The signal peptide (SP) of integrated membrane proteins is removed cotranslationally or posttranslationally in the endoplasmic reticulum, while GP64, a membrane fusion protein of Bombyx mori nucleopolyhedrovirus (BmNPV), retains its SP in the mature protein and virion. In this study, we revealed that uncleaved SP is a key determinant with additional functions in infection. First, uncleaved SP endows BmNPV with strong virulence; second, SP retention-induced BmNPV infection depends on cholesterol recognition amino acid consensus domain 1 (CRAC1) and CRAC2. In contrast, the recombinant virus with SP-cleaved GP64 has reduced infectivity, and only CRAC2 is required for BmNPV infection. Furthermore, we showed that cholesterol in the plasma membrane is an important fusion receptor that interacts with CRAC2 of GP64. Our study suggested that BmNPV GP64 is a key cholesterol-binding protein and uncleaved SP determines GP64\'s unique dependence on the CRAC domains. IMPORTANCE BmNPV is a severe pathogen that mainly infects silkworms. GP64 is the key membrane fusion protein that mediates BmNPV infection, and some studies have indicated that cholesterol and lipids are involved in BmNPV infection. A remarkable difference from other membrane fusion proteins is that BmNPV GP64 retains its SP in the mature protein, but the cause is still unclear. In this study, we investigated the reason why BmNPV retains this SP, and its effects on protein targeting, virulence, and CRAC dependence were revealed by comparison of recombinant viruses harboring SP-cleaved or uncleaved GP64. Our study provides a basis for understanding the dependence of BmNPV infection on cholesterol/lipids and host specificity.

Estrogen receptor (ER) alpha is commonly thought to bind to a consensus estrogen response element (ERE) as a homodimer, but previous experiments have not ruled out the presence of other proteins in the ERalpha/ERE complex. To characterize this interaction in more detail, we overexpressed mouse (m) ERalpha in a baculovirus system, using the selective advantage of the apoptosis inhibitor p35. Recombinant mERalpha possesses the predicted molecular weight and binds 17beta-estradiol and an oligonucleotide containing a consensus vitellogenin ERE with high affinity. Over a wide concentration range of mERalpha protein (0.1-50 nM), only one complex was detected between mERalpha and vitellogenin ERE in gel shift assays. The ratio of E2:vitellogenin ERE bound by mERalpha was close to 2:1, and each complex contained only one ERE. The molecular weight of the complex was determined to be 160 000, very close to that predicted for two mERalpha proteins and one ERE oligonucleotide, therefore providing strong evidence that no other proteins were present. Recombinant mERalpha was purified such that it was the only protein observable by silver stain. Purified mERalpha and mERalpha in a nuclear extract behaved identically in Ferguson analysis, providing more evidence that only mERalpha was binding to the ERE. Purified mERalpha bound vitellogenin ERE with high affinity (Kd = 0. 92 +/- 0.20 nM), indicating that no other proteins are necessary for high-affinity mERalpha interaction with a consensus ERE. To determine whether ERalpha in an estrogen-responsive mammalian tissue behaves the same as the overexpressed mERalpha, we tested rat uterine cytosol by Ferguson analysis. ERalpha in rat uterine cytosol behaved identically to overexpressed mERalpha, suggesting that ERalpha in the uterine extract also binds to DNA predominantly as a homodimer with no additional proteins.

Synthetic genes encoding the 146 and 155 amino acid forms of human basic fibroblast growth factor (bFGF) were constructed with codon usage biased towards the polyhedrin-encoding gene of Autographa californica nuclear polyhedrosis virus (AcNPV). Expression of both bFGF genes in Spodoptera frugiperda (SF-21) suspension cell culture using a recombinant baculovirus yielded approximately 2.5 mg of mitogenically fully active protein per 10(9) cells following heparin-affinity chromatography. To improve translational efficiency, the Kozak consensus sequence was introduced and it was found that neither the replacement of a pyrimidine by a purine at position -3, nor the nature of the base at position +4 had any noticeable effect on the final levels of bFGF expression in SF-21 cells. The bases at these critical points in the consensus do not therefore play a major role in expression levels of the bFGF synthetic genes. The two synthetic genes were also expressed in Escherichia coli as native proteins using the T7 expression system. 5 mg of mitogenically fully active bFGF were obtained from 1 l of bacterial culture. Both insect cell- and E. coli-derived bFGF were equally mitogenic for Swiss 3T3 fibroblasts.