Lessons from Chimpanzee-based Research on Human Disease: The Implications of Genetic Differences
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
ATLA 39, 527–540, 2011 527
Lessons from Chimpanzee-based Research on Human
Disease: The Implications of Genetic Differences
Jarrod Bailey
New England Anti-Vivisection Society (NEAVS), Boston, USA and British Union for the Abolition of
Vivisection (BUAV), London, UK
Summary — Assertions that the use of chimpanzees to investigate human diseases is valid scientifically are
frequently based on a reported 98–99% genetic similarity between the species. Critical analyses of the rel-
evance of chimpanzee studies to human biology, however, indicate that this genetic similarity does not
result in sufficient physiological similarity for the chimpanzee to constitute a good model for research, and
furthermore, that chimpanzee data do not translate well to progress in clinical practice for humans.
Leading examples include the minimal citations of chimpanzee research that is relevant to human medi-
cine, the highly different pathology of HIV/AIDS and hepatitis C virus infection in the two species, the lack
of correlation in the efficacy of vaccines and treatments between chimpanzees and humans, and the fact
that chimpanzees are not useful for research on human cancer. The major molecular differences underly-
ing these inter-species phenotypic disparities have been revealed by comparative genomics and molecular
biology — there are key differences in all aspects of gene expression and protein function, from chromo-
some and chromatin structure to post-translational modification. The collective effects of these differences
are striking, extensive and widespread, and they show that the superficial similarity between human and
chimpanzee genetic sequences is of little consequence for biomedical research. The extrapolation of bio-
medical data from the chimpanzee to the human is therefore highly unreliable, and the use of the chim-
panzee must be considered of little value, particularly given the breadth and potential of alternative
methods of enquiry that are currently available to science.
Key words: animal model, chimpanzee, genetics, Pan troglodytes.
Address for correspondence: Jarrod Bailey, NEAVS, 333 Washington Street, Suite 850 Boston, MA
02108-5100, USA; BUAV, 16a Crane Grove, London N7 8NN, UK.
E-mail: jarrod.bailey@mac.com
Introduction research for the animals involved (5, 6), profound
ethical considerations have always been the foun-
The validity of using chimpanzees (Pan dation for strong opposition to chimpanzee
troglodytes) in research to investigate the diseases research (7, 8). In addition, in recent years a bur-
that affect humans is frequently based on the geoning scientific case has been made against the
reported genetic similarity of the two species. It is use of chimpanzees in research, which seriously
often noted that chimpanzees are 98–99% geneti- questions the usefulness and human relevance of
cally identical to humans — a figure that was ini- chimpanzee data (9–13; and see http://www.
tially derived 40 years ago (1, 2) and was confirmed releasechimps.org/pdfs/chimp-efficacy-paper-main.
more recently by using superior technologies (3). pdf and http://www.releasechimps.org/pdfs/Chimp-
Advocates of experimentation with chimpanzees CA-suppl.pdf). Furthermore, it is apparent that
claim it is axiomatic that this genetic similarity the claimed 98–99% genetic similarity between
leads to equivalent biological and physiological humans and chimpanzees is superficial and decep-
similarity, and that because chimpanzees are the tive. The overall similarity, when all relevant fac-
closest relatives of humans, they must constitute a tors are taken into account, is likely to be nearer to
good — indeed the best — animal species to use for 95% (14, 15), or even approaching 93% (16, 17). As
the study of human disease (4). Thus, the use of elaborated in this paper, important differences at
chimpanzees in biomedical research is justified by all levels of gene expression and protein function,
its proponents, alongside claims of the crucial role from chromosome structure to post-translational
played by such research in fighting many grave modification, combine to generate significant and
human diseases. widespread biological and physiological differ-
However, this justification has become increas- ences.
ingly controversial. Given the cognitive and emo- In recent years, such concerns have led to legis-
tional capacities of chimpanzees, and the harmful lation that bans, or at least very severely restricts,
consequences of laboratory life and invasive chimpanzee experimentation in the relatively few528 J. Bailey
countries where it was ever practised (see http:// Results
www.releasechimps.org/mission/end-chimpanzee-
research/country-bans/). The exception is the USA,
where approximately 1,000 chimpanzees remain in Cytogenetic differences: Fusions, inversions
laboratories (see http://www.releasechimps.org/ and translocations
mission/end-chimpanzee-research/), many of them
the last survivors of an intensive breeding pro- There are numerous well-known structural differ-
gramme for HIV/AIDS research that began in 1986 ences between the human and chimpanzee
(18). The ethical and scientific concerns that have genomes. For example, human chromosome 2
been raised have been reflected in substantial pub- results from a fusion of ancestral chromosomes cor-
lic opposition to chimpanzee experimentation (see responding to chimpanzee chromosomes 12 and 13
http://www.releasechimps.org/mission/end- (28). Compared to chimpanzees, human chromo-
chimpanzee-research/public-opinion/), which has somes 1 and 18, as well as others, contain large
translated into some degree of legal protection: the inverted stretches, and many human chromosomes
passage of the Chimpanzee Health, Improvement, contain large translocated or rearranged segments
Maintenance and Protection (CHIMP) Act in 2000 (26). These rearrangements exert a significant
(19, 20), and the more-recent legislative efforts in influence on the expression of the affected genes
both the US House of Representatives and the US via the ‘position effect’, which results from an
Senate in the form of the Great Ape Protection Act alteration in the location of a gene. The relocation
(GAPA) in 2009 (21, 22), which was reintroduced of a gene can influence its expression via changes
in 2011 as the Great Ape Protection and Cost in the proximity and/or nature of cis-acting pro-
Savings Act (GAPCSA; H.R.1513/S.810; 23, 24). moters and enhancers. Furthermore, the local
This act would end invasive experimentation on structural environment of chromatin can alter the
captive chimpanzees in the USA, bringing the accessibility of transcriptional proteins, and can
country into line with the rest of the world, and exert gene-silencing effects via the influence of
would facilitate the retirement of all federally nearby heterochromatic DNA (29). One study
owned chimpanzees into permanent sanctuaries. ‘conservatively estimated’ that 18% of genes —
The aim of the present review is to aid the numbering some 3,200 — have genomic neigh-
debate and discussion surrounding the use of bourhoods that differ between humans and chim-
chimpanzees in US laboratories, by examining panzees, and may therefore be subject to gene
the ‘genetic similarity’ argument in the light of expression changes due to position effects (30).
current knowledge regarding the comparative Another major difference is found between the Y
molecular biology of humans and chimpanzees chromosomes of humans and chimpanzees. The
and their inter-species differences. Salient infor- male-specific region (MSY) of each of these chro-
mation from recent years, along with a consider- mosomes has evolved rapidly over the past six mil-
ation of the performance and human relevance of lion years, and as a result, the chromosomes now
the chimpanzee model, is synthesised, in order to ‘differ radically in sequence, structure and gene
illuminate and critically evaluate this argument. content’ (31). The chimpanzee MSY contains twice
as many massive palindromes as its human coun-
terpart and has lost genes, while the human MSY
Methods has gained genes. X-degenerate regions of the
chimpanzee Y chromosome have lost four out of 16
Papers describing important genetic differences genes, due to inactivating mutations. In total, the
between humans and chimpanzees, which have chimpanzee MSY contains only two-thirds of the
significant or potentially significant functional number of genes/gene families in the human MSY,
biological consequences, were located via the and just 50% of the protein-coding transcription
GoPubMed database (25). Other salient reports units (31).
were located in two exceptional reviews of
human/chimpanzee comparative genomics that
were published in 2001 (26) and 2006 (27). A total Mobile DNA elements and copy number
of 178 papers, dating from the past decade variation
(2001–2010), were examined to form the basis of
a review, the aim of which is to illustrate, with a The position effect can also be invoked by the
sound and comprehensive basis, that there are action of ‘mobile elements’ or ‘transposable ele-
crucial inter-species differences with extensive ments’ in the genome — sections of DNA that can
and far-reaching effects. These differences per- move or ‘transpose’ themselves to other genomic
meate all aspects of gene expression and protein locations. One subclass of these elements can be
function, from chromosome and chromatin struc- classified as long interspersed nuclear elements
ture all the way through to post-translational (LINEs) and short interspersed nuclear elements
modification. (SINEs; 26); these vary between humans andChimpanzee-based research on human disease 529 chimpanzees and cause variations in gene expres- the speciation process of humans and chim- sion between them. Mobile elements comprise panzees, and to have contributed greatly to genetic almost half of the entire human genome (26, 32), variation between the two species (33). A signifi- and significantly affect gene function by physically cant proportion of these inversions, as well as of disrupting genes, by generating alternative sites other chromosomal rearrangements, such as the for the splicing of RNA molecules during gene segmental duplications mentioned earlier, are transcription, or by interfering with gene promoter caused by the actions of the aforementioned mobile and enhancer regions (33). Among other things, (transposable) elements. Together, these elements they are responsible for an as yet undefined pro- have acted dynamically to generate numerous, and portion of large duplicated regions of DNA called often substantial, chromosomal rearrangements, segmental duplications, which in total comprise a which are the basis of many of the genetic differ- significant 5% of the human genome. While ences between humans and chimpanzees, includ- approximately two-thirds of these duplications are ing gene gain and gene loss (33). common to humans and chimpanzees, one-third In 2005, Feuk and colleagues compared the are species-specific and have contributed greatly to genomes of the two species and identified almost the biological differences between humans and 1,600 inverted regions, which are equivalent to chimpanzees (34). approximately 5% of the human genome (39). They These duplications, in a phenomenon known as noted that 151 of these inversions contained either copy number variation, are responsible for signifi- entire or partial genes, or a breakpoint that inter- cant disparities in gene expression and subsequent sected a gene. Furthermore, they found 140 genes biological consequences, such as susceptibility to with human-specific CNDs and 15 genes with disease (35). They are known to be one of the most chimpanzee-specific CNDs, of which some were significant causes of genetic variation among pri- important immunologically. A more-recent study mates (36), and are at the root of many aspects of identified 252 inversion loci between human and intra-species and inter-species diversity, differen- chimpanzee lineages, the junctions of which could tially affecting many human and chimpanzee be easily characterised. Analysis of these junctions genes involved in critical processes, such as directly implicated the aforementioned LINEs and immune and inflammatory responses, cell prolifer- an abundant class of SINEs, known as Alu ele- ation and tumour formation (37). Clearly, copy ments, in the inversion process (33). This study number variation greatly affects the appropriate- also revealed a novel mechanism for some inver- ness of chimpanzee use in the study of human dis- sions that was responsible for 27 human-specific eases. events and for 22 chimpanzee-specific events. In 2006, Bailey and Eichler reported that over 76 These events affected exonic and intronic regions, million base pairs of DNA are differentially dupli- and had an impact on gene function via gene dis- cated between chimpanzees and humans (35). ruption, the generation of alternative splice sites, More recently, Armengol and co-workers examined and effects on gene regulatory regions. There are over 16,000 genes and identified 23 human-specific approximately one million Alu elements in the copy number differences (CNDs) versus chim- human genome with the power to effect these out- panzees, and 26 chimpanzee-specific CNDs versus comes (40), and thereby greatly influence gene humans, with some of the affected genes having expression (41). neuronal functions (36). An earlier analysis Alu elements can also give rise to completely revealed 177 complete and partial gene duplica- new exons, often in existing functional genes, with tions in humans, but not in chimpanzees; the oppo- diverse splicing patterns (40). One such Alu- site was true for another 94 genes (34). Half of the derived exon has been implicated in a form of mus- 271 genes showed differences in expression — in cular dystrophy (42). More generally, Alu elements fact, genes in duplicated regions are often up to 10 have been associated with a number of diseases times more likely to show differences in expression that arise from disrupted gene function, including levels than those in other regions (35). It is various cancers, haemophilia, neurofibromatosis, axiomatic that many genes with important func- type-2 diabetes and Alzheimer’s disease (43). tions would be affected: the multi-organ cancer Interestingly, the chimpanzee genome has 100,000 susceptibility gene CHEK2, for example, is present fewer Alu elements than the human genome, and in nine copies in chimpanzees, but there are 13–16 although 572 evolutionarily young Alu elements copies in a typical human genome (38). have been identified in the human genome, just Inversions of genetic material are also highly 160 such elements have been found in the chim- prevalent, and are different, in the human and panzee genome (43). These great differences speak chimpanzee genomes. They are important because once again to the inappropriateness of equating of their effects on gene expression in the vicinity of research on chimpanzees to investigations based the DNA breakpoints where they occur, as well as on human subjects. their direct interruption of some genes. Such inver- Large insertions and deletions of genetic mate- sions are considered to have been major drivers in rial, or indels, are another source of significant
530 J. Bailey
variation between human and chimpanzee gen- implicated in immunity to viral infection and sur-
omes. Volfovsky and colleagues compared two- veillance for cancer, differs between the two
thirds of human chromosome 21 and its equivalent species (52).
in chimpanzees, chromosome 22 (there is an alter-
native chromosome nomenclature, in which
human and chimpanzee chromosomes are num- Gene expression
bered similarly; based on this alternative nomen-
clature, the comparison involved human and Even in the majority of cases in which genes are
chimpanzee chromosome 22). This comparison common to humans and chimpanzees, there are
revealed 683 indels in the proximity of known substantial and widespread differences in their
human genes, which potentially affected their expression. For example, a study examining the
expression (44). Twenty-three of these indels were expression of around 12,000 genes in the pre-
within exons, or within splice-site regions, and ten frontal cortex of the brain found that almost 1,000
altered the sequence of the corresponding protein. were expressed in the human, but not in the chim-
Indels have also been shown to affect major histo- panzee, while the reverse occurred for 344 genes.
compatibility complex (MHC) genes, which are In addition, of the genes that were expressed in
critical for immune responses. Indeed, indels are both species, 20% showed a different expression
responsible for a large 95kb deletion in the MHC profile — for example, 19 genes linked to
locus that resulted in a virtual fusion of the MICA Alzheimer’s, Parkinson’s and Huntington’s dis-
and MICB human genes into the single hybrid eases in humans were found to be expressed dif-
chimpanzee MIC gene (45). An additional 64 indels ferently in chimpanzees (53). In the cerebral
of 100bp were identified in this locus, which is cortex, at least 169 genes are expressed differently
genetically linked to differences in the handling of — many of which are involved in neuroprotection
various infections, including HIV, hepatitis B and and synaptic transport (54) — and 916 genes are
C viruses, and the malarial parasite, Plasmodium expressed at least two-fold differently in the cere-
falciparum, as well as susceptibility to autoim- bellum (55). Furthermore, many genes involved in
mune diseases. oxidative metabolism and mitochondrial function
are expressed to a higher degree in the human
brain than in the chimpanzee brain (56–58).
Gene complement Not surprisingly, such differences are not con-
fined to the brain. One study analysed an average
Many genes are present in humans but entirely of approximately 10,500 genes in various human
absent in chimpanzees, or vice versa, largely due to and chimpanzee organs, and found that a striking
the duplication and deletion of large genomic 34% showed differential expression in the brain,
regions already noted. It has been reported that, 25% in the liver, 33% in the kidney, 35% in the
since the evolutionary split of humans and chim- heart, and 62% in the testes (59). Another study,
panzees, humans have gained 689 genes and lost which analysed even more genes (17,231), identi-
86, while chimpanzees have gained 26 genes and fied 16.3% of genes in the liver, 19.5% in the kid-
lost 729 that are still present in humans (46). This ney, and 18.5% in the heart, that were expressed
means that humans differ by 6.4% in terms of their at different levels within the two species (57).
gene complement alone (1,418 genes out of 22,000), These percentages amount to many thousands of
which has significant consequences. An example is genes differentially expressed in various organs.
the golgin gene subfamily, which has been linked Notably, differences in the expression of a single
to systemic autoimmune diseases such as lupus gene can have crucially important consequences:
erythematosus (47). In humans, this subfamily the natural cytotoxicity receptor gene (NKp44), for
comprises 49 genes, but only 23 genes in chim- example, shows five-fold higher expression in
panzees. chimpanzees than in humans, and it exhibits dif-
A comparison of protease genes, which encode ferential increases upon activation. Its protein
enzymes important in a wide variety of cellular product is involved in the recognition of HIV-1, and
processes, showed that five genes had been com- in the sensing and killing of virus-infected and
pletely deleted in the human genome, while the cancerous cells. It is thought that these differences
corresponding number of deletions in the chim- could be at least partly responsible for the more
panzee genome was just two (48, 49). Other stud- benign course of HIV infection in chimpanzees
ies have suggested: that 27% of human kinase (60).
genes might not have a close homologue in chim- It is highly likely that species variation in the
panzees (50); that 23% of human genes encoding expression of just one gene, which encodes a mem-
zinc-finger proteins, many of which modulate gene ber of the inhibitory sialic acid-recognising Ig-
transcription, might be absent in chimpanzees superfamily lectins (Siglecs), has a major effect on
(51); and that the presence of genes encoding cell- comparative general immune responses to many
surface receptors of natural-killer cells, which are and varied pathogens. For example, Soto and co-Chimpanzee-based research on human disease 531
workers (61) showed that inhibitory Siglecs gener- ated with genes involved in the development and
ally, but in particular Siglec-5, are expressed at a maintenance of this organ.
much lower level in humans than in chimpanzees. Species differences in transcription factors cre-
The greater abundance of inhibitory Siglecs in ate variations in gene expression, and thus gener-
chimpanzees serves to dampen excessive immune ate physiological changes between humans and
responses relative to humans, leaving human T- chimpanzees. In addition, the role of transcription
lymphocytes and B-lymphocytes, in comparison to factors in differential gene expression is aug-
those of chimpanzees, over reactive to a variety of mented by variations in transcription factor bind-
stimuli. This may explain species differences in ing sites between species. DNA sequence
diseases that involve immunopathology, including variability at transcription factor binding sites is
HIV, hepatitis C, asthma, psoriasis and rheuma- normal between individuals of the same species:
toid arthritis (61). for example, 25% of binding sites for RNA poly-
In addition, terminal sialic acids are produced in merase II vary between individual humans.
larger amounts in the hearts of chimpanzees than However, a statistically significant 32% of these
they are in those of humans (62). As these mole- binding sites were found to have meaningful dif-
cules are frequently the targets of viral pathogens ferences between humans and chimpanzees (67).
that can initiate fibrosis, their greater abundance Differences in conserved non-coding sequences
in chimpanzees makes myocardial fibrosis more flanking genes have been shown to have serious
common in that species (63). Indeed, this type of implications for gene expression, as they may
heart disease is the major cause of cardiac arrest affect three-quarters of human autosomal genes
and progressive heart failure in chimpanzees, but (75.4% in one particular study; 68).
it is very different from the main cause of heart As an example of species-specific variability in
disease in humans, which is coronary artery ather- the influence of neighbouring non-coding regions, a
osclerosis (63). In common with the apparent comparative analysis of 100 representative DNA
genetic factors underlying a greater incidence of repair-associated genes in humans and chim-
myocardial fibrosis in chimpanzees, it seems that panzees revealed that 13% and 32%, respectively,
the increased incidence of atherosclerosis in exhibited evidence of accelerated evolution in their
humans might also have a genetic cause — the promoter and intronic regions (69). Some of those
lack of Neu5Gc sialic acid (described further in the genes also showed inter-species differences in
Differences in Genetic Sequences section). A theory expression, which are thought to affect DNA repair
is that humans ingest Neu5Gc sialic acid (as part mechanisms. A faster rate of DNA repair in chim-
of their diet), and that this ‘foreign’ molecule trig- panzee cells than in human cells, might contribute
gers an immune response, causing inflammation to the different incidences of cancer between the
and exacerbating atherosclerosis. This chain of species.
events is augmented by the increased reactivity of Gene expression is also affected by DNA methy-
human T-cells caused by their lower expression of lation patterns, which act epigenetically through
inhibitory Siglecs, as noted above (61). In sum- histone modification and chromatin remodelling to
mary, it appears that a few differences in genetic mediate transcription (70). In common with all
sequence and gene expression may be responsible alterations to cis-acting sequences, minor varia-
for very different pathologies of heart disease in tions, or differences, in DNA methylation can
humans and chimpanzees (63). affect the expression of many genes. It has been
discovered that over 12% of methylation sites, cov-
ering one-third of all genes, might be differentially
Factors affecting gene expression methylated in humans and chimpanzees (71).
There are major differences in other factors that
It is significant that many of the genes that are dif- control gene expression. For example, in recent
ferentially expressed, often at a higher level in years microRNAs — small, 21–22 nucleotide RNA
humans, encode transcription factors, as these pro- molecules that can negatively regulate gene
teins can affect the regulation of many hundreds of expression by interfering with mRNA molecules,
genes and can result in significant phenotypic either causing them to degrade or by blocking tran-
effects (64). Despite their understandably strict scription — have received increasing attention (72,
evolutionary conservation, many transcription fac- 73). There are thought to be over 1,000 miRNAs,
tors have evolved under directional selection in each of which can repress the expression of hun-
humans (57, 65). A study by Nowick et al. illus- dreds of genes in highly complex regulatory net-
trates this elegantly, as they identified 90 tran- works (74). They play crucial roles in cell
scription factor genes with significantly different development, e.g. in signalling and differentiation,
expression levels in human and chimpanzee brains and in cell fate, e.g. in apoptosis. As miRNAs regu-
(66). These gene networks were enriched for pri- late many oncogenes, perturbation of their func-
mate-specific KRAB-ZNF genes, which are central tion can induce the initiation and progression of
to human and chimpanzee brains and are associ- tumours (75). Interestingly, comparisons of human532 J. Bailey
and chimpanzee precursor miRNAs have shown throughout evolution. However, 20 of the 333
that only 60% are identical, and that just 73% genes, some of which are definitively involved in
show no change in their secondary structure (76). tumour formation, had some level of difference.
In addition, around 10% of mature miRNAs are This may help to explain why chimpanzees have
different to some degree, and there are at least 39 an extraordinarily low incidence of cancers, espe-
miRNAs that are present in the human but not in cially those of the breast, prostate and lung, can-
the chimpanzee (75). As a result of the fast evolu- cers that account for over 20% of human deaths in
tion of miRNA binding sites (76), there are modern populations (12). Another analysis, this
undoubtedly species differences in the expression one of human and chimpanzee protease genes
levels of miRNAs (and in the selection and expres- (which are involved in many essential biological
sion of target genes). At least 10% of miRNAs in processes), revealed important differences: a high
embryonic stem cells, which can differentiate into degree of similarity was discovered between the
over 200 different types of cell during develop- 559 chimpanzee and 561 human protease genes,
ment, differ between humans and chimpanzees but seven genes were deleted entirely (i.e. two
(77). genes present in chimpanzees were absent in
Adenosine-to-inosine RNA editing, another humans, and the opposite was true for five genes;
important factor that affects gene expression, is 48). Many other genes contained small insertions,
catalysed by an enzyme that post-transcriptionally deletions or sequence changes that can lead to
converts adenosine to inosine in RNA molecules, gene inactivation, and several of these genes were
resulting in RNA sequences in which inosines, in involved in the function of the immune system
place of the genomically-encoded adenosines, are (48).
read as guanosine residues by the cell’s transla- Another study identified 54 genes in humans
tional and splicing machinery. This process has and 162 genes in chimpanzees that had undergone
obvious consequences for any gene that is affected, positive selection since the evolutionary split
but importantly the process of editing is consider- between the species. Just 17 genes had been posi-
ably more prominent in humans than in chim- tively selected in both species (80). Notably, the
panzees (78). In humans, it occurs predominantly genes positively selected in humans have been
in the previously described Alu elements, and it implicated in epithelial cancers, schizophrenia and
affects myriad loci across tens of thousands of other cognitive disorders, ataxia and migraine,
genes. When Alu elements in non-coding regions autoimmune diseases such as lupus and rheuma-
are edited, gene expression can be altered; when toid arthritis, and Alzheimer’s disease, all of which
coding regions are edited, the amino acid differ in prevalence and symptoms between
sequences, properties and functions of the encoded humans and chimpanzees (80). The positively
proteins are changed. Because the brain is most selected genes included those involved in: DNA
affected in this way, many of the targets of editing repair, general tumour progression, gastric and
are associated with the genes involved in neuroge- breast cancers, susceptibility to and progression of
nesis. Altered editing is therefore associated with a melanoma, neurological development, various
number of neuropathological disorders. The differ- ataxias, migraine, dystonia, epilepsy, bipolar dis-
ential editing between humans and chimpanzees order and cognitive impairment. Differences in
has distinct effects on gene expression and func- transcription factor genes were also found, includ-
tion via alternative splicing and differences in ing the HIVEP3 gene, which activates gene expres-
mRNA stability, nuclear retention and miRNA bio- sion in HIV and which may account for some of the
genesis and targeting (78). species differences in HIV/AIDS, and other genes
involved in regulating gene expression such as
MOV10, which is involved in the silencing of
Differences in genetic sequences mRNA molecules.
Although Volfovsky et al. (44), in the study pre-
In addition to the large-scale genomic rearrange- viously reported, identified 683 large indels
ments, such as inversions, indels, duplications, between chimpanzee chromosome 22 and its
and so forth already discussed, as well as the dis- human counterpart chromosome 21, an earlier,
parities in gene complement, there are smaller more-detailed analysis of these chromosomes
apparent differences that can have a considerable revealed a total of around 68,000 indels of all sizes,
impact on human and chimpanzee physiology and resulting in genetic differences in the 231 genes
that are responsible for notable inter-species dif- therein, leading to differences between humans
ferences. and chimpanzees in 83% of the protein products
A recent analysis of chimpanzee genes ‘equiva- (81). In addition, around 1 in 10 of these genes
lent’ to 333 human genes implicated in cancer (79) showed significant differences in their expression
found that the genes shared a high degree of simi- levels. Another study found that about 8%
larity. This would be expected, because genes con- (1,109/13,487) of the human/chimpanzee gene
trolling cell growth are often tightly conserved pairs analysed were affected by premature stopChimpanzee-based research on human disease 533 codons in the chimpanzee version, resulting in mutation, is further reflected in the array of truncated mRNA transcriptional products that are human diseases with which they are associated, either degraded, or that give rise to truncated pro- including: cancer, arthritis, neurodegenerative and teins with altered or ablated functions (17). cardiovascular disorders, haemophilia, Alzheimer’s Elsewhere, an analysis of all human and chim- disease, hereditary pancreatitis, and progeroid panzee exons showed that humans have 1,931 dis- syndromes (49). rupted exons (there are more than 500,000 in A deletion of two base pairs in the human total), while the chimpanzee has 3,742 disrupted MYH16 gene resulted in a frameshift that reduced exons (82). In humans, functional consequences of the sizes of jaw muscles in humans and allowed a these disruptions identified to date include larger brain to develop (85). A single base pair sub- melanoma, breast cancer, neuroblastoma, some stitution caused an inactivating, premature stop inflammatory diseases, problems of immune and codon in the human type I hair keratin gene, which cardiovascular development, lupus and stress resulted in smooth, hairless skin in humans, in responses. contrast to the very hairy skin of chimpanzees (86). Specific examples of small genetic changes with Sequence differences in the FOXP2 gene, resulting major, or at least potentially major, consequences in just two amino acids that are different in the are commonplace. For example, the CMP-sialic protein product, have been linked with a species acid hydroxylase gene in humans contains a small difference of almost immeasurable magnitude and deletion of a 92bp exon that is not found in the consequence — the acquisition of human language chimpanzee gene. This deletion causes a frame- (87). shift that results in loss of function of the protein. While inter-species sequence disparities for any Deficiency of this enzyme activity means that particular gene may have obvious direct conse- humans, unlike chimpanzees, do not have a partic- quences due to the potential variability of that ular sialic acid residue (N-glycolyl-neuraminic gene’s protein products and their immediate func- acid, or Neu5Gc) on the surface of almost all of the tional activities, it must also be appreciated that cells in their bodies. The absence of this residue indirect effects may be significant as well. The has implications for intercellular interactions and aforementioned FOXP2 gene, for example, in addi- many other biological processes, as well as for sus- tion to its own structural and functional differ- ceptibility (and resistance) to microbial pathogens ences between humans and chimpanzees, (83). The CDR4 gene, which encodes the CDR4 cell differentially and significantly affects the expres- receptor that is a major part of the receptor com- sion of a further 116 genes. In humans, but not in plex for HIV and thus its route of infection of chimpanzees, 61 genes are up-regulated and immune cells, differs between humans and chim- another 55 are down-regulated by the FOXP2 pro- panzees significantly. This difference affects CDR4 tein (88). The genes involved are important for regulation, expression, density on the cell surface brain development and function (for example, and glycosylation — factors that influence its those involved in craniofacial formation and in structure, stability, and ability to interact with establishing the neural circuitry and physical other immune cells. These small genetic differ- structures needed for spoken language via cerebel- ences between humans and chimpanzees therefore lar motor function), and in the formation of carti- have serious consequences for HIV infection and lage and connective tissue. pathology, and underlie some of the differences seen in HIV/AIDS between the species (84). The differences in the genomic complement of Differences in RNA splicing protease genes (48), which were discussed earlier, may be significant, despite the overall similarity of Splicing factors are complexes of proteins and RNA the human and chimpanzee degradomes and their that bind to specific sequences in pre-mRNA inter- constituent genes. Important differences also exist mediates during transcription, and splice out in protease genes that are common to the two intronic sequences to generate multiple mRNA species, particularly in those genes related to host molecules from one gene. Their presence, nature, defence, such as neutrophil granule proteases and function and repertoire, therefore have far-reach- processors of pro-inflammatory cytokines (49). ing consequences for the expression of many other These differences are notable, because of the wide genes. They greatly increase the complexity of the variety of critical functions in which the enzymes genome and proteome, and over 80% of human are involved, such as: cell cycle progression; cellu- genes are subject to their action (89). It is critical lar proliferation, differentiation and migration; to note that many splicing factors are differentially embryonic development; tissue remodelling; angio- expressed in humans and chimpanzees; for exam- genesis; apoptosis, autophagy and senescence; fer- ple, 43 in the testes and 20 in the brain (90). This tilisation; immunity; wound healing; and alone will result in many protein variants, which haemostasis (49). The importance of proteases, and may have distinct functions in the tissues and of the effects of their differential regulation or organs of humans and chimpanzees. As previously
534 J. Bailey
described, the inter-species consequences of differ- ceived and actual importance of chimpanzee
ential alternative splicing are further amplified by research, amid claims from its advocates that it is
the differential action of mobile elements and crucial and takes place only when absolutely nec-
adenosine-inosine RNA editing in generating and essary (as expressed, for example, by VandeBerg
deleting alternative splice sites (33, 40, 44, 78). and Zola [4]). If these claims were actually well-
founded, papers reporting chimpanzee research
should be heavily cited in the literature describing
Post-translational differences advances in human medicine, but the 2007 analy-
sis found this not to be the case. By using a statis-
The genomic and transcriptomic differences tically representative sample of almost 100
described earlier obviously translate into pro- published papers, the analysis revealed that more
teomic differences, with consequent inter-species than 85% of the chimpanzee studies had not been
diversity in protein function, biochemistry and cited at all in papers relevant to human medicine.
physiology. In addition, there are comparative pro- Furthermore, a thorough and detailed analysis of
teomic studies that directly inform our knowledge the papers cited in a human medical context, and
of human/chimpanzee differences at the protein of those that had cited them, revealed that the
level. For example, 80% of orthologous proteins chimpanzee studies had contributed little, if any-
were found to differ in their amino acid sequences thing, to the outcome of the human studies. Often,
(91) by an average of two amino acids per protein the results from research involving chimpanzees
(92), which has clear implications for evaluating were duplications of human outcomes that had
the applicability of data from chimpanzees. already been reported in earlier papers, or the con-
Furthermore, a comparative analysis of proteins clusions were inconsistent with data on humans or
that interact with HIV-1 revealed that, of 1,447 other non-human primates. In contrast, the analy-
such human proteins, 77 had no orthologue in non- sis found it demonstrable that in vitro research,
humans, including chimpanzees (93). The same human clinical and epidemiological investigations,
study revealed significant differences between and molecular and genomic methods had con-
humans and chimpanzees in the group of tributed most to the findings of the 95 papers in
APOBEC3 proteins, which are involved in host the sample.
defence against retroviruses. Of the 1,370 human A review that examined the use of chimpanzees
and chimpanzee orthologous proteins that were in HIV/AIDS research (11) found that the number
compared, each species had more than 600 species- of AIDS-related chimpanzee studies fell by nearly
specific phosphorylation sites, potentially affecting 90% between 1998 and 2005, apparently because of
thousands of protein–protein interactions, enzyme the poor clinical translation of, and need for, that
activities and protein stabilities (93). line of research. Responses to HIV vaccines were
highly discordant in humans and chimpanzees, so
the review concluded that responses to vaccination
The consequences of genetic differences: in chimpanzees could not be considered predictive
Evaluation of the translation of data from of responses in humans. In fact, according to the
chimpanzee research to humans same review, 85 diverse HIV vaccines that were
developed by using chimpanzees and other pri-
Several critical analyses have reflected serious mates and shown to have protective, or therapeu-
concerns about the human relevance of chim- tic effects, were evaluated in 197 human clinical
panzee research and the capacity of chimpanzee trials. Yet, protection and/or significant therapeu-
data to produce tangible clinical benefits. While tic effects in humans have still not been demon-
each of these representative studies set out with strated by any HIV vaccine, despite decades of
the hypothesis that chimpanzee experimentation effort and many millions of dollars of research
was not a productive, or necessary, research funding. Thus, contrary to claims that chim-
methodology, their conclusions are supported by panzees “are still important for testing vaccines
substantial data. Furthermore, it must be noted aimed at preventing HIV-1 infection or reducing
that each study provides numerous peer-reviewed the virus load in infected individuals”, for example
publications, as well as professional opinion and (4), it was concluded that neither claim has any sci-
comment to substantiate its conclusions. Caveats entific foundation, and that a return to the use of
surrounding the efficacy of chimpanzee research chimpanzees in AIDS research/vaccine develop-
are indeed widespread, and they emanate from ment would be without scientific justification.
many and varied qualified sources. Chimpanzees have scarcely been used in any
In 2007, a citation analysis by Knight (13) and form of cancer research, which may be surprising
Bailey and colleagues (see http://www.release when one considers that cancer is one of the great-
chimps.org/pdfs/chimp-efficacy-paper-main.pdf est killers of human beings. A review of the subject
and http://www.releasechimps.org/pdfs/Chimp-CA- showed that there have been very few scientific
suppl.pdf) provided a general overview of the per- publications relevant to human cancer that usedChimpanzee-based research on human disease 535 chimpanzees, and that tumours in chimpanzees area of hepatitis C research, the arguments are are extremely rare, especially of the types that are strongly against a scientifically based requirement responsible for most of the deaths from cancer for the use of chimpanzees and very much in among humans (12). The review found that chim- favour of concentrating research in human-specific panzee tumours are biologically different from clinical and in vitro technologies. those of humans, and this includes the tumour ini- tiation and progression phases. These observations demonstrate that, despite their overall apparent Discussion and Conclusions genetic similarity to humans, chimpanzees consti- tute a poor research model for human cancer. The conduct of invasive research on captive chim- Furthermore, these animals are not essential for panzees may never have been more controversial the development of therapeutic monoclonal anti- than it is today. In recent years, experimentation on bodies for cancer treatment. Finally, the review chimpanzees has been outlawed, or at least severely found that published papers describing potential restricted, in the relatively few scientifically new cancer therapies tested in chimpanzees often advanced nations where it was ever practised (see included significant caveats concerning differences http://www.releasechimps.org/mission/end-chim- in the species, acknowledged that research with panzee-research/country-bans/). Nevertheless, some chimpanzees was no better than research with observers still insist that such restriction is a mis- other animals, and described interventions that take (4). Legislation banning experimentation on had not been pursued clinically, presumably due to chimpanzees has been implemented largely due to adverse preclinical results (12). These conclusions ethical concerns — namely, that research with are supported by many of the genetic differences chimpanzees causes extreme pain and suffering to between humans and chimpanzees described in animals with very highly developed cognitive and the present paper. emotional capacities. This rationale is being Finally, two in-depth reviews by the author of increasingly augmented by a burgeoning scientific the present paper have recently been completed. argument — that data from chimpanzee experi- These address claims that research with chim- ments are flawed and invalid with regard to human panzees must continue in order to provide a better medicine, and that superior alternatives, with understanding of hepatitis C, and to realise treat- greater human relevance, do exist. This argument, ments and cures for this devastating disease that in turn, is bolstered by our increasing knowledge of affects millions of people worldwide (9, 10). This is the genetic differences between humans and chim- important, not only due to the impact of hepatitis panzees, as was summarised in this review, which C on the human population, but also because it serve to explain the negative results and conclu- represents the area of greatest use of chimpanzees sions reached in critical assessments of the rele- in laboratories at present. These two reviews vance of the chimpanzee for many areas of research demonstrate that researchers have, for many on human disease. These negative results and con- years, cited numerous serious caveats and prob- clusions, briefly summarised above, also give value lems with the chimpanzee model. They have also and meaning to those genetic differences, showing stressed the urgent need for in vitro viral culture that they do indeed matter and that they combine to systems to accelerate progress, which occurred make the chimpanzee a very different species from much earlier for other viruses, such as polio and the human. measles. Paralleling the case of HIV/AIDS, these Yet, in spite of this substantial and compelling caveats, and the need for in vitro culture systems, knowledge, the ostensibly extreme genetic similar- are due to major pathological differences between ity of chimpanzees and humans is frequently used human and chimpanzee hepatitis C virus infection. to assert the validity and indispensability of chim- The reviews show that in vitro and clinical panzee research. An example from 2005 is a high- approaches, rather than chimpanzee experiments, profile opinion piece by two prominent chimpanzee have provided the most substantial, comprehen- researchers, from two institutions housing hun- sive and important data. It is now possible to dreds of research chimpanzees, in which they investigate the entire life cycle of the hepatitis C make a plea for increased research on captive virus, immune responses and host factors, identify chimpanzees (4). It is conceded that the chim- therapeutic targets, and test new therapies and panzee is the species most genetically similar to vaccines, in a human, and therefore completely rel- humans, although the complexity of aligning such evant, context. The reviews show that chimpanzee huge genomes makes it difficult, if not impossible, use is declining markedly in research on hepatitis to measure this similarity in a truly precise way. C; indeed, it has fallen by over 60% in the last 20 Initial estimates suggested this similarity was years and is now at a historical low of just one- around 98.5% to 99% (King and Wilson [94], third of its 1985 peak level. In contrast, non-ani- reviewed by Varki [95]), a figure that has been con- mal research on hepatitis C has increased 80-fold firmed more recently, although with certain over the same time period. In summary, in the caveats (reviewed by Volfovsky et al. [44],
536 J. Bailey Watanabe et al. [81], and the Chimpanzee for example, are more susceptible to a variety of Sequencing and Analysis Consortium [92], among “infectious diseases, cardiovascular diseases, obe- others). For instance, the figure of 98.5–99% takes sity, type II diabetes, autoimmune diseases, major into account nucleotide substitutions only. When psychoses, and neurodegenerative diseases” (69), genomic insertions and deletions are considered, are more prone to developing cancer (12, 59), and the overall similarity decreases to approximately are differently susceptible to malarial parasites 95–96% (14, 15), or to as low as 93.5% (16). Even when they are compared to chimpanzees (97). when repeat-sequence and low-complexity DNA Experimentation with chimpanzees has failed to are excluded, the overall difference is still thought translate to meaningful progress in HIV/AIDS to be approaching 2.5%, which is double the origi- (11), hepatitis C (9, 10) and cancer (12), as well as nal and highly cited estimate (16). more generally for other diseases (13; see also It has been widely acknowledged for some years http://www.releasechimps.org/pdfs/chimp-efficacy- that a genetic difference of only 1–2% is actually a paper-main.pdf and http://www.releasechimps. ‘myth’, but its fragile underpinnings have appar- org/pdfs/Chimp-CA-suppl.pdf). ently been overlooked by advocates of chimpanzee The use of chimpanzees has declined greatly in research in their defence of the practice. Their all areas of research, and it cannot be shown to be assertion of a high genetic similarity and of only a necessary in any current field of biomedical 1–2% difference is even more surprising when one research. It has been falsely credited with major appreciates the extent of the awareness within the breakthroughs in the past, where human-specific scientific community of other important inter- methods were the true contributors, while modern species disparities and their phenotypic effects, technologies and cutting-edge alternative methods and for how long this awareness has existed. As negate any value it may have had historically. The long ago as 1975, the authors of a seminal study presumed benefits are arguable in themselves. It (94) that indicated a 1% difference between must therefore be concluded that any justification humans and chimpanzees insisted that ‘‘the of chimpanzee research based on genetic similarity genetic distance between humans and the chim- is superficial and incorrect. And, based on a deeper panzee is probably too small to account for their and more thorough appreciation of profound inter- substantial organismal differences’’. In other species differences in gene expression, rather than words, differences in gene regulation and expres- simple genetic similarity, it must be concluded sion, rather than protein-coding mutations, must that the chimpanzee does not, and can never, con- play a major role (46, 94, 96). Knowledge of other stitute a crucial and indispensable model for factors with a great influence on gene expression human biomedical research. and human/chimpanzee phenotypes also precedes When the scientific case against chimpanzee and surrounds such claims. As this review demon- experimentation is considered alongside the strates, various phenomena — including genomic numerous and varied concerns about animal wel- rearrangements, mobile DNA elements (e.g. fare, ethical issues and financial concerns (7, 8), LINEs and SINEs), gene duplications and dele- the argument becomes even more compelling and tions, copy number variation, differences in tran- formidable. Studies have revealed post-traumatic scription factors and binding sites, DNA stress disorder in ex-research chimpanzees now in methylation, miRNAs and binding sites, gene edit- sanctuary (5, 98), and have detailed physical and ing and splicing and protein phosphorylation — psychological trauma suffered by chimpanzees contribute to human/chimpanzee biological differ- that have been raised in various human/chim- ences. Many of these factors affect gene expression panzee contexts and then used in research, as well and function, and are deemed to have played a as the chimpanzees’ ability to recover from such greater role than simple nucleotide substitutions trauma once in sanctuary (6). Tens of millions of in the evolution of species-specific phenotypes. As taxpayer dollars could be saved each year by trans- opined in a review outlining significant gene gain ferring all federally owned and supported chim- and loss in humans and chimpanzees: “Without panzees from laboratories to sanctuaries, where accounting for differences in the total DNA unique they would receive superior care and enjoy a to each species, we cannot hope to take a proper higher quality of life (Capaldo and Owens, submit- accounting of the meaningful genetic divergence ted). Public opinion polls show that twice as many between humans and chimpanzees” (46). Americans support a ban on chimpanzee research As the recent analyses summarised in this as oppose one, and that 71% of the American pub- review show, the divergence between humans and lic think chimpanzees used in research for more chimpanzees is considerable, and it translates to than 10 years should be retired (this figure would major differences in susceptibility to disease, in now include over 90% of all chimpanzees in US lab- symptoms and in pathologies. These differences oratories; see http://www.releasechimps.org/mis- have “eluded any molecular explanation within sion/end-chimpanzee-research/public-opinion/). this supposedly 1% diversity range” (45), a diver- Finally, chimpanzee experimentation is a ‘special sity which “is clearly misleading” (91). Humans, case’. This is evidenced not only by public opinion,
Chimpanzee-based research on human disease 537
but also by various policies and laws in the USA, on 26 August, 2010. Sincere gratitude is expressed
such as: the CHIMP Act passed in 2000 (99); the to the New England Anti-Vivisection Society
extension, in 2007, of a moratorium in place since (NEAVS) and the British Union for the Abolition of
1995 on breeding chimpanzees which applies to Vivisection (BUAV) for funding this work. There
federally supported chimpanzees in US laborato- are no financial conflicts of interest. The author is
ries (see http://www.releasechimps.org/2007/05/ supported by these two not-for-profit groups,
22/breeding-moratorium-decision-good-news/); the which campaign against animal experimentation.
many countries that have recently banned or
severely limited the use of great apes (see http://
www.releasechimps.org/mission/end-chimpanzee- Received 30.06.11; received in final form 14.09.11;
accepted for publication 14.09.11.
research/country-bans/); and the considerable leg-
islative support in both the US House of
Representatives and the US Senate for GAPCSA
(the Great Ape Protection and Cost Savings Act),
References
which seeks to end invasive chimpanzee research 1. Kohne, D.E. (1970). Evolution of higher-organism
and retire all federally owned chimpanzees in US DNA. Quarterly Reviews of Biophysics 3, 327–375.
laboratories to permanent sanctuary (23, 24). 2. Sibley, C.G. & Ahlquist, J.E. (1984). The phylogeny
To the best available knowledge, the USA is the of the hominoid primates, as indicated by
only country in the world that actively conducts DNA–DNA hybridization. Journal of Molecular
invasive research on captive chimpanzees to any Evolution 20, 2–15.
3. Ebersberger, I., Metzler, D., Schwarz, C. & Pääbo, S.
significant degree. For the sake of the approxi- (2002). Genomewide comparison of DNA sequences
mately 1,000 chimpanzees in US laboratories, who between humans and chimpanzees. American
almost without exception are elderly and have suf- Journal of Human Genetics 70, 1490–1497.
fered greatly, there is a strong argument for the 4. VandeBerg, J.L. & Zola, S.M. (2005). A unique bio-
USA to join the rest of the world and end this medical resource at risk. Nature, London 437,
research. There is also a strong argument to end 30–32.
5. Bradshaw, G.A., Capaldo, T., Lindner, L. & Grow,
this research for the sake of humanity, which G. (2008). Building an inner sanctuary: Complex
stands to benefit from superior, more humane, and PTSD in chimpanzees. Journal of Trauma &
more human-relevant scientific inquiry, should the Dissociation 9, 9–34.
use of chimpanzees cease. The evidence reviewed 6. Bradshaw, G.A., Capaldo, T., Lindner, L. & Grow,
here shows that chimpanzees are simply not G. (2009). Developmental context effects on bi-cul-
‘human enough’ to constitute valuable models for tural post-trauma self repair in chimpanzees.
Developmental Psychology 45, 1376–1388.
research on human diseases: important differ- 7. Sauer, U.G. (2000). Reasons for not using primates
ences, with serious biological consequences, run in research. ALTEX 17, 217–220. [In German.]
right through each level of gene expression and 8. Thew, M. (2002). Are results of primate research
protein function, from chromosome and chromatin worth the suffering it causes? Nature, London 418,
structure to post-translational modification. This 273.
results in non-predictive data of poor relevance to 9. Bailey, J. (2010). An assessment of the use of chim-
panzees in hepatitis C research past, present and
humans, that rarely translate to the clinic and that
future: 1. Validity of the chimpanzee model. ATLA
actually impede progress in human medicine. The 38, 387–418.
argument that chimpanzees must constitute a 10. Bailey, J. (2010). An assessment of the use of chim-
good model for research on human diseases, based panzees in hepatitis C research past, present and
on their ostensible genetic similarity to humans, is future: 2. Alternative replacement methods. ATLA
specious and should be dismissed. Only by moving 38, 471–494.
11. Bailey, J. (2008). An assessment of the role of chim-
away from chimpanzee research, and by fully
panzees in AIDS vaccine research. ATLA 36,
embracing and adopting superior human-specific 381–428.
alternatives, can treatments or cures for the many 12. Bailey, J. (2009). An examination of chimpanzee
diseases that blight the lives of hundreds of mil- use in human cancer research. ATLA 37, 399–416.
lions of people be realised quickly and safely. 13. Knight, A. (2007). The poor contribution of chim-
panzee experiments to biomedical progress. Journal
of Applied Animal Welfare Science 10, 281–308.
14. Britten, R.J. (2002). Divergence between samples of
Acknowledgements chimpanzee and human DNA sequences is 5%,
counting indels. Proceedings of the National
Jarrod Bailey was the sole author of this manu- Academy of Sciences of the USA 99, 13,633–13,635.
script and was responsible for its conception, 15. Varki, A. & Altheide, T.K. (2005). Comparing the
research, and preparation. The work herein has human and chimpanzee genomes: Searching for
needles in a haystack. Genome Research 15,
not been presented anywhere else in print prior to
1746–1758.
this publication, and is based on an oral presenta- 16. Wetterbom, A., Sevov, M., Cavelier, L. & Berg-
tion given at the ‘Animals, Research and strom, T.F. (2006). Comparative genomic analysis
Alternatives’ conference in Washington, DC, USA, of human and chimpanzee indicates a key role forYou can also read
