Improved yields of bovine blastocysts from in vitro-produced oocytes. II. Media and co-culture cells

Influence of sperm fertilising concentration, sperm selection method and sperm capacitation procedure on the incidence of numerical chromosomal abnormalities in IVF early bovine embryos

The occurrence of numerical chromosomal aberrations, widely described as a major cause of mortality in in vitro-produced (IVP) embryos, has been linked to several factors. In the present study we investigated the effect of sperm fertilising concentration and semen handling (sperm selection and capacitation) before IVF on the rate of numerical chromosomal abnormalities in bovine embryos. In all, 466 IVP cattle embryos were karyotyped throughout three sequential experiments, analysing the effects of sperm fertilising concentration (0.1, 1.0 or 10×10(6) spermatozoa mL(-1)), selection method (unselected or Percoll-selected spermatozoa) and capacitation medium (bovine serum albumin (BSA), heparin or their combination). The percentage of normal (diploid) and aberrant (haploid, polyploid or aneuploid) embryos was noted in each experiment. The rate of numerical chromosomal abnormalities was mainly affected by sperm fertilising concentration (P<0.01) and, to a lesser extent, by the sperm capacitation medium (P<0.05). Polyploidy and haploidy rates were only affected by sperm fertilising concentration (P<0.05). Interestingly, the sperm selection technique used in the present study did not reduce the incidence of chromosome abnormalities in IVP cattle embryos (P>0.05). Finally, aneuploidy rates were not affected during the experiments (P>0.05), which suggests that they are not related to sperm-related factors. On the basis of these results, we conclude that sperm fertilising concentration is the 'paternal' key factor that affects the rate of numerical chromosomal abnormalities in IVP bovine embryos. By making small adjustments to fertilising protocols, the rate of cytogenetically aberrant embryos can be markedly reduced.

Effects of oocyte quality, incubation time and maturation environment on the number of chromosomal abnormalities in IVF-derived early bovine embryos

Chromosomal aberrations are one of the major causes of embryo developmental failures in mammals. The occurrence of these types of abnormalities is higher in in vitro-produced (IVP) embryos. The aim of the present study was to investigate the effect of oocyte morphology and maturation conditions on the rate of chromosomal abnormalities in bovine preimplantational embryos. To this end, 790 early cattle embryos derived from oocytes with different morphologies and matured under different conditions, including maturation period (24 v. 36 h) and maturation media (five different serum supplements in TCM-199), were evaluated cytogenetically in three sequential experiments. The rates of normal diploidy and abnormal haploidy, polyploidy and aneuploidy were determined in each embryo. Throughout all the experiments, the rate of chromosomal abnormalities was significantly (P < 0.05) affected by oocyte morphology and maturation conditions (maturation time and culture medium). Lower morphological quality was associated with a high rate of chromosome abnormalities (P < 0.05). Moreover, polyploidy was associated with increased maturation time (P < 0.01), whereas the maturation medium significantly (P < 0.05) affected the rates of haploidy and polyploidy. In general, supplementing the maturation medium with oestrous cow serum or fetal calf serum resulted in higher rates of chromosomal aberrations (P < 0.05) compared with the other serum supplements tested (bovine steer serum, anoestroues cow serum, bovine amniotic fluid and bovine serum albumin). On the basis of the results of the present study, we conclude that the morphological quality of oocytes and the maturation conditions affect the rate of chromosomal abnormalities in IVP bovine embryos.

Efeito do estradiol e da progesterona no desenvolvimento e na qualidade de embriões bovinos produzidos in vitro

Avaliaram-se o desenvolvimento e a qualidade de embriões bovinos, cocultivados com células epiteliais do oviduto bovino (CEOBs) expostas ou não ao estradiol e à progesterona. Os ovócitos foram maturados in vitro por 24h e, então, fertilizados utilizando-se sêmen congelado, em estufa de CO2 a 5% e 38,5ºC. As CEOBs foram cultivadas em TCM-199 com ou sem estradiol (E2) (24 horas), nas mesmas condições da maturação e fertilização in vitro (MIV e FIV), e, em seguida, adicionadas aos diferentes grupos em CR2 com ou sem progesterona (P4) (G1=P4+E2); (G2=E2); (G3=P4) e (G4=controle). Após 18h da FIV, as células foram cultivadas nos diferentes sistemas. Nenhuma diferença (P>0,05) foi observada nas taxas de clivagem entre G1, G2 e G4 (53,5%; 56,3%; 51,7%) e nos padrões de blastocistos (BLs) (29,3%; 31,2%, 28,7%). Índices menores (P<0,05) foram obtidos no G3 para ambas as variáveis (34,5%; 16,4%). G1 e G2 apresentaram taxas de eclosão maiores (P<0,05) que os outros grupos (23,3%; 23,2%), sendo G4 (19,3%) diferente de G3 (16,1%). Em G1, G2 e G3, o número de células nos BLs aumentou 125,9; 128,4 e 123,6, respectivamente (P<0,05), em relação ao G4 (112,5). Conclui-se que o tratamento das CEOBs com o E2, nas primeiras 24 horas de cultivo, pode ser usado isoladamente ou em combinação com a progesterona, a fim de melhorar a qualidade de embriões bovinos produzidos in vitro

Global poly(A) mRNA expression profile measured in individual bovine oocytes and cleavage embryos

The objective of this article was to estimate quantitative differences for GAPDH transcripts and poly(A) mRNA: (i) between oocytes collected from cumulus–oocyte complexes (COCs) qualified morphologically as grades A and B; (ii) between grade A oocytes before and afterin vitromaturation (IVM); and (iii) amongin vitro-produced embryos at different developmental stages. To achieve this objective a new approach was developed to estimate differences between poly(A) mRNA when using small samples. The approach consisted of full-length cDNA amplification (acDNA) monitored by real-time PCR, in which the cDNA from half of an oocyte or embryo was used as a template. The GAPDH gene was amplified as a reverse transcription control and samples that were not positive for GAPDH transcripts were discarded. The fold differences between two samples were estimated using delta Ct and statistical analysis and were obtained using the pairwise fixed reallocation randomization test. It was found that the oocytes recovered from grade B COCs had quantitatively less poly(A) mRNA (p< 0.01) transcripts compared with grade A COCs (1 arbitrary unit expression rate). In the comparison with immature oocytes (1 arbitrary unit expression rate), the quantity of poly(A) mRNA did not change during IVM, but declined following IVF and varied with embryo culture (p< 0.05). Amplification of cDNA by real-time PCR was an efficient method to estimate differences in the amount of poly(A) mRNA between oocytes and embryos. The results obtained from individual oocytes suggested an association between poly(A) mRNA abundance and different morphological qualities of oocytes from COCs. In addition, a poly(A) mRNA profile was characterized from oocytes undergoing IVM, fertilization and blastocyst heating.

Immunolocalization of bovine sperm protease BSp120 by light and electron microscopy during capacitation and the acrosome reaction: Its role in in vitro fertilization

Mammalian fertilization involves various steps in which the participation of specific enzymes has been demonstrated by numerous studies. Acrosin is one of the most widely acrosomal protease in mammalian spermatozoa studied, including bovine; however, other proteases have also been described. A new trypsin-like serine protease named bovine serine protease of 120 kDa (BSp120) and its pre-cursor BSp66 (66 kDa) were identified in bovine spermatozoa. Cytological and ultrastructural immunolocalization studies on BSp120 were performed in live and fixed cells. Immunoflorescence assays with specific polyclonal antibodies revealed localization of BSp120 on the sperm head, with a signal homogeneously distributed over the acrosome resembling a horseshoe. After the acrosome reaction, sperm showed a patchy pattern in the acrosomal cap. Immune electron microscopy analysis indicated that BSp120 is located over the head plasma membrane of capacitated spermatozoa and acrosome reacting spermatozoa. To assess BSp120 function in sperm-oocyte interaction, in vitro fertilization studies were conducted. Oocytes were incubated with spermatozoa pre-treated with anti-BSp120, anti-guinea pig acrosin, and anti-BSp120 plus anti-guinea pig acrosin. Pre-treatment of bovine spermatozoa with antibodies towards each protein did not significantly modify fertilization rates. However, when both anti-acrosin and anti-BSp120 antibodies were simultaneously added, there was a significant decrease in the fertilization rate, suggesting that both enzymes may be required for fertilization. Altogether, the results from the present study described the localization of BSp120 over the acrosome of bovine sperm, and suggest its involvement in fertilization.

Cell donor influences success of producing cattle by somatic cell nuclear transfer

To assess sources of variation in nuclear transfer efficiency, bovine fetal fibroblasts (BFF), harvested from six Jersey fetuses, were cultured under various conditions. After transfection, frozen-thawed lung or muscle BFF donor cells were initially cultured in DMEM in 5% CO(2) and air and some were transferred to MEM, with 5% or 20% O(2) or 0.5% or 10% serum and G418 for 2-3 wk. Selected clonal transfected fibroblasts were fused to enucleated oocytes. Fused couplets (n = 4007), activated with ionomycin and 6-dimethylaminopurine, yielded 927 blastocysts, and 650 were transferred to 330 recipients. Fusion rate was influenced by oxygen tension in a fetus-dependent manner (P < 0.001). Blastocyst development was influenced in a number of ways. Hip fibroblast generated more blastocysts when cultured in MEM (P < 0.001). The influence of serum concentration was fetus dependent (P < 0.001) and exposing fibroblast to low oxygen was detrimental to blastocyst development (P < 0.001). Cells from two of the six fetuses produced embryos that maintained pregnancies to term, resulting in eight viable calves. Pregnancy rates 56 days after transfer for the two productive donor fetuses, was at least double that of other recipients and may provide a fitness indicator of BFF cell sources for nuclear transfer. We conclude that a significant component in determining somatic cell nuclear transfer success is the source of the nuclear donor cells.

Effect of Dietary Energy and Protein on Bovine Follicular Dynamics and Embryo Production In Vitro: Associations with the Ovarian Insulin-Like Growth Factor System1

Heifers were assigned either low or high (HE) levels of energy intake and low or high concentrations of dietary crude protein. The effect of these diets on the plasma concentrations of insulin, insulin-like growth factor (IGF)-I, and urea on follicular growth and early embryo development is described. We propose that the observed dietary-induced changes in the ovarian IGF system increase bioavailability of intrafollicular IGF, thus increasing the sensitivity of follicles to FSH. These changes, in combination with increased peripheral concentrations of insulin and IGF-I in heifers offered the HE diet, contribute to the observed increase in growth rate of the dominant follicle. In contrast to follicular growth, increased nutrient supply decreased oocyte quality, due in part to increased plasma urea concentrations. Clearly a number of mechanisms are involved in mediating the effects of dietary energy and protein on ovarian function, and the formulation of diets designed to optimize cattle fertility must consider the divergent effects of nutrient supply on follicular growth and oocyte quality.

Factors determining competence of in vitro produced cattle embryos

Concerns have developed in regard to problems associated with pregnancies and calves produced after use of cattle blastocysts made in the laboratory for embryo transfer. For both empirical studies and commercial purposes, there is a need for assurance that the product of these biotechnologies results in a normally functioning entity of its kind. Ability to use more genetic material from a donor female and in producing blastocysts needs to be improved to increase the efficiency of utilizing in vitro biotechnologies in animal production agriculture and for biomedical purposes. The role of gametes used as raw materials for laboratory production cattle embryos and adequacy of culture systems in supporting development of embryos are discussed in relation to competency of embryos produced in vitro.

Practical measures to improve in vitro blastocyst production in the bovine

Media, chemicals, sera and protocols used for in vitro production of bovine embryos are different from one laboratory to another. This paper describes some of the critical steps required to produce embryos in vitro, suggests quality control measures to lower variations in blastocyst yield and describes different tests which may be used when implementing new procedures in a routine production system.

Responses of oocytes and embryos to the culture environment

Embryo development is strongly influenced by events occurring during oocyte maturation. Although many immature oocytes are capable of completing meiosis in vitro, only a small percentage of the original pool of immature oocytes is competent to continue development to the blastocyst stage and subsequently result in a pregnancy. This indicates that maturation of oocytes in vitro may not be occurring in an entirely normal manner. Cytoplasmic changes occurring during maturation, collectively termed cytoplasmic maturation, are essential for embryonic development. The cytoplasm of the oocyte may play a crucial role in assembling the correct metabolic machinery for production of sufficient energy for cellular functions during maturation, cleavage and blastocyst formation. A better understanding of the structural, functional and metabolic characteristics of the oocyte during maturation, and the consequence of changes in these parameters on developmental competence is needed. Understanding the role of cytoplasmic changes during oocyte maturation will help increase the efficiency of in vitro embryo production. Better embryo production strategies will facilitate basic research into the control of early development, improve implementation in endangered species, provide a source of high quality oocytes for nuclear transfer and transgenic technologies and benefit the commercial embryo transfer industry.

Culture of blastomeres from in vitro-matured, fertilized, and cultured bovine embryos

A culture system was devised to study the differentiation of bovine blastomeres. Blastomeres (2-13 per well) from embryos produced by in vitro maturation, fertilization, and culture of oocytes obtained from slaughterhouse ovaries were cultured for 10 days in 24-well culture plates on feeder layers in blastomere culture medium (BCM: equal parts tissue culture medium 199 and low-glucose Dulbecco's modified Eagle's medium with 10% fetal bovine serum). Ovine embryonic fibroblasts and STO cells were superior to bovine and mouse embryonic fibroblasts as mitotically inactivated feeder cells. Over five studies in which four blastomeres from an embryo were added to each culture well, an average of one colony per well formed from the blastomeres. The colonies continued to grow throughout the culture period, and most colonies resembled trophectoderm in their cellular characteristics, although some cultures contained a mixture of trophectoderm and endoderm. When the number of blastomeres cultured in each well was varied from 2-8, the number of colonies formed was proportional to the number of blastomeres added. Blastomeres from day 5 and day 6 embryos produced fewer colonies than did those from day 4 embryos, perhaps as a result of differentiation and tighter blastomere adhesion resulting in damage during their separation. The absence of serum did not alter the number of colonies formed. A number of growth factors, including LIF, OM, PGDF alpha, and FGF4, had no effect on the number of colonies, the size of colonies, or their alkaline phosphatase staining score beyond that provided by the feeder layer on serum when present. Blastomeres did not form colonies in the absence of feeder layers.

A comparison of Menezo's B2 and tissue culture Medium-199 for in vitro production of bovine blastocysts

The objectives of this study were, first, to evaluate the effectiveness of 2 culture media, Menezo's B2 (B2) and Tissue Culture Medium-199 (M-199), for the production of bovine blastocysts in a commercial embryo transfer program; and, second, to characterize the stage of development, quality grade and cell number of blastocysts produced in each medium. One-cell bovine embryos were produced using in vitro maturation and fertilization procedures. After fertilization, the embryos were co-cultured on Buffalo rat liver (BRL) cell monolayers in either B2 or M-199+1% BSA (M-199) medium. Both media were supplemented with 10% fetal calf serum (FCS) and penicillin/streptomycin. Embryo cultures were continued undisturbed to either Day 7 or Day 8 post-insemination. In the Day 7 cultures, all blastocysts were removed for evaluation on Day 7, and the remaining embryos were cultured for a further 24 h. Any additional blastocysts that formed were removed for evaluation and designated as Day 8 disturbed embryos. All blastocysts were classified for stage and quality grade. Embryos were fixed and stained for determination of cell number. Overall, the proportion of blastocysts was greater (P = 0.0003) with B2 medium (46%) than with M-199 (33%). This was due to a larger (P = 0.0001) proportion of blastocysts produced in B2 medium when cultures were left undisturbed for 8 d (50 vs 28% for B2 vs M-199). The proportion of blastocysts on Day 7 of culture tended to differ (P = 0.073) between media (33 vs 24% for B2 vs M-199). In addition, there were more (P = 0.007) blastocysts at advanced stages of development in B2 medium on Day 7. There was no effect of type of medium on the distribution of embryo quality grades on any day examined. The number of cells per blastocyst did not differ between media but did vary significantly (P < .05) with both stage and grade. In conclusion, B2 medium was superior to M-199 medium when used in a co-culture system with BRL cells for the production of bovine blastocysts.

Evaluation of mitomycin-treated vero cells as a co-culture system for IVM/IVF-derived bovine embryos

Support of the in vitro development of IVM/IVF-derived bovine embryos by Vero cells was evaluated by comparing the following treatment groups: 1) proliferating (Unt-Vero) vs nonproliferating (Mit-Vero) cells; 2) supplementation of medium with estrous cow serum (ECS) vs bovine serum albumin (BSA); 3) Mit-Vero cells vs bovine oviduct epithelial cells (BOECs); and 4) addition of leukemia inhibitory factor (LIF) to Mit-Vero cell co-cultures at Day 1 vs Day 4. Mit-Vero cells stimulated higher rates of blastocysts (Day 7, 40 vs 27%) and hatched blastocyst (Day 10, 38 vs 12%) formation than Unt-Vero cells. These rates were comparable to those obtained with BOECs; blastocyst hatching was slightly higher following co-culture with Mit-Vero cells (36%) than BOECs (29%). Blastocyst formation was similar in ECS- vs BSA-supplemented medium; however, hatching was greatest (37%) during co-culture in medium +10% ECS. While the addition of LIF throughout the co-culture period was ineffective, addition of the cytokine beginning at Day 4 slightly increased blastocyst formation rates. Evaluation of LIF secretion using ELISA revealed detectable levels of the cytokine in Mit-Vero-conditioned medium (50 pg/10(5) cells); this may explain the minimal influence of exogenous LIF during embryo co-culture. Mit-Vero cells provided comparable support of bovine embryo development when used even up to 2 wk after establishment as monolayers. In conclusion, Mit-Vero cells provide a readily-available, safe and easy-to-use co-culture method which is at least as supporting of bovine embryo development as BOECs. One contribution of these cells may be secretion of the cytokine LIF.

Cell allocation to the inner cell mass and the trophectoderm in bovine embryos cultured in two different media

Data from other laboratories have shown that speed of bovine blastocyst development is higher when Ménézo B2 is used for coculture compared to TCM199. It was our purpose to investigate whether this early blastocyst formation was also indicative of embryo quality by studying the allocation of inner cells in embryos generated by B2-coculture and by TCM199-coculture. For this purpose, a differential staining technique was used. General embryo development was similar for TCM199- and B2-embryos expressed as rate of cleavage at day 3 and morula-blastocyst formation at day 8 (P > 0.05), but significantly different when expressed as number of eight-cell stages at day 3 and expanded or hatched blastocysts at day 8 (P < 0.01). B2-embryos cultured until day 5, 6, and 7 post insemination, had total cell numbers of 24, 65, and 109 respectively, which was significantly higher than the cell number of TCM199-embryos cultured over the same time period (18, 41, and 71 respectively, P < 0.001). Morphological differentiation was significantly more advanced for B2-embryos at day 7 and 8 (P < 0.0001 and P < 0.001, respectively). First presumptive inner cells appeared in eight- to 16-cell stages at day 3. Because the determination of inner cells by differential staining is depending upon the presence of functional tight junctions, we concluded that the establishment of the tight junction seal in B2-embryos differed from that in TCM199-embryos: Inner cells appeared 0.56 cell cycle later in B2-embryos (P < 0.001) and a larger variation existed in the number of ICM-cells in B2-blastocysts (P < 0.001). The higher total cell number of B2-expanded blastocysts was mainly acquired by trophectoderm growth (P < 0.06). These data indicate that the apparent better quality of B2-embryos (faster cleavage, earlier blastocyst formation) is not reflected in a reliable number of inner cells of B2-blastocysts.

Developmental competence of bovine embryos derived from oocytes collected at various stages of the estrous cycle

The developmental competence of bovine oocytes collected from donors at various stages of the estrous cycle and fertilized in vitro was investigated by comparing the yields of embryos obtained from oocytes isolated from the ovaries of cows slaughtered on estrous cycle Days 7 and 14, 8 and 15, 9 and 16 and on Days 19, 20 and 2. The percentages of oocytes that developed into blastocysts at Day 8 after exposure to spermatozoa were: 11.9 vs 20.0; 13.2 vs 30.5; 20.8 vs 29.8; and 11.7, 4.4 and 16.9, respectively. A significantly higher proportion of oocytes developed into blastocysts following isolation on cycle Days 14 to 16 (24.3 %) than following recovery on Days 7 to 9 (13.0 %; P < 0.05), Days 19 to 20 (6.6 %; P < 0.05) or Day 2 (16.9 %; P < 0.05). Embryo development was also faster in oocytes isolated at the end of the luteal phase (Days 14 to 16). These results demonstrate that the stage of the estrous cycle may influence the developmental potential of oocytes and in vitro embryo production.

Culture of in vitro fertilized bovine embryos with bovine oviductal epithelial cells, Buffalo rat liver (BRL) cells, or BRL-cell-conditioned medium

Co-culture with various cell types can enhance development of bovine embryos, especially through the transition from maternal to embryonic mRNA utilization, a stage of growth refractory to most in vitro methods. Bovine oviductal epithelial (BOE) cells have been particularly successful for culturing embryos through the refractory stage; however, Buffalo rat liver (BRL) cells are a readily available, long-lived, easy-to-care-for alternative. This study compared the embryotrophic activity of BOE to BRL cells with particular emphasis on the transition stage of growth. A total of 7158 immature bovine oocytes, matured and fertilized in vitro, were divided into 4 different culture treatments: Treatment 1: BRL conditioned medium for 72 h then BRL co-culture; Treatment 2: BRL co-culture; Treatment 3: BOE co-culture for 72 h in 5% oxygen then BRL co-culture; and Treatment 4: BOE co-culture for 72 h in 5% oxygen followed by BOE co-culture in air. Those same treatments were used to evaluate embryotrophic differences of early (4 to 5) versus late (14 to 15) passage BRL cells maintained in M-199 medium with 10% serum. Two bulls were also evaluated to determine if there exists a bull-by-culture system interaction. Treatment 3 resulted in the best development after 9 d; 9.1% of selected immature oocytes developed to expanded blastocyst. Early passage BRL cells were significantly more embryotrophic than later passage cells; this was most pronounced for Treatment 2. There was a treatment-by-bull interaction, which should be considered when comparing results among similar studies.

Development of early bovine embryos in co-culture with ksom and taurine, superoxide dismutase or insulin

Three experiments, utilizing 2578 embryos, were designed to test the effects of media, taurine, Superoxide dismutase and insulin on the development of embryos produced by in vitro maturation and in vitro fertilization (IVM/IVF). Embryos showing at least 1 cleavage during culture for 40 to 44 h after IVM/IVF were selected for further culture under various conditions for 6 d at 39 degrees C in 5% C0(2):95% air. A Buffalo rat liver (BRL) cell co-culture was used in all 3 experiments. Experiment 1 was a 3 x 2 factorial arrangement with KSOM (a high potassium simplex optimization-derived medium containing only 12 ingredients), Menezo B(2) and TCM-199 media with or without 7 mM taurine. Blastocyst production in the 3 media, respectively, was 48, 36 and 29% (P<0.05). Addition of 7 mM taurine increased the percentage of blastocysts from 34 to 42 (P<0.05). In Experiment 2, Superoxide dismutase (SOD) did not improve blastocyst development (P>0.05). In Experiment 3, insulin (75 ng/ml) added to KSOM resulted in 46% morulae plus blastocysts compared with 35% for the control (P<0.05). These results indicate that the co-culture of embryos in KSOM with taurine or insulin added is superior to commonly used complex media for efficient production of blastocysts following IVM/IVF.

In vitro pluripotency of epiblasts derived from bovine blastocysts

Two experiments were conducted to compare the utility of in vitro- and in vivo-derived bovine blastocysts for the isolation of pluripotent epiblasts. In experiment 1, the inner cell masses (ICMs) of in vivo-collected blastocysts yielded a higher proportion of epiblasts after culture on STO feeder cells than ICMs from in vitro-produced blastocysts (P = .0157). In experiment 2, ICMs of in vivo-collected blastocysts that hatched on day 8 yielded a greater proportion of epiblasts after culture on STO feeder cells than ICMs from in vitro-produced blastocysts that hatched on day 8. The difference was reversed but smaller for blastocysts that hatched on day 9 (Interaction, P = .0125). Epiblasts from blastocysts that hatched on day 8 regardless of their source generated more differentiated cell lines in extended culture than did blastocysts that hatched on day 9. Extended epiblast culture yielded cells identifiable as products of the three embryonic germ layers that included epithelial cells, fibroblasts, neuronal cells, hepatocyte-like cells, and macrophage-like cells. Alkaline phosphatase activity combined with cell morphology identified the bovine epiblast cells and distinguished them from trophectoderm and endoderm that frequently contaminated epiblast cell cultures. In vivo-derived blastocysts, especially from early-hatching blastocysts, were a superior source of pluripotent epiblasts. Epiblast cells in this study all differentiated or senesced indicating that standard conditions for mouse embryonic stem cell culture do not maintain bovine epiblast cells in an undifferentiated state.