The conserved L1089 in the S2 subunit of avian infectious bronchitis virus determines viral kidney tropism by disrupting virus-cell fusion.

The virulence of avian gamma-coronavirus infectious bronchitis viruses (IBV) for the kidney has led to high mortality in dominant-genotype isolations, but the key sites of viral protein that determine kidney tropism are still not fully clear. In this study, the amino acid sequences of the S2 subunit of IBVs with opposing adaptivity to chicken embryonic kidney cells (CEKs) were aligned to identify putative sites associated with differences in viral adaptability. The S2 gene and the putative sites of the non-adapted CN strain were introduced into the CEKs-adapted SczyC30 strain to rescue seven mutants. Analysis of growth characteristics showed that the replacement of the entire S2 subunit and the L1089I substitution in the S2 subunit entirely abolished the proliferation of recombinant IBV in CEKs as well as in primary chicken oviduct epithelial cells. Pathogenicity assays also support the decisive role of this L1089 for viral nephrotropism, and this non-nephrotropic L1089I substitution significantly attenuates pathogenicity. Analysis of the putative cause of proliferation inhibition in CEKs suggests that the L1089I substitution affects neither virus attachment nor endocytosis, but instead fails to form double-membrane vesicles to initiate the viral replication and translation. Position 1089 of the IBV S2 subunit is conservative and predicted to lie in heptad repeat 2 domains. It is therefore reasonable to conclude that the L1089I substitution alters the nephrotropism of parent strain by affecting virus-cell fusion. These findings provide crucial insights into the adaptive mechanisms of IBV and have applications in the development of vaccines and drugs against IB.

Chondroitin sulfate E downregulates N-cadherin and suppresses myotube formation.

Myogenesis, the formation of muscle fibers, is affected by certain glycoproteins, including chondroitin sulfate (CS), which are involved in various cellular processes. We aimed to investigate the mechanism underlying CS-E-induced suppression of myotube formation using the myoblast cell line C2C12. Differentiated cells treated with 0.1 mg/ml CS-E for nine days showed multinucleated and rounded myotubes with myosin heavy chain positivity. No difference was found between the CS-E-treated group with rounded myotubes and CS (-) controls with elongated myotubes in the levels of phospho-cofilin, a protein involved in the dynamics of actin cytoskeleton. Interestingly, N-cadherin, which is involved in the gene expression of myoblast fusion factors (myomaker and myomixer), was significantly downregulated at both the mRNA and protein levels following CS-E treatment. These results suggest that N-cadherin downregulation is one of the mechanisms underlying the CS-E-induced suppression of myotube formation.

Interspecies cell fusion between mouse embryonic stem cell and porcine pluripotent cell.

In the area of stem cell research, fusion of somatic cells into pluripotent cells such as mouse embryonic stem (ES) cells induces reprogramming of the somatic nucleus and can be used to study the effect of trans-acting factors from the pluripotent cell on the pluripotent state of somatic nucleus. As many other groups, we previously established a porcine pluripotent cell line at a low potential. Therefore, here, we performed experiments to investigate if the fusion with mouse ES cell could improve the pluripotent state of porcine pluripotent cell. Our data showed that resultant mouse-porcine interspecies fused cells are AP positive, and could be passaged up to 20 passages. Different degrees of increases in expression of porcine pluripotent genes proved that pig-origin gene network can be programmed by mouse ES. Further differentiation study also confirmed these fused cells' potential to form three germ layers. However, unexpectedly, we found that chromosome loss and aberrant (especially in porcine chromosomes) is severe after the cell fusion, implying that interspecies cell fusion may be not suitable to study porcine pluripotency without additional supportive conditions for genome stabilization.

Phytohemagglutinin facilitates the aggregation of blastomere pairs from Day 5 donor embryos with Day 4 host embryos for chimeric bovine embryo multiplication.

Multiplication of bovine embryos by the production of aggregation chimeras is based on the concept that few blastomeres of a donor embryo form the inner cell mass (ICM) and thus the embryo proper, whereas cells of a host embryo preferentially contribute to the trophectoderm (TE), the progenitor cells of the embryonic part of the placenta. We aggregated two fluorescent blastomeres from enhanced green fluorescent protein (eGFP) transgenic Day 5 morulae with two Day 4 embryos that did not complete their first cleavage until 27 hours after IVF and tested the effect of phytohemagglutinin-L (PHA) on chimeric embryo formation. The resulting blastocysts were characterized by differential staining of cell lineages using the TE-specific factor CDX2 and confocal laser scanning microscopy to facilitate the precise localization of eGFP-positive cells. The proportions of blastocyst development of sandwich aggregates with (n = 99) and without PHA (n = 46) were 85.9% and 54.3% (P < 0.05), respectively. Epifluorescence microscopy showed that the proportion of blastocysts with eGFP-positive cells in the ICM was higher in the PHA group than in the no-PHA group (40% vs. 16%; P < 0.05). Confocal laser scanning microscopy revealed that the total cell numbers of blastocysts from the PHA group of aggregation chimeras (n = 17; 207.8 ± 67.3 [mean ± standard deviation]) were higher (P < 0.05) than those of embryos without ZP and exposed to PHA (n = 30; 159.6 ± 42.2) and of handling control embryos (n = 19; 176.9 ± 53.3). The same was true for ICM cell counts (56.5 ± 22.0 vs. 37.7 ± 14.2 and 38.7 ± 12.4) and TE cell counts (151.2 ± 58.0 vs. 121.9 ± 37.4 and 138.3 ± 53.0), whereas the ICM/total cell number ratio was not different between the groups. Of the 17 chimeric blastocysts analyzed by confocal laser scanning microscopy, nine had eGFP-positive cells (three of them in the ICM, three in the TE, and three in both lineages). When integration in the ICM occurred, the number of eGFP-positive cells in this compartment was 8.3 ± 2.3 (mean ± standard error of the mean). We conclude that PHA is advantageous for the formation of aggregation chimeras, but the approach tested in the present study with only two donor blastomeres and two host embryos did not result in multiplication of genetically valuable donor embryos.

Factors affecting the development of somatic cell nuclear transfer embryos in Cattle.

Nuclear transfer is a complex multistep procedure that includes oocyte maturation, cell cycle synchronization of donor cells, enucleation, cell fusion, oocyte activation and embryo culture. Therefore, many factors are believed to contribute to the success of embryo development following nuclear transfer. Numerous attempts to improve cloning efficiency have been conducted since the birth of the first sheep by somatic cell nuclear transfer. However, the efficiency of somatic cell cloning has remained low, and applications have been limited. In this review, we discuss some of the factors that affect the developmental ability of somatic cell nuclear transfer embryos in cattle.

Effects of trichostatin A on In vitro development and DNA methylation level of the satellite I region of swamp buffalo (Bubalus bubalis) cloned embryos.

Trichostatin A (TSA), a histone deacetylase inhibitor, has been widely used to improve the cloning efficiency in several species. This brings our attention to investigation of the effects of TSA on developmental potential of swamp buffalo cloned embryos. Swamp buffalo cloned embryos were produced by electrical pulse fusion of male swamp buffalo fibroblasts with swamp buffalo enucleated oocytes. After fusion, reconstructed oocytes were treated with 0, 25 or 50 nM TSA for 10 h. The results showed that there was no significant difference in the rates of fusion (82-85%), cleavage (79-84%) and development to the 8-cell stage (59-65%) among treatment groups. The highest developmental rates to the morula and blastocyst stages of embryos were found in the 25 nM TSA-treated group (42.7 and 30.1%, respectively). We also analyzed the DNA methylation level in the satellite I region of donor cells and in in vitro fertilized (IVF) and cloned embryos using the bisulfite DNA sequencing method. The results indicated that the DNA methylation levels in cloned embryos were significantly higher than those of IVF embryos but approximately similar to those of donor cells. Moreover, there was no significant difference in the methylation level among TSA-treated and untreated cloned embryos. Thus, TSA treatments at 25 nM for 10 h could enhance the in vitro developmental potential of swamp buffalo cloned embryos, but no beneficial effect on the DNA methylation level was observed.

Production of chimeric embryos by aggregation of bovine egfp eight-cell stage blastomeres with two-cell fused and asynchronic embryos.

Embryo disaggregation allows the production of two to four identical offspring from a single cow embryo. In addition, embryo complementation has become the technique of choice to demonstrate the totipotency of embryonic stem cells and induced pluripotent stem cells. Therefore, the aim of this study was to generate a new and simple method by aggregation in the well-of-the-well system to direct each single enhanced green fluorescent protein (egfp) eight-cell blastomere derived from bovine in vitro fertilization embryos to the inner cell mass (ICM) of chimeras produced with fused and asynchronic embryos. To this end, the best conditions to generate in vitro fertilization-fused embryos were determined. Then, the fused (F) and nonfused (NF) embryos were aggregated in two distinct conditions: synchronically (S), with both transgenic and F embryos produced on the same day, and asynchronically (AS), with transgenic embryos produced one day before F embryos. The highest fusion and blastocysts rates were obtained with two pulses of 40 V. The 2ASF and 2ASNF groups showed the best number of blastocysts expressing the EGFP protein (48% and 41%, respectively). Furthermore, the 2ASF group induced the highest localization rates of the egfp-expressing blastomere in the ICM (6/13, 46% of ICM transgene-expressing blastocysts). This technique will have great application for multiplication of embryos of high genetic value or transgenic embryos and also with the generation of truly bovine embryonic stem cells and induced pluripotent stem cells.

Pluripotent cells in farm animals: state of the art and future perspectives.

Pluripotent cells, such as embryonic stem (ES) cells, embryonic germ cells and embryonic carcinoma cells are a unique type of cell because they remain undifferentiated indefinitely in in vitro culture, show self-renewal and possess the ability to differentiate into derivatives of the three germ layers. These capabilities make them a unique in vitro model for studying development, differentiation and for targeted modification of the genome. True pluripotent ESCs have only been described in the laboratory mouse and rat. However, rodent physiology and anatomy differ substantially from that of humans, detracting from the value of the rodent model for studies of human diseases and the development of cellular therapies in regenerative medicine. Recently, progress in the isolation of pluripotent cells in farm animals has been made and new technologies for reprogramming of somatic cells into a pluripotent state have been developed. Prior to clinical application of therapeutic cells differentiated from pluripotent stem cells in human patients, their survival and the absence of tumourigenic potential must be assessed in suitable preclinical large animal models. The establishment of pluripotent cell lines in farm animals may provide new opportunities for the production of transgenic animals, would facilitate development and validation of large animal models for evaluating ESC-based therapies and would thus contribute to the improvement of human and animal health. This review summarises the recent progress in the derivation of pluripotent and reprogrammed cells from farm animals. We refer to our recent review on this area, to which this article is complementary.

Effect of post-fusion holding time, orientation and position of somatic cell-cytoplasts during electrofusion on the development of handmade cloned embryos in buffalo (Bubalus bubalis).

The present study was conducted primarily to optimize electrofusion conditions for efficient production of zona-free nuclear transfer embryos in buffalos (Bubalus bubalis). We found that 4V AC current for proper triplet alignment and single step fusion method, using a single DC pulse of 3.36 kV/cm for 4-µs duration, produced the most convincing results for efficient reconstitution of zona-free cloned embryos. Lysis rate was very high (84.28 ± 2.59%) when triplets were in physical contact with negative electrode after applying DC current, however, cleavage rate and blastocyst rate were found to be similar when the triplets were not in physical contact with either positive or negative electrodes or when they were in physical contact with the positive electrode. Significant improvement in blastocyst production was observed when the somatic cell faced the positive electrode than when it faced the negative electrode (39.17 ± 2.74% vs. 25.91 ± 2.00%, respectively) during electrofusion. Similarly, the blastocyst rate (52.0 ± 3.4%) was found to be significantly higher when reconstructed embryos were activated 6 h post electrofusion as compared to 0, 2, 4 and 8 h (16.04 ± 6.3%; 18.36 ± 1.4%; 22.44 ± 3.7% and 30.02 ± 4.6%, respectively). This study establishes the application of zona-free nuclear transfer procedures for the production of handmade cloned buffalo embryos through optimization of electrofusion parameters and post fusion holding time for enhancing their preimplantation development.

Nuclear-cytoplasmic incompatibility and inefficient development of pig-mouse cytoplasmic hybrid embryos.

Inter-species somatic cell nuclear transfer (iSCNT) embryos usually fail to develop to the blastocyst stage and beyond due to incomplete reprogramming of donor cell. We evaluated whether using a karyoplast that would require less extensive reprogramming such as an embryonic blastomere or the meiotic spindle from metaphase II oocytes would provide additional insight into the development of iSCNT embryos. Our results showed that karyoplasts of embryonic or oocyte origin are no different from somatic cells; all iSCNT embryos, irrespective of karyoplast origin, were arrested during early development. We hypothesized that nuclear-cytoplasmic incompatibility could be another reason for failure of embryonic development from iSCNT. We used pig-mouse cytoplasmic hybrids as a model to address nuclear-cytoplasmic incompatibility in iSCNT embryos. Fertilized murine zygotes were reconstructed by fusing with porcine cytoplasts of varying cytoplasmic volumes (1/10 (small) and 1/5 (large) total volume of mouse zygote). The presence of pig cytoplasm significantly reduced the development of mouse zygotes to the blastocyst stage compared with control embryos at 120 h post-human chorionic gondotropin (41 vs 6 vs 94%, P<0.05; 1/10, 1/5, control respectively). While mitochondrial DNA copy numbers remained relatively unchanged, expression of several important genes namely Tfam, Polg, Polg2, Mfn2, Slc2a3 (Glut3), Slc2a1 (Glut1), Bcl2, Hspb1, Pou5f1 (Oct4), Nanog, Cdx2, Gata3, Tcfap2c, mt-Cox1 and mt-Cox2 was significantly reduced in cytoplasmic hybrids compared with control embryos. These results demonstrate that the presence of even a small amount of porcine cytoplasm is detrimental to murine embryo development and suggest that a range of factors are likely to contribute to the failure of inter-species nuclear transfer embryos.

Emperipolesis-like invasion of neoplastic lymphocytes into hepatocytes in feline T-cell lymphoma.

Twelve cases of feline malignant lymphoma with emperipolesis-like invasion of neoplastic lymphocytes were examined microscopically, immunohistochemically and ultrastructurally. Intracytoplasmic invasion of neoplastic cells varied in severity between the cases, between hepatic lobules and between areas within the lobules. The number of infiltrating neoplastic cells ranged from one to several per hepatocyte. Neoplastic cells exhibited widely varying morphology from case-to-case and cell-to-cell within each case, and contained eosinophilic cytoplasmic granules in four cases. Immunohistochemical examination revealed that neoplastic cells in 11 of the 12 cases expressed one or both T-cell markers (CD3 and TIA-1). Diagnosis of T-cell lymphoma was also confirmed by assessment of clonality by polymerase chain reaction. Ultrastructural analysis revealed that the neoplastic lymphocytes were contained within an invagination of the cell membrane of the hepatocyte, rather than directly infiltrating into the cytoplasm of the cell. There was no evidence that the invasive neoplastic lymphocytes had a cytotoxic effect.

Effective donor cell fusion conditions for production of cloned dogs by somatic cell nuclear transfer.

As shown by the birth of the first cloned dog 'Snuppy', a protocol to produce viable cloned dogs has been reported. In order to evaluate optimum fusion conditions for improving dog cloning efficiency, in vivo matured oocytes were reconstructed with adult somatic cells from a female Pekingese using different fusion conditions. Fusion with needle vs chamber methods, and with low vs high pulse strength was compared by evaluating fusion rate and in vivo development of canine cloned embryos. The fusion rates in the high voltage groups were significantly higher than in the low voltage groups regardless of fusion method (83.5 vs 66.1% for the needle fusion method, 67.4 vs 37.9% for the fusion chamber method). After embryo transfer, one each pregnancy was detected after using the needle fusion method with high and low voltage and in the chamber fusion method with high voltage, whereas no pregnancy was detected using the chamber method with low voltage. However, only the pregnancy from the needle fusion method with high voltage was maintained to term and one healthy puppy was delivered. The results of the present study demonstrated that two DC pulses of 3.8 to 4.0 kV/cm for 15 µsec using the needle fusion method were the most effective method for the production of cloned dogs under the conditions of this experiment.

Definition of fusion medium and electric parameters for efficient zygote electrofusion in the Pacific oyster (Crassostrea gigas).

Cell electrofusion has been widely used in the induction of tetraploidy in mammals, but little attention has been paid in molluscs. This work pursued the establishment of fusion medium (ionic vs. non-ionic) and electric parameters in the electrofusion of Pacific oyster zygotes (prior to the completion of the first mitotic division), minimizing all deleterious effects possible to D-larval stage. The tested combinations of electric field intensity (Vcm(-1)) and number of square DC pulses applied (for 50 micros each) were (Voltage x N degrees pulses): 400 x 1, 400 x 2, 400 x 3 and 600 x 1, 600 x 2, 600 x 3. When pulses were applied for first time, it was determined that an ionic fusion medium (microfiltered seawater) offered better conditions than the non-ionic fusion media previously used (0.6 M sucrose or 0.6 M mannitol) in terms of embryo survival and lysis rates. In this fusion medium, two different combinations of electric parameters (3 square DC pulses of 400 Vcm(-1) for 50 micros each at 26 degrees C and 1 square DC pulse of 600 Vcm(-1) for 50 micros at 26 degrees C) offered the best technical results of fusion (57 and 79% respectively) and survival until D-larva (44 and 41% respectively). In conclusion, these electric parameters could be established, using seawater as electrofusion medium, for further approaches to evaluate individual ploidy and survival beyond spat.

Porcine circovirus type 2 (PCV2) induces cell proliferation, fusion, and chemokine expression in swine monocytic cells in vitro.

Granulomatous lymphadenitis is one of the pathognomonic lesions in post-weaning multisystemic wasting syndrome (PMWS)-affected pigs. This unique lesion has not been reported in direct association with viral infection in pigs. The objective of the present study was to evaluate whether porcine circovirus type 2 (PCV2) alone is able to induce functional modulation in porcine monocytic cells in vitro to elucidate its possible role in the development of granulomatous inflammation. It was found that the proliferation activity of blood monocytes (Mo) and monocyte-derived macrophages (MDM) was significantly enhanced by PCV2. During monocyte-macrophage differentiation, the PCV2 antigen-containing rate and formation of multinucleated giant cells (MGC) were significantly increased in MDM when compared to those in Mo. The MDM-derived MGC displayed a significantly higher PCV2 antigen-containing rate than did the mono-nucleated MDM. Supernatants from PCV2-inoculated MDM at 24 h post-inoculation induced an increased tendency of chemotactic activity for blood Mo. At the same inoculation time period, levels of mRNA expression of the monocytic chemokines, monocyte chemoattractant protein-1 and macrophage inflammatory protein-1, also significantly increased in PCV2-inoculated MDM. The results suggest that PCV2 alone may induce cell proliferation, fusion, and chemokine expression in swine monocytic cells. Thus, PCV2 itself may play a significant role in the induction of granulomatous inflammation in PMWS-affected pigs.

An allogeneic hybrid-cell fusion vaccine against canine mammary cancer.

Mammary cancer is among the most prevalent of canine tumors frequently resulting in death due to metastatic disease. Most tumors fail to raise an effective immune reaction making improving immune recognition a priority. Hybrid-cell fusion strategies have been employed to load dendritic cell populations with tumor cell antigens to stimulate immune recognition; however, recovery, heterogeneity and quality of primary cells from patients present enormous challenges. We employed allogeneic cell lines to develop an improved hybrid-cell fusion strategy and evaluated immune reactions in normal laboratory beagles. Such a strategy relies on enhanced immune recognition of allogeneic tumor cell antigens by antigen presenting cells. Optimized PEG-promoted fusions between uniquely stained canine mammary tumor CMT12 or CMT28 cells and a dendritic cell-like DH82 cell fusion partner resulted in greater than 40% hybrid-cell fusion populations by flow cytometry and fluorescence microscopy. Hybrid-cell fusions were delivered by direct ultrasound guided injection into popliteal lymph nodes of laboratory beagles. Only hybrid-cell fusions provided statistically significant enhancement of cell-mediated immunity ((51)Cr-release assay) compared to innate reactions in naïve vehicle injected dogs while dogs vaccinated with either single cell component alone did not. Vaccination with hybrid-cell fusions enhanced IFN-gamma expression in sorted CD8+ and CD4+ cells but not in CD4-/CD8- cells consistent with a CTL response. Cell-mediated immune assays revealed strong reactions against matched (vaccine component) CMT cells and unmatched CMT cells indicative of an immune response to mammary cancer antigens common to both cell lines. These results provide proof of principle for development of an allogeneic vaccination strategy against canine mammary cancer.

The cell agglutination agent, phytohemagglutinin-L, improves the efficiency of somatic nuclear transfer cloning in cattle (Bos taurus).

One of the several factors that contribute to the low efficiency of mammalian somatic cloning is poor fusion between the small somatic donor cell and the large recipient oocyte. This study was designed to test phytohemagglutinin (PHA) agglutination activity on fusion rate, and subsequent developmental potential of cloned bovine embryos. The toxicity of PHA was established by examining its effects on the development of parthenogenetic bovine oocytes treated with different doses (Experiment 1), and for different durations (Experiment 2). The effective dose and duration of PHA treatment (150 microg/mL, 20 min incubation) was selected and used to compare membrane fusion efficiency and embryo development following somatic cell nuclear transfer (Experiment 3). Cloning with somatic donor fibroblasts versus cumulus cells was also compared, both with and without PHA treatment (150 microg/mL, 20 min). Fusion rate of nuclear donor fibroblasts, after phytohemagglutinin treatment, was increased from 33 to 61% (P < 0.05), and from 59 to 88% (P < 0.05) with cumulus cell nuclear donors. The nuclear transfer (NT) efficiency per oocyte used was improved following PHA treatment, for both fibroblast (13% versus 22%) as well as cumulus cells (17% versus 34%; P < 0.05). The cloned embryos, both with and without PHA treatment, were subjected to vitrification and embryo transfer testing, and resulted in similar survival (approximately 90% hatching) and pregnancy rates (17-25%). Three calves were born following vitrification and embryo transfer of these embryos; two from the PHA-treated group, and one from non-PHA control group. We concluded that PHA treatment significantly improved the fusion efficiency of somatic NT in cattle, and therefore, increased the development of cloned blastocysts. Furthermore, within a determined range of dose and duration, PHA had no detrimental effect on embryo survival post-vitrification, nor on pregnancy or calving rates following embryo transfer.

Development of a nonmechanical enucleation method using x-ray irradiation in somatic cell nuclear transfer.

Irradiation of in-vitro-matured bovine oocytes with x-rays of different durations was performed to develop an alternative to conventional mechanical enucleation methods in somatic cell nuclear transfer. No significant difference in embryo development to the blastocyst stage was detected between nonmechanical and mechanical methods, and cytologic analyses of karyotype and microtubule formation showed the potential availability of x-ray irradiation.

Establishment of a chicken monoclonal antibody panel against mammalian prion protein.

A panel of chicken monoclonal antibodies (mAbs) was developed against prion protein (PrP), the sequence of which is a highly conserved molecule among mammals. A portion of the splenocytes from chickens immunized with recombinant mouse PrP was fused with the chicken B cell line, MuH1. The remaining splenocytes were used to generate the recombinant mAbs by phage display. A total of 36 anti-PrP mAbs, 2 from cell fusion and 34 from phage display were established. The specificity of these mAbs was determined by Western blot and ELISA using various PrP antigens including recombinant PrPs, synthetic PrP peptides and PrPs from brains or scrapie-infected neuroblastoma cell line. These mAbs were classified into three main groups, protease K (PK)-sensitive (Group I), PK cleavage site proximal (Group II) and PK-resistant (Group III), based on their abilities to recognize PrP following PK-treatment. Some mAbs were found to selectively recognize different glycoforms of PrP as well as the metabolic fragments of PrP. Furthermore, we found that PrP recognition by chickens differed from that by PrP-knockout mouse. These results indicate that these newly generated PrP antibodies from chickens will help to research the PrP and to establish the diagnosis of prion disease.

Veterinary sources of nonrodent monoclonal antibodies: interspecific and intraspecific hybridomas.

The generation of monoclonal antibodies from species other than rats and mice has developed slowly over the last 20 years. The advent of antibody engineering and realization of the advantages of nonmurine antibodies, in terms of their superior affinities and specificities, and their potential as components of human and veterinary therapeutics has increased their relevance recently. There have been significant advances in the development of myeloma and heteromyeloma fusion partners. This is an opportune moment to consolidate experiences of MAb production across the range of species of veterinary interest and place it into context with other developments in the field of monoclonal antibodies. The background to the development of antibodies from species other than the mouse is discussed. The species and antigens used to date are reviewed, as are the methods and results reported. A suggested protocol is provided for first attempts to exploit the huge potential of this aspect of hybridoma technology and suggestions are made for its further expansion.