Unique structure of the M loop region of β1-tubulin may contribute to size variability of platelets in the family Felidae


Mary K. Boudreaux, Department of Pathobiology, College of Veterinary Medicine, Auburn University, 166 Greene Hall, Auburn, AL 36849-5519, USA
E-mail: boudrmk@auburn.edu


Background: Platelet size is relatively uniform in mammals except for domestic cats. Uniform platelet production by megakaryocytes can be disrupted if microtubule assembly or dynamics is impaired. Mutations in the gene encoding β1-tubulin have been documented in dogs and people, and the resulting microtubule effects have been associated with production of large platelets.

Objectives: The objectives of this study were to evaluate morphology of platelets on feline blood smears, determine the gene sequences encoding β1-tubulin in members of the family Felidae, and compare the findings with those in other mammalian species to determine whether predicted structural differences in β1-tubulin that might affect microtubule stability or assembly were present.

Methods: At least 100 platelets/smear on blood smears from 15 domestic cats and 88 big cats were evaluated to assess platelet size variability. Platelet-derived cDNA obtained from a domestic cat and genomic DNA isolated from blood samples of domestic cats and other members of the family Felidae were analyzed by PCR using primers specific for β1-tubulin. Gene sequences obtained were compared with those of other common mammals.

Results: Two differences in gene sequence were found in a highly conserved region encoding the M loop of β1-tubulin in members of the family Felidae compared with sequences from other species. Platelet size variation was present in big cats and domestic cats. In addition, a rare amino acid change was documented in the C-terminal region encoding the H11 helix in domestic cats.

Conclusion: Members of the family Felidae have an altered M loop region in β1-tubulin compared with other mammals. This variation may contribute to the observed platelet size variability.


Although people and most domestic animals have relatively uniform-sized platelet populations, marked differences in platelet size are a common occurrence in cats. In most species platelet size ranges from 2 to 4 μm; however, in cats platelet size may range from 2 to 6 μm or larger. On average, cats have mean platelet volumes (MPV) that are significantly higher than MPVs in people and other domestic animals owing to the high prevalence of large platelets.1–3 Similar to platelets of other species examined by light microscopy, feline platelets are anucleate cytoplasmic fragments with centrally located azurophilic granules. However, in cats, large platelets equal in size to RBCs are commonly observed.

Uniform platelet production by megakaryocytes can be disrupted if microtubule stability or assembly is altered. Cavalier King Charles Spaniels with inherited macrothrombocytopenia have a mutation in the gene encoding β1-tubulin that is thought to result in a change in microtubule stability with production of large platelets.4 Recently, a similar β1-tubulin gene mutation was documented in people with inherited macrothrombocytopenia, thus illustrating the importance of microtubules for uniform platelet production.5 We hypothesized that altered amino acid sequences in the M loop of β1-tubulin, the region responsible for the lateral interaction between protofilaments within microtubules, play a role in modifying microtubule stability and contribute to size variability of platelets in domestic cats. To partially test this hypothesis, we compared the M loop β1-tubulin gene sequences of other members of the family Felidae (big cats) to that of domestic cats and evaluated blood smears to assess platelet size variability.

Materials and Methods


Blood samples collected in potassium (K3) EDTA tubes (Tyco Healthcare Group Lp, Mansfield, MA, USA) from domestic cats were submitted to the clinical pathology laboratory in the Veterinary Teaching Hospital at the College of Veterinary Medicine, Auburn University for routine testing. Samples from 15 domestic cats, including 7 Ragdoll cats, were evaluated. Blood samples in EDTA and blood smears from big cats were provided over an 18-month period (August 2008 to February 2010) by the San Diego Wild Animal Park and San Diego Zoo and by Colleges of Veterinary Medicine at the University of Tennessee, Kansas State University, and the University of Georgia. Samples were collected during routine evaluations and not solely for the purpose of this study. A total of 88 samples from big cats were evaluated and were from 47 tigers (1 Bengal, 3 Siberian, 3 Indochinese, 2 Sumatran, and 38 subspecies not indicated), 9 Eastern cougars, 14 African lions, 7 leopards (2 Amur, 1 Snow, 1 North Chinese, and 3 subspecies not indicated), 4 cheetahs, 1 caracal, 4 servals, and 2 bobcats. DNA was isolated from the EDTA-anticoagulated blood samples within 1 month of blood collection; in some cases, blood samples were batched and stored in the refrigerator until shipments were made. Blood smears were prepared within 1 hour of blood collection and were then stained. Platelet size and morphology were evaluated by light microscopy. Platelet numbers were estimated to be at least 100,000/μL in all of the samples included in the study. Platelet size variability was assessed by comparing the size of platelets with that of RBCs and by measurement with a calibrated micrometer. At least 100 platelets were evaluated on each blood smear. Blood smears from domestic cats and dogs were used as the standards for comparison.

Molecular studies

Domestic feline platelet-derived cDNA was used as a template in initial PCR assays. Platelet-rich plasma (PRP) was isolated from EDTA-anticoagulated blood. Prostaglandin E1 (Sigma, St. Louis, MO, USA) was added to PRP samples at a final concentration of 3 μM before centrifugation of PRP at 1500g to form pellets. Platelet pellets were resuspended in a small volume of autologous plasma and transferred to RNase-free tubes and centrifuged again. Residual plasma was removed from the pellets, and the pellets were frozen at −80°C for approximately 6 months until used for RNA isolation. Total RNA was isolated from platelet pellets using a commercially available kit (Micro to Midi Total RNA Purification Kit, Invitrogen, Carlsbad, CA, USA). cDNA synthesis was performed using a separate commercially available kit (iScript cDNA synthesis kit, BioRad, Hercules, CA, USA).

Primers were designed based on sequences within the feline genome that aligned with known human sequences for β1-tubulin (NCBI-BLAST). PCR products were electrophoresed by agarose gel electrophoresis and DNA was extracted from target bands using the QIAquick Gel Extraction Kit (Qiagen Inc., Valencia, CA, USA). Extracted DNA was sequenced using an ABI 3100 Genetic Analyzer (Applied Biosystems by Life Technologies Corp., Carlsbad, CA, USA). The cDNA sequence obtained for feline β1-tubulin was submitted to GenBank (GI:197359561); this sequence matched the coding region of genomic DNA determined in later experiments. Primers for the rest of the study were based on the sequence for domestic cats and designed to amplify an 874 base-pair segment encoding most of exon 4. Primers used for this study were CACGCTGCTGCTGAGCCGGAT (forward) and CACGACGTCTCCGTCTTCCTC (reverse). DNA obtained from domestic cats and big cats was isolated from EDTA-anticoagulated blood for these studies (QIAamp DNA Blood Mini Kit, Qiagen Inc.).

The feline amino acid sequence for β1-tubulin was aligned with amino acid sequences, obtained from GenBank, from other species. Changes in amino acids involved in lateral or longitudinal contacts6 that resulted in a change in polarity or charge and were unique to the cat were considered significant. The predicted amino acid sequence encoded by exon 3 for equine β1-tubulin that was available on GenBank (GI:194224589) was not correct due to a missing equine genomic sequence. Primers were designed around the missing sequence and PCR was performed on equine genomic DNA to generate the correct coding sequence for exon 3. This sequence was used along with the available GenBank sequence to determine the correct coding sequence for equine β1-tubulin.


The platelets of all evaluated members of the family Felidae had some degree of size variability by light microscopy; in all instances variability was greater than that observed on canine blood smears (Figure 1). The proportion of platelets of domestic cats and big cats that were 4 μm or larger was 10–25%, whereas only 5% or less of canine platelets were 4 μm or larger. Large elongated platelets were prevalent on blood smears obtained from 1 Ragdoll cat (Figure 2), and this cat appeared to be mildly thrombocytopenic (approximately 100,000/μL) based on evaluation of the blood smear. Large elongated platelets were not frequently observed on blood smears obtained from big cats or other domestic cats, including other Ragdoll cats.

Figure 1.

 Examples of peripheral blood smears obtained from members of the family Felidae illustrating the platelet size variability typical of these animals. (A) Domestic cat. (B) African lion. (C) Cheetah. (D) Serval. (E) Cougar. (F) Sumatran tiger. Wright–Giemsa, × 100 objective.

Figure 2.

 Peripheral blood smear from a Ragdoll cat homozygous for the F385Y change in the gene encoding β1-tubulin. Note the elongated platelets. Wright–Giemsa, × 100 objective.

Alignment of the feline amino acid sequence for β1-tubulin, predicted from the nucleotide sequence obtained from feline platelet-derived cDNA and genomic DNA, with human, canine, equine, bovine, and porcine amino acid sequences revealed 2 significant differences in the M loop, a region involved in lateral contact: a polar acidic aspartic acid instead of a nonpolar neutral glycine at position 277 and a polar strongly basic arginine instead of a polar neutral glutamine at position 279 (Figure 3). These same amino acid differences within the M loop region were also present in all the samples from big cats. Significant variations were not present in the regions involved in longitudinal contacts. Rarely, an amino acid variation was also observed in domestic cats at amino acid position 385 within alpha helix 11, resulting in a polar tyrosine instead of a nonpolar phenylalanine. This change was seen in 3 Ragdoll cats; 1 cat was homozygous and 2 were heterozygous for the change. The change was not seen in 4 other Ragdoll cats. Large elongated platelets were prevalent on blood smears from the Ragdoll cat homozygous for this amino acid change (Figure 2), but not in the heterozygous Ragdoll cats or any cats without the change. This amino acid change was not documented in any of the big cats.

Figure 3.

 The feline amino acid sequence for β1-tubulin, predicted from the nucleotide sequence obtained from feline platelet-derived cDNA. DNA is aligned with human, canine, equine, bovine, and porcine amino acid sequences. Amino acid residues involved in lateral contacts within microtubules are highlighted in yellow, and amino acids in regions involved in longitudinal contacts are highlighted in blue. Significant amino acid differences in the M loop (boxed area) are highlighted in pink. The 2 changes are substitution of a polar acidic aspartic acid (D) for a nonpolar neutral glycine (G) at position 277 (G277D) and substitution of a polar strongly basic arginine (R) for a polar neutral glutamine (Q) at position 279 (Q279R). The underlined amino acid D denotes the location of the mutation D249N associated with macrothrombocytopenia in Cavalier King Charles Spaniels. Rarely, an amino acid variation was also observed in Ragdoll cats at amino acid position 385 (white letter highlighted in blue) within alpha helix 11 (highlighted in green), resulting in substitution of a nonpolar phenylalanine for a polar tyrosine (F385Y).


Microtubules are essential for normal platelet formation by megakaryocytes and maintenance of discoid platelet shape.7β1-tubulin is a protein component within microtubules that is expressed exclusively in platelets and megakaryocytes and is estimated to account for 90% of the total β-tubulin content in blood platelets.8 As megakaryocytes mature, microtubule assembly within the cytoplasm is a critical component of proplatelet formation and ultimate platelet production. Proplatelet elongation is thought to occur as motor proteins bind to microtubules allowing microtubules to slide past one another. Thus, microtubule structure important for assembly, stability, and protein-binding must be maintained for orderly platelet formation to occur.9,10

Microtubules are made up of laterally attached protofilaments, which are polymers of αβ-tubulin dimers.11β1-tubulin is specifically synthesized in megakaryocytes and is normally not expressed in other tissues.8 The M loop is located on the lateral region of both the α- and β-monomers (Figure 4). This region has a higher affinity for the same subunit, so α–α and β–β interactions are preferred. As the protofilaments spiral to form the microtubule, the protofilaments are progressively offset and thus α- and β-monomers do not line up in the area referred to as the seam.12 In the seam region α–β and β–α interactions occur, and this region constitutes an area of weakness in the microtubule, partly owing to the lowered affinity at the M loop. The M loop region is a critically important contributor to microtubule polymerization.13,14 Variations in the M loop region can affect microtubule flexibility by altering interprotofilament interactions. Taxol, a microtubule-stabilizing agent, binds near the β1-tubulin M loop region and is thought to have an impact on microtubule assembly by altering M loop flexibility (Figure 4).15 In one study evaluating microtubule primary structures of Antarctic fishes, M loop variations were documented in α- and β-tubulins. These changes were hypothesized to play a role in microtubule stability and assembly and were thought to be part of an adaptation to changes in temperature.16

Figure 4.

 Model of a bovine αβ-tubulin dimer stabilized with taxol. Arrows indicate the location of the G277D and Q279R changes documented in domestic cats and big cats in the M loop of β1-tubulin. The locations of D249N and F385Y are also depicted. Taxol (solid blue) is labeled. Model obtained from RCSB Protein Data Bank Protein Workshop (http://www.pdb.org/pdb/static.do?p=general_information/about_pdb/policies_references.html).18,19

Amino acids within the M loop region are highly conserved across species. The findings of this study suggest that a region of variation in the M loop is present in domestic cats and in other members of the family Felidae. The nucleic acid differences noted are predicted to encode an aspartic acid instead of a glycine (G277D) and an arginine instead of a glutamine (Q279R) within this highly conserved region. These variations would result in a change in charge and may have an impact on the stability or assembly of platelet microtubules, resulting in the production of variably sized platelets. An interesting difference in cats compared with dogs and people with mutations in the gene encoding β1-tubulin is that, although cats have variably sized platelets, they do not have thrombocytopenia. The mutations described in Cavalier King Charles Spaniels and in people with macrothrombocytopenia are located at or near the interface between α- and β-subunits (intraprotofilament effects), whereas the amino acid differences described in cats are within the M loop region (interprotofilament effects). These findings suggest the possibility that intraprotofilament changes that have an impact on microtubule assembly or dynamics are more likely to result in macrothrombocytopenia, whereas interprotofilament changes result in altered platelet size without an affect on number. Alternatively, these findings may be related to the existence of a higher platelet mass in domestic cats compared with other species.3 Although dogs, people, and cats have comparable platelet numbers, in dogs and people the MPV is 7–9 fL whereas in cats the MPV is 11 fL, resulting in an overall lower platelet mass in dogs and people compared with cats. The β1-tubulin gene mutations resulting in macrothrombocytopenia in dogs and people may result from the overall platelet mass being met fairly quickly as many large platelets enter circulation, thus requiring fewer overall numbers of platelets to be produced to achieve adequate hemostasis. The presence of large and presumably more functional platelets in cats may also contribute to the overall enhanced platelet reactivity observed in cats. However, feline platelet function is comparable to that observed in people, a species in which the platelet mass is lower and comparable to that found in dogs, thus casting some doubt on this hypothesis.

The finding of the rare F385Y change in the H11 helix within the C-terminal domain in Ragdoll cats was interesting. Phenylalanine 385 is located in the H11 alpha helix of β1-tubulin (Figure 4). The H11 alpha helix is within the C-terminal domain and forms part of a prominent longitudinal ridge on the outer protofilament surface.11 This region is involved in the binding of microtubule-associated proteins and motor proteins involved in microtubule assembly and stabilization.11,17 Elongated platelets were not observed in the samples from big cats except for a few mildly elongated platelets observed in blood smears from 1 cheetah. In a previous study, large elongated platelets were noted in blood smears from 1 of 6 king cheetahs.2 More studies are needed to determine whether there is a correlation with formation of large elongated platelets and the presence of the F385Y mutation in domestic cats and big cats. It is interesting to speculate that the alteration in polarity introduced by this amino acid change might affect the binding of motor proteins that facilitate proplatelet extension. The production of large elongated platelets may be enhanced in cats responding to thrombocytopenia if, under these conditions, proplatelet formation is occurring more rapidly and with higher frequency, placing an additional stress on the mechanics of platelet production. Identification of factors contributing to feline platelet size heterogeneity could further our understanding of megakaryocyte maturation and platelet formation. Additional studies are required to determine whether M loop or other amino acid differences play a role in production of variably sized and shaped platelets in cats.


EDTA anticoagulated blood samples and blood smears from big cats were graciously provided over an 18-month period by Ms. McCaffree at the San Diego Wild Animal Park and the San Diego Zoo and by Drs. McCain, Shoemaker, and Camus at the Colleges of Veterinary Medicine at the University of Tennessee, Kansas State University, and the University of Georgia, respectively. Christina Osborne received partial funding from the Morris Animal Foundation, and Adam Herre and Elliot Ramos Rivera received partial funding from Merck-Merial. The authors thank Kevin King of the Platelet Laboratory at the College of Veterinary Medicine, Auburn University for his valuable technical assistance.

Disclosure: The authors have indicated they have no affiliations or financial involvement with any organization or entity with a financial interest in, or in financial competition with, the subject matter or materials discussed in this paper.