Generation of a Specific-Pathogen–Free Baboon Colony

Abstract

We undertook establishing an SPF baboon colony in response to requests from researchers. To enable the widest possible future use of SPF baboons, our aim was to develop an SPF colony of baboons (Papio hamadryas anubis) free of 12 target viruses: 5 herpesviruses, 4 retroviruses, simian virus 40, measles, and monkeypox. Infant baboons were removed from their mothers within 24 h of birth and nursery-reared. Groups of 3 to 8 age-matched conspecifics were isolated in separate rooms for 1 y while undergoing repeated testing for target viruses. During the initial 7 y of the SPF program, 171 infants were enrolled, of which 76 (44.4%) subsequently were removed from the program. Of those removed, 54 (71.0%) were culled due to breaks in virus-free status, 12 (15.8%) died of various causes, 4 (5.3%) developed seizures, and 6 (7.9%) were removed for other reasons. The most problematic viruses were baboon cytomegalovirus (25.9% of culls), Herpesvirus papio 1 (51.9%), and simian foamy virus (7.4%). Using conspecific groups of 3 to 4 infants reduced first-year program losses as compared with groups of 6 to 8. There have been 17 births in the SPF colony, and all these infants have been free of all target viruses since birth. On the basis of these results, early removal of infants from their dams, housing in small peer groups, frequent virus testing, and aggressive culling of virus-positive animals is an effective approach for development of a baboon colony free of multiple viruses.

Wild-caught and conventionally raised baboons have many indigenous viruses, bacteria, and parasites.⁸ In addition to pathologies occurring in infected animals, many of these agents can influence the results of biomedical research using baboons. Furthermore, some zoonotic agents can pose an occupational risk to human caretakers and research staff.¹⁴ For all of these reasons, SPF animals of many different species have been produced and are used widely in research.

Retroviruses and herpesviruses are of particular concern to biomedical research due to their ability to influence the immune system and to persist in the host after the initial infection event. Reactivation of latent herpesviruses is an important problem in immunosuppressed subjects,^{6,11,19,20,44} and several herpesviruses encode genes for proteins that directly affect the host immune response.^{2,34,45,49,52} Because monkey herpesviruses are closely related to their human virus homologs, testing of human herpesvirus vaccines in conventional nonhuman primates is problematic due to antigenic crossreactivity between human and enzootic simian herpesviruses.^{1,7,9,12,25,29,35} In addition, the ability of most baboon herpesviruses to infect human cells in vitro raises the specter of zoonotic infections.

Similarly retroviruses are important due to their zoonotic potential and ability to alter immune responses. Simian retrovirus type D (SRV) and SIV can both have an immunosuppressive effect on infected animals. Simian T-lymphotropic virus (STLV) has been associated with lymphoma in baboons, and simian foamy virus (SFV) is a considerable problem for researchers due to its reactivation and destruction of ex vivo cell cultures. Simian virus 40 (SV40) is a papovavirus of rhesus monkeys, and baboons carry antigenically similar viruses.^33,48 Although little is known about SV40 in baboons, neurologic lesions in immunosuppressed macaques have been described.⁵ There is also a danger of zoonotic disease, as evidenced by primate handlers that have been infected with SV40, SRV, or SFV.^{5,10,14,23,31,42} Therefore, many reasons support the development of SPF colonies of baboons free of enzootic viral infections.

A number of institutions have developed SPF macaque colonies, all of which originally targeted 4 viruses designated as problematic by the NIH (SRV, SIV, STLV, and Macacine herpesvirus 1 [monkey B virus]).³⁰ Since these colonies were established, a need for animals that are free of additional viruses has become clear. In response, several institutions have proceeded to rederive SPF macaques that are free of several additional viruses (so-called ‘superclean’ SPF animals). This multistaged approach is a very expensive process, requiring additional facilities, personnel, and years to establish a second SPF colony. Directly targeting many viruses at the outset to develop a superclean SPF colony had not been attempted previously.

Because most herpesviruses and retroviruses are spread through direct contact, theoretically it should not be much more difficult to develop SPF animals free of many of these viruses rather than just a few. In 1997, 3 infant baboons were necessarily delivered by cesarean section at our facility (the University of Oklahoma Health Sciences Center). To test the ability to keep infant baboons free of multiple viruses (cytomegalovirus [BaCMV], Herpesvirus papio 1 and 2 [HVP1, HVP2], simian varicella virus [SVV], and SFV), these 3 infants were hand-reared and subsequently housed in the same room as conventional baboons but kept in separate cages without allowing any direct contact with conventional baboons. Serologic testing revealed that 2 of these baboons remained free of all the target viruses for more than 3 y, whereas the third baboon acquired BaCMV after 2 y (BaCMV typically is acquired by infants younger than 1 y³³). Therefore, deriving SPF baboons free of multiple target viruses appeared to be an obtainable goal. This observation together with data on the prevalence and transmission dynamics of many of the target viruses in our conventional baboon colony³³ suggested that a one-step derivation process would be sufficient to establish an SPF colony of baboons free of multiple viruses. The many inquiries we have received over the years from investigators regarding the availability of SPF baboons prompted us to attempt establishment of such an SPF colony. Here we present the results to date and summarize some of the problems encountered during the process of establishing a superclean SPF baboon colony.

Materials and Methods

Animal care.

All housing and procedures were performed in compliance with the Guide for the Care and Use of Laboratory Animals¹⁸ and an environmental enhancement program was in place. Protocols for maintenance of baboon colonies (conventional and SPF) were approved by the University of Oklahoma Health Sciences Center Institutional Animal Care and Use Committee.

Derivation of SPF baboons.

Daily records are kept routinely on the estrus cycle of all adult females in the conventional breeding colony, thereby allowing accurate prediction of the date of delivery. Neonates to be enrolled into the SPF program were selected antepartum according to paternity and predicted delivery date such that 2 to 4 infants would be enrolled within a 2-d period in order to form age-matched peer groups. This practice also allowed condensation of the constant 24-h husbandry needed to hand-rear infants into discrete time intervals. All infants were delivered naturally in the conventional baboon colony, which is housed in large indoor-outdoor corrals. Baboons typically deliver between dusk and dawn, and neonates were removed from their dams as soon as possible on the morning after birth. Dams were separated from the colony into a squeeze cage, anesthetized (10mg/kg ketamine), and their infants were removed. Neonates were transported to the SPF nursery to be hand-raised and dams returned to their group. Care was taken to not remove successive infants from female baboons. Approximately 20 to 30 infants were recruited into the SPF program each year.

Strict barrier principles were put in place to prevent disease transmission from conventional baboons housed in the same building. Baboons were cared for by technicians dedicated to the SPF program who had no assignments with conventional baboons. Before entering SPF areas, all personnel were required to take a wet shower and change into SPF-exclusive scrubs and shoes. Furthermore, contact with conventional baboons could not occur within the 24 h prior to entering the SPF areas. All equipment and supplies were exclusive to the SPF areas. Room air pressure was set up so that the SPF areas were positive with respect to conventional areas. Sick or injured baboons were treated within the SPF area if possible. On the rare occasion when an animal had to be taken out for diagnostic purposes or surgery, it was not allowed to return to the SPF program.

Newly arriving neonates received a warm bath to remove all apparent soil and then allowed to air dry after a quick dip in chlorhexidine disinfectant. They were individually housed at 29 °C in stainless steel cages (61 × 61 × 41 cm) for the first 2 wk. The cages included a hanging surrogate and bottle holder (Figure 1) so that interactions with human personnel were kept to a minimum. Beginning at 2 wk of age, infants spent increasing periods of time in a larger cage (183 × 122 × 196 cm) with other infants until they were living together full-time at 3 mo of age. Infant baboons were fed infant formula (Similac Advance, Abbott, Abbott Park, IL) at the regular concentration recommended for human babies until being weaned onto commercial chow (LabDiet Primate Diet, Northeastern Products, Warrensburg, NY) at approximately 3 mo of age. At about 1 y of age, baboons were placed into large aluminum and concrete gang cages. Enrichment included commercial product (LabDiet Monkey Morsels, Northeastern Products), bananas, apples, and ‘novelty produce’ (for example, celery, grapes, cucumbers) given 3 to 4 times each week.

Figure 1.

Initial infant caging arrangement. A 1-wk-old infant baboon is shown with the hanging surrogate and bottle holder cage arrangement used during their first 3 mo in the SPF program.

Virus testing.

All SPF baboons underwent rigorous serologic and PCR testing for 12 target viruses (Table 1). Serologic testing by ELISA for all herpesviruses, SV40, and SFV were performed as described.³³ Although chacma baboons (P. cynocephalus ursinus) are known to be infected with baboon polyoma virus 1 (SA12), we have not detected this specific virus in the conventional colony of yellow and olive baboons. SV40 has been detected in the colony, and there is some evidence that another related polyoma virus may also be present in some animals.⁴² Because SA12 and SV40 are very closely related, SV40 antigen was used to test for polyoma virus infections. SIV, STLV, and SRV testing was performed at a reference laboratory (Pathogen Detection Laboratory, California National Primate Research Center, University of California at Davis, Davis, CA). Serologic tests were performed to screen the samples, and those testing positive or equivocal by serology were confirmed by using Western blot. Additional serologic testing was performed by commercial laboratories to independently confirm some inhouse results.^24,26,50

Table 1.

Viruses targeted for exclusion in our SPF program

Virus group	Baboon virus	Prevalence in adult (>3 y) baboons at our facility (%)9,32	Human homolog
Herpesvirus	BaCMV	100	Human CMV
	Baboon rhabdinovirus	100	Human herpesvirus 8 (also called Kaposi Sarcoma herpesvirus)
	HVP1	95	EBV
	HVP2	93	Herpes simplex virus 1 and 2
	SVV	72	VZV
Retrovirus	SFV	100	none
	STLV	67	HTLV1
	SIV	11	HIV
	SRV	5	none
Papovavirus	SV40	85	BK and JC viruses
Paramyxovirus	Measles	0	Measles
Poxvirus	Monkeypox, Vaccinia	0	Smallpox, Vaccinia

SPF baboons could find use in studies involving various human vaccine and bioterrorism agents. Therefore, the colony also was tested for measles and monkeypox despite the fact that SPF animals were not vaccinated against measles, and no poxvirus was known to be present in the conventional colony. Vaccinia virus was used to test for monkeypox, because the 2 viruses exhibit substantial antigenic crossreactivity.²¹ These ELISA tests were developed by using detergent extracts of infected cells as described.³²

Turnaround time for most virus testing was very short. All herpesvirus, SFV, SV40, measles, and poxvirus tests were inhouse assays, and serum samples were tested simultaneously for these 9 viruses. Inhouse testing generally was performed within 3 to 10 d of collecting a sample. When needed, samples could be tested within 1 to 3 d of collection. Results for serum samples submitted to offsite laboratories ranged from 3 to 21 d. Because viruses included in the inhouse ELISA test panel included those that were most problematic, the rapid turnaround time for inhouse ELISA testing permitted cull–retain decisions to be made in a timely manner. Any serum sample testing positive for a target virus was retested immediately to confirm the test results and, if still positive, the animal was removed immediately from the program. In cases where the test results were equivocal, an additional serum sample was collected 1 to 2 wk later for retesting. In addition, all baboons with equivocal test results and cohorts of positive animals were retested. Because the use of additional assays for confirmatory testing and redrawing of additional serum samples for retesting required additional time, any baboon with a positive test or consecutive equivocal test results was removed from the SPF program. All animals sold from the program were tested prior to sale to confirm their virus-free status.

Results

In years 1 and 2 of the baboon SPF program, large peer groups (6 to 8) were used, and infants were tested for viruses at 2, 4, 6, 9, and 12 mo of age. Both the group size and test schedule proved untenable, requiring changes in subsequent years. The size of peer groups used to house infants until 1 y of age proved to be a critical factor in the development of the SPF colony. Initially 6 to 8 age-matched conspecifics were housed in separate rooms until 1 y of age, with the intent of decreasing abnormal behavior by providing extensive peer contact. However, we found that infants more than about 3 wk apart in age did not get along well, with older animals being too aggressive toward younger ones.

In addition, in year 1 we also found that when virus breaks did occur, they rapidly spread to involve an entire peer group. Of the 15 female and 10 male baboons recruited into the program in year 1, only 8 (one group, 32%) remained in the program. One infant died of pneumonia of unknown etiology, and 13 were lost due to BaCMV breaks in 2 peer groups. In one case, all 6 animals in the peer group showed decreasing levels of maternal antiBaCMV antibodies at 2 and 4 mo of age, but at 6 mo antiBaCMV titers had risen and increased further at 9 mo of age. In the second group, 2 of 7 infants showed an initial slight increase in BaCMV IgG levels between 2 to 4 mo of age, followed by a substantial increase in titers at 6 mo of age. Therefore, in each peer group, maternal antibodies had begun to wane but had not disappeared completely prior to the outbreak of BaCMV infection.

Losses in year 2 of the program were similar. One infant developed seizures, 3 died (one of HVP2 pneumonia presumably contracted at birth⁵¹), and 14 were culled due to breaks in virus-free status. In one outbreak of HVP1, virus spread rapidly from one initial infant to all of its cagemates, resulting in loss of the entire group of 8 infants. As a result, only 9 of 27 (33%) of baboons recruited during year 2 ultimately were retained in the SPF program.

As a result of the high losses experienced in years 1 to 2 of the baboon SPF program due to rapid spread of virus infections within peer groups, several changes were made to the program. First, the age difference of infants within groups was kept to less than 2 wk. This change was successful in decreasing aggression among cagemates. Second, the virus testing schedule and cull approach was changed in year 3 of the program to allow earlier identification of virus breaks. All infants were bled at 1, 2, 3, 4, 6, 9, and 12 mo of age, with initial tests being run at 3 mo of age. At each subsequent testing, the current and prior 1 to 2 serum samples were tested together. This pattern allowed detection of nondecay of maternal antibody levels (1, 2, and 3 mo sera) for earlier identification of virus outbreaks and provided a means of detecting any slight increases in reactivity between successive bleeds that might be indicative of an animal in the early stages of infection. In addition, whenever an infant developed even a slight increase in reactivity to any virus, they were bled 1 wk later and retested. Any animals positive on a retest immediately were culled from the program and their remaining cagemates tested monthly for 3 mo to ensure that they had not acquired the infection.

A third change made in year 3 of the program regarded housing. Infants were kept in groups of 3 to 4 until 6 mo of age. In addition, the decision was made to cull all infants in a peer group in which an animal had become infected with virus. With these changes, 10 of the 19 (53%) baboons enrolled in year 3 of the program were retained, with 9 being culled due to virus breaks. In year 4 of the program cagemates of infants that developed viral infections again were culled as a preventative measure, resulting in 14 of 20 (70%) of enrolled baboons being culled. Because BaCMV and HVP1 (due to their efficient spread among peers) were the 2 primary causes of program losses, in year 5 cagemates of infants that became infected with viruses other than BaCMV or HVP1 were retained and received monthly serologic testing; all cagemates of animals that developed BaCMV or HVP1 infections were still culled from the program. As a result in years 5 and 6, losses of baboons younger than 1 y due to breaks in virus status were reduced to 24% and 33%, respectively. Overall, 67% of baboons culled from the SPF program were less than 1 y of age, and 68% of these losses were due to breaks in virus-free status (Table 2).

Table 2.

Causes of losses of baboons from SPF program

Age (y)	Virus	Seizures	Death	Other	Total
<1	34	3	9	4	50
1–2	12	0	2	1	15
2–3	5	0	0	0	5
>3	3	1	1	1	6
All	54	4	12	6	76

As SPF baboons reached 1 y of age, larger peer groups were formed in gang cages. As the program progressed, these groups were developed into multiaged harem groups. At present (7 y into the program), there are 5 groups consisting of 1 adult male and 5 to 7 adult female baboons as well as infants and juveniles (Table 3). Animal losses due to virus breaks continued to occur in groups of older animals, although very infrequently. The majority of these breaks have been due to HVP1.

Table 3.

Summary of the SPF baboon colony status

Year	No. enrolled	No. removed	No. sold	SPF colony census (as of December 2009)
				Age (y)	No. of males	No. of females	Total
2001	25	17	1	6–7	1	6	7
2002	27	18	4	5–6	2	3	5
2003	19	9	4	4–5	0	6	6
2004	20	14	1	3–4	0	5	5
2005	21	5	9	2–3	0	7	7
2006	27	9	5	1–2	5	8	13
2007	32	4	0	<1	12	16	28
SPF born	17	4	0	<1	8	5	13
Total	188	80	24		28	56	84

The behavioral development of SPF baboons was a concern, given that they were removed from their dams soon after birth and were hand-reared. Virtually all infants exhibited abnormal clinging behavior when young (Figure 2), but most lost this behavior at 3.5 to 4 y of age. Infants reared in age-matched peer groups of more than 2 individuals reacted more normally when introduced to other baboons or new situations than did individually or pair-raised baboons. Species-typical behaviors such as grooming have been observed to occur as frequently in SPF animals as in conventional baboons, and stereotypies and self-directed injurious behaviors have been uncommon. As the SPF baboons housed in multiaged harem groups matured, normal mounting and presentation behavior occurred. To date, 17 infants have been born in the colony to SPF dams. One female baboon had 3 infants die as a result of trauma, and she was removed from the program. All other dams have been attentive mothers and have shown typical maternal care toward their infants. In addition, younger troop members display normal interest in new infants. Behaviorally, the SPF baboons appear to be doing quite well.

Figure 2.

Four 9-mo-old nursery-reared baboons exhibiting abnormal clinging behavior. This behavior disappeared by 3.5 to 4 y of age.

Even before initial derivation of this colony, a number of inquiries had been received regarding the possible availability of SPF baboons. Despite the newness of the SPF program and the small size of the colony, we were able to partially meet this need through sale of animals from the program (Table 3). However, such sales were selective in that 3 specific groups of animals were sold. First, animals culled from the program due to seroconversion to 1 to 2 viruses were made available to researchers. Second, animals in the SPF colony that became problematic through rising nonspecific serologic immunoreactivity were sold from the colony. And third, excess male SPF baboons were sold. This approach allowed us to partially meet needs of the research community while maintaining overall growth of the SPF colony.

Discussion

Attempting to generate an SPF primate colony free of 12 viruses has not been attempted previously to our knowledge. The rapidity with which some of the target viruses can be transmitted among peers resulted in considerable losses to the program during the first 2 y. Modification of the program to include a combination of 1) caging of infants in peer groups of 3 or 4 animals, 2) frequent serologic testing until 6 mo of age, 3) testing the previous serum sample alongside each new serum sample to detect increases in reactivity, 4) biasing screening assay cutoffs to pick up all true positives (that is, acceptance of some false positives), and 5) aggressive culling of animals testing positive proved to be the best approach for deriving baboons free of multiple target viruses. Although this protocol undoubtedly contributed to a higher-than-necessary cull rate due to removal of animals with equivocal or false-positive test results, this excess was considered acceptable in order to be confident that the colony was indeed free of the target viruses.

Given the age of acquisition of the target viruses in our conventional baboon breeding colony, it was evident that whereas some viruses were acquired later in life, others were very efficiently transmitted to infants younger than 1 y.³³ Consistent with these observations, the viruses most commonly associated with breaks in virus status of baboons less than 1 y of age were BaCMV and HVP1, followed by SFV (Figure 3). By the time a baboon tested positive for one of these viruses and a confirmatory test was completed, the majority of their cagemates had also become infected. Because strict barrier principles were used to maintain isolation of the SPF colony from conventional baboons, the index cases most probably were infected in utero or during the birthing process. Many outbreaks occurred at 3 to 6 mo of age, when maternal antibodies begin to wane. Whether the infection in index cases was maintained as a low level chronic infection by the presence of maternal antibodies or the virus was latent and reactivated at 3 to 6 mo of age when maternal antibodies waned remains an open question.

Figure 3.

Viruses responsible for removal of animals from our SPF program. The number of culls for each virus is shown by age group. The total number of culls due to each virus are indicated by the numbers in parentheses in the key.

Most other viruses, including 3 herpesviruses (SVV, HVP2, and baboon rhadinovirus), 3 retroviruses (STLV, SIV, and SRV), and polyoma viruses (SV40), were less of a problem than expected. Single animals were culled due to infection with most of these viruses, with baboon rhadinovirus being the only one of these viruses that caused an outbreak (an infection that spread to several additional animals). This pattern is consistent with baboon rhadinovirus being efficiently spread to baboons less than 1 y of age in the conventional colony.³³ Although varicella zoster virus is spread very efficiently among humans by the respiratory route,⁴⁶ the same does not appear to be true for SVV in baboons. We have noted sporadic seroconversion to SVV in the conventional colony but no cluster of cases in infant or juvenile baboons, unlike what is typical of varicella zoster virus epidemics. In addition, although SVV can cause very serious infections in other primate species,^13,17,39 we have never observed clinically apparent SVV infections in our conventional baboon colony. Furthermore, because SVV likely is transmitted through inhalation as is varicella zoster virus and because the incidence of SVV in the conventional colony is low, it is unlikely that an infant would acquire SVV during or immediately after birth. Therefore, SVV does not appear to be spread efficiently among baboons, and the lack of SVV outbreaks in the SPF program is not unexpected. The remaining viruses (HVP2, STLV, SIV, SRV, and SV40) all typically require intimate contact for transmission, so the lack of outbreaks due to these viruses is again not surprising. Positive retroviral antibody results were not confirmed by PCR or virus isolation. Therefore it is possible that these early test results were false positives due to cross reactivity with endogenous retroviruses.

Two of the viruses selected for exclusion in the SPF program, measles and monkeypox, are absent in adults of our conventional colony according to current serologic testing (Table 1) and generally are not included in other SPF programs. However, we included these viruses due to previous recognition of infections in baboons by these viruses, the seriousness of disease they can produce in baboons, their potential to cause zoonotic infections, and the possible use of SPF animals in research studies involving one of these agents. Sporadic measles infections in conventionally housed baboons and seropositive wild adult baboons have been reported.^8,22 In addition, baboons can be infected experimentally with measles,⁴⁰ and fatal infection of other nonhuman primate species has been reported, with mortality rates of up to 100%.^4,27 Measles virus is a human pathogen and is believed to be passed to captive nonhuman primates by infected humans;^28,41 this scenario most likely explains why measles virus has not been a problem during derivation of the SPF colony. Monkeypox virus has been shown to infect baboons,^15,16 and infection of other nonhuman primate species has been demonstrated to be fatal.³ In addition, monkeypox virus expresses proteins that can affect the normal function of the immune system.⁴⁷ Taking these observations into account along with the outbreak of monkeypox in the United States in 2003,³⁶ we decided to include monkeypox as one of our targeted agents for routine screening and removal.

Although most culling due to virus breaks occurred in animals younger than 1 y, some animals were culled at older ages. A few of these baboons developed high IgG titers to HVP1, indicative of an active virus infection. We suspect that these animals may have been infected in utero, during birth, or within their first few hours while still with the dam and that maternal antibodies suppressed the infection although latent infections were established. Reactivation of HVP1 at a later point in a seronegative animal would explain these late-age breaks. However, most of the culling of animals older than 1 y were due to equivocal test results. We noted that some animals exhibit higher background levels of reactivity than do others, and some baboons tended to exhibit a gradual increase in nonspecific reactivity to several viral antigens over time. ELISA optical density values in these animals never reached levels typical of truly infected baboons but did reach level at which they exceeded the equivocal range to a point where they were culled from the program as a precautionary measure. Because this reactivity involved multiple antigens, we suspect that it reflected nonspecific reactivity rather than being indicative of actual viral infections. That one infant born in the colony exhibited this equivocal reactivity to multiple viral antigens when first tested at 3 mo of age and that this reactivity has persisted to the present (now 1 y old) seems to support this reasoning. We do not have a viable explanation for what causes this phenomenon.

Investigators and laboratory animal programs using baboons frequently request baboons that are free of certain gastrointestinal parasites. Specifically, Trichuris trichuria, Strongyloides stercoralis, and Strongyloides fulleborni are enzootic in our outdoor-housed conventional baboon colony. These parasites were not originally on our list of agents targeted for exclusion from the SPF colony but have recently been included. Fortunately, testing has failed to detect either species of Strongyloides in the SPF baboons. This parasite seems to have been eliminated by the SPF derivation process. However, Trichuris is present within the SPF baboon colony. We previously showed that treatment of baboons with fenbendazole or fenbendazole-containing primate diet can eliminate Trichuris from single- and group-housed baboons.^37,38 Accordingly, we are presently working to develop a program to eradicate this parasite from the SPF colony.

A primary goal for our SPF program was to have the first-generation hand-reared infants become progenitors of a self-sustaining SPF breeding colony and to self-raise their offspring; therefore, we wanted their behavior to be as species-typical as possible. Studies have shown that monkeys reared individually or in small groups exhibit more abnormal behaviors than do those raised in larger groups.⁴³ Our previous experience with hand-raising baboons indicated that the only thing worse than raising a single baboon independently from conspecifics was raising 2 baboons apart from conspecifics. In that situation, the 2 infants fixate on each other, exhibiting abnormal clinging behavior and failing to interact normally with other conspecifics when the group size is increased. For this reason, we initially used peer groups of 6 to 8 infants. However, this practice resulted in considerable losses to the program due to rapid spread of viral infections within a group. Ultimately, the use of smaller groups of 3 to 4 infants proved optimal for retention of recruits through their first year while maintaining relatively species-typical behavior, as evidenced by our collective increased retention rates achieved during years 3 to 6 compared with years 1 and 2.

Aside from the problems encountered with pair-caging of infants, we have not encountered noteworthy problems regarding abnormal behavior of baboons recruited into the SPF program. Although some infants do exhibit excessive clinging to peers early in life, this behavior disappears with increasing age when infant groups of 3 to 4 are used. No baboons have had to be culled from the program due to bad behavior toward peers, and only one adult female baboon has been removed from the program due to poor maternal skills. Because the SPF colony is still in the early stages of development, as a precautionary measure, the colony has been organized into small harem groups rather than a single large, multimale group. This approach has proven successful, in that several infants have been born into the SPF program and, with only one exception, the dams have provided excellent maternal care. The behavior of other members of the troops toward new infants has reflected that of baboons in the conventional breeding colony. Therefore, despite hand-rearing of first-generation SPF animals, the adult baboons are able to breed and raise their own infants.