all kinds of transformations although fortechnical reasons, it is usually impossible to prove conclusivelytheir existence in the living host. The solution of such intricateproblems requires expert knowledge of modern experimental-pathological methods, and should not be attempted by those whomerely know the "test-tube" and not the "animal" phases of amicrobe.

Infectiousness and pathogenicity are relative conditions whichare not dependent on the parasite alone, but in more important

133

13K F. MEYER

degree, on the individual, racial and species characteristics of thehost. Aside from a disposition for a manifest disease, the hostin all probability possesses an equally important yet independentsusceptibility for infections without pathological consequences.Latency is conditioned by the behavior of the infected macro6rgan8sm.For the sake of completeness, the 'various possibilities may bebriefly noted. The possession of one factor or factors, whichantagonize or neutralize the injurious effect of the microbe, forexample, antitoxin content of the blood in a diphtheria carrier,may be important. Or a preceding infection of like etiology orphases of the disease may change the reactivity of the organismto such an extent that a reinfection or persistent infection mayappear or remain in a clinically latent state. Since these modifi-cations in the clinical course are the result of immunity reactions,Doerr has proposed the term "immunosatoric latency." Whythe defensive mechanism remains incomplete and then suddenlyceases to function so that the disease becomes manifest as arelapse or is so intensified that a carrier state after months andyears is promptly healed has not been explained. It must befrankly admitted that the method by which the host participatesin these reactions is not known. These resistance variationsshould be correlated with phagocytic activity. Studies in thisdirection are obviously indicated. Latency is rather frequentin the infections in which the phagocytic mononuclear cellsactively participate in the defense, or furnish the environmentmost conducive toward a favorable development of the infectiveagent. Extensive studies on the histopathology of Brucellainfections and psittacosis suggest an experimental approach to aclearer understanding of the immunosatoric latency.Dependent on certain conditions of the host are those forms of

latency which already as primary infections remain free fromclinical signs. The latent yellow fever and typhus fever ofchildren, dengue in adults, typhoid bacilli in the blood of healthypersons and various Rickettsia infections in laboratory animalsbelong to this group. Since the effects of active or passive im-munity can with certainty be excluded in the typhus infectionsof laboratory animals, it is strange that this state of individual

134 '

LATENT INFECTIONS

tendency to "latent infection" has not been correlated with theoperation of genetic factors. Even the inmunosatoric latencyis in part influenced by these forces. Competent observers havein recent years pointed out that focalization and chronicity of amicrobian disease process is not merely conditioned by individualimmunizing subinfections but ultimately may be traced to theselection of resistant generations (Hirzfeld). Though the analysisof constitutional immunity is in its infancy, the data thus faravailable suggest that latency is often the expression of the host'sinheritance rather than the variability of the disease agent.However, it must be stated with frankness that it will be impos-sible to list or discern more in detail the causes of latency untilaccurate information concerning the factors which promote orsuppress symptoms of an infectious disease is available. Theproblem of "inapparente infections" is a part of the unsolvedproblem of pathogenicity. In its broader aspects it is the key tothe understanding of latent epidemization and as an evolutionaryphase participates in the rise and decline of certain diseases.The results of the scientific investigations of this importantproblem will doubtless furnish a rationale for the natural fluctua-tions of epidemics which now cannot be completely explainedeither through the frequency of the manifest infections or by thevariability of the causative agent. Finally, this analysis hasshown that a comprehension of mass disposition is intimatelydependent on a clearer understanding of individual dispositionand disease symptomatology.

135