Signal peptide

A signal peptide is a short (3-60 amino acids long) peptide chain that directs the transport of a protein. Signal peptides may also be called targeting signals, signal sequences, transit peptides, or localization signals. There is some confusion relating to the precise meaning of the term 'signal peptide'. Some sources refer to signal peptides as only the pre-sequence that targets the propeptide to the endoplasmic reticulum and through the secretory pathway. Used in this system it does not refer to 'transit peptides'.[1] A 'transit peptide' used in this system refers to the part of the pre-sequence that targets the protein to other organelles, such as mitochondria, chloroplasts and apicoplasts. The amino acid sequences of signal peptides direct proteins (which are synthesized in the cytosol) to certain organelles such as the nucleus, mitochondrial matrix, endoplasmic reticulum, chloroplast, apoplast and peroxisome. Some signal peptides are cleaved from the protein by signal peptidase after the proteins are transported.

Types by protein destination

Endoplasmic reticulum
Almost all proteins that are transported to the endoplasmic reticulum have a sequence consisting of 5-10 hydrophobic amino acids on the N-terminus. The protein is guided to the ER by a signal-recognition particle (SRP), which moves between the ER and the cytoplasm. It binds to the signal peptide. The SRP binds to the signal peptide as soon as it is synthesized and extruded from the ribosome. This causes a pause in protein synthesis, most probably allowing the ribosome-SRP complex time to bind to the SRP receptor on the target ER membrane. The SRP is thought to be a regulatory GTP protein. Conformational changes may therefore lead to the SRP release. The protein may then be threaded through the ER membrane by a translocator pore. Most of these proteins are transported from the endoplasmic reticulum to the Golgi apparatus. If these proteins have a particular 4-amino-acid retention sequence, KDEL (lys-asp-glu-leu), on the C-terminus, they are retained in the endoplasmic reticulum or are routed back to the ER (in instances where they escape) via interaction with the KDEL receptor in the Golgi apparatus. Many records now,state that c-terminal KDEL in mammals, HDEL variants in yeast are conserved to be the re-router signals rather than ER retention signals.

Nucleus

A nuclear localization signal (NLS) is a signal peptide directing to the nucleus and is often a unit consisting of five basic, positively-charged amino acids. The NLS normally is located anywhere on the peptide chain.

Nucleolus
The nucleolus within the nucleus can be targeted with a sequence called a nucleolar localization signal (abbreviated NoLS or NOS).

Mitochondrial matrix
The signal peptide that directs to the mitochondrial matrix has a sequence consisting of an alternating pattern (amphipathic helix)with a few hydrophobic amino acids and a few positively charged amino acids at the N terminus. It is usually called the mitochondrial targeting signal (MTS).

Peroxisome
There are two types of signal peptides directing to peroxisome, which are called peroxisomal targeting signals (PTS). One is PTS1, which is made of three amino acids on the C-terminus. The other is PTS2, which is made of a 9-amino-acid sequence often present on the N-terminus of the protein.

Types
Following is a list of types of signal peptides:
y y

N-terminus signal peptides often target the mitochondrial matrix, endoplasmic reticulum and peroxisome. C-terminus signal peptides often target the peroxisome.

Examples of signal peptides

Transport to the nucleus (NLS) Transport to the endoplasmic reticulum Leu-Leu-Val-Gly-Ile-Leu-PheTrp-Ala-Thr-Glu-Ala-Glu-GlnLeu-Thr-Lys-Cys-Glu-Val-PheGlnRetention to the endoplasmic reticulum Transport to the mitochondrial matrix Thr-Arg-Thr-Leu-Cys-Ser-SerArg-Tyr-Leu-Leu-Pro-Pro-Lys-Lys-Lys-Arg-Lys-ValH2N-Met-Met-Ser-Phe-Val-Ser-Leu-

-Lys-Asp-Glu-Leu-COOH H2N-Met-Leu-Ser-Leu-Arg-Gln-SerIle-Arg-Phe-Phe-Lys-Pro-Ala-

Transport to the peroxisome (PTS1) Transport to the peroxisome (PTS2)

-Ser-Lys-Leu-COOH H2N-----Arg-Leu-X5-His-Leu-

H2N is the N-terminus of a protein. COOH is the C-Terminus of a protein.

Glycosylation
Glycosylation is the enzymatic process that attaches glycans to proteins, lipids, or other organic molecules. This enzymatic process produces one of the fundamental biopolymers found in cells (along with DNA, RNA, and proteins). Glycosylation is a form of co-translational and posttranslational modification. Glycans serve a variety of structural and functional roles in membrane and secreted proteins.[1] The majority of proteins synthesized in the rough ER undergo glycosylation. It is an enzyme-directed site-specific process, as opposed to the non-enzymatic chemical reaction of glycation. Glycosylation is also present in the cytoplasm and nucleus as the O-GlcNAc modification. Five classes of glycans are produced:
y y y y y

N-linked glycans attached to a nitrogen of asparagine or arginineside-chains O-linked glycans attached to the hydroxyoxygen of serine, threonine, tyrosine, hydroxylysine, or hydroxyproline side-chains, or to oxygens on lipids such as ceramide phospho-glycans linked through the phosphate of a phospho-serine; C-linked glycans, a rare form of glycosylation where a sugar is added to a carbon on a tryptophan side-chain glypiation, which is the addition of a GPI anchor that links proteins to lipids through glycan linkages.

Purpose
The carbohydrate chains attached to the target proteins serve various functions.[2] For instance, some proteins do not fold correctly unless they are glycosylated first.[1] Also, polysaccharides linked at the amide nitrogen of asparagine in the protein confer stability on some secreted glycoproteins. Experiments have shown that glycosylation in this case is not a strict requirement for proper folding, but the unglycosylated protein degrades quickly. Glycosylation may play a role in cell-cell adhesion (a mechanism employed by cells of the immune system), as well.

Mechanisms
There are various mechanisms for glycosylation, although most share several common features:[1]
y y

Glycosylation, unlike glycation, is an enzymatic process The donor molecule is often an activated nucleotide sugar

The process is site-specific.

Types of glycosylation

N-linked glycosylation

Comparative overview of the major types of vertebrate N-glycan subtypes and some representative C. elegans N-glycans. N-linked glycosylation is important for the folding of some eukaryotic proteins. The N-linked glycosylation process occurs in eukaryotes and widely in archaea, but very rarely in bacteria. In Eukaryotes, most N-linked oligosaccharides begin with addition of a 14-sugar precursor to an asparagine in the polypeptide chain of the target protein. The structure of this precursor is common to most eukaryotes, and contains 3 glucose, 9 mannose, and 2 N-acetylglucosamine molecules. A complex set of reactions attaches this branched chain to a carrier molecule called dolichol, and then it is transferred to the appropriate point on the polypeptide chain as it is translocated into the ER lumen.

There are three major classes of N-linked saccharides resulting from this core: high-mannose oligosaccharides, complex oligosaccharides and hybrid oligosaccharides.[1]
y

High-mannose is, in essence, just two N-acetylglucosamines with many mannose residues, often almost as many as are seen in the precursor oligosaccharides before it is attached to the protein. Complex oligosaccharides are so named because they can contain almost any number of the other types of saccharides, including more than the original two N-acetylglucosamines.

Proteins can be glycosylated by both types of oligosaccharides on different portions of the protein. Whether an oligosaccharide is high-mannose or complex is thought to depend on its accessibility to saccharide-modifying proteins in the Golgi. If the saccharide is relatively inaccessible, it will most likely stay in its original high-mannose form. If it is accessible, then it is likely that many of the mannose residues will be cleaved off and the saccharide will be further modified by the addition of other types of group as discussed above. The oligosaccharide chain is attached by oligosaccharyltransferase to asparagine occurring in the tripeptide sequence Asn-X-Ser or Asn-X-Thr where X could be any amino acid except Pro. This sequence is known as a glycosylation sequon. After attachment, once the protein is correctly folded, the three glucose residues are removed from the chain and the protein is available for export from the ER. The glycoprotein thus formed is then transported to the Golgi where removal of further mannose residues may take place. However, glycosylation itself does not seem to be as necessary for correct transport targeting of the protein, as one might think. Studies involving drugs that block certain steps in glycosylation, or mutant cells deficient in a glycosylation enzyme, still produce otherwise-structurally-normal proteins that are correctly targeted, and this interference does not seem to interfere severely with the viability of the cells. Mature glycoproteins may contain a variety of oligomannoseN-linked oligosaccharides containing between 5 and 9 mannose residues. Further removal of mannose residues leads to a 'core' structure containing 3 mannose, and 2 N-acetylglucosamine residues, which may then be elongated with a variety of different monosaccharides including galactose, N-acetylglucosamine, N-acetylgalactosamine, fucose and sialic acid. GalNAc, glucose, and rhamnose linked to asparagines have been observed as well, although mostly in less complex organisms or bacteria. Glucose linked to the guanidinium group of arginine in sweet corn amyelogenin is the only reported example of N-linked glycosylation on an amino acid other than asparagine.

O-linked glycosylation
O-linked glycosylation is a form of glycosylation occurring in the Golgi apparatus.[1] O-N-acetylgalactosamine (O-GalNAc)

O-linked glycosylation occurs at a later stage during protein processing, probably in the Golgi apparatus. This is the addition of N-acetyl-galactosamine to serine or threonine residues by the enzyme UDP-N-acetyl-D-galactosamine:polypeptide N-acetylgalactosaminyltransferase (EC2.4.1.41), followed by other carbohydrates (such as galactose and sialic acid). This process is important for certain types of proteins such as proteoglycans, which involves the addition of glycosaminoglycan chains to an initially unglycosylated "proteoglycan core protein." These additions are usually serine O-linked glycoproteins, which seem to have one of two main functions. One function involves secretion to form components of the extracellular matrix, adhering one cell to another by interactions between the large sugar complexes of proteoglycans. The other main function is to act as a component of mucosal secretions, and it is the high concentration of carbohydrates that tends to give mucus its "slimy" feel. Proteins that circulate in the blood are not normally O-glycosylated, with the exception of IgA1 and IgD (two types of antibody) and C1-inhibitor. O-fucose O-fucose is added between the second and third conserved cysteines of EGF-like repeats in the Notch protein, and other substrates by GDP-fucose protein O-fucosyltransferase 1, and to Thrombospondin repeats by GDP-fucose protein O-fucosyltransferase 2. In the case of EGF-like repeats, the O-fucose may be further elongated to a tetrasaccharide by sequential addition of Nacetylglucosamine (GlcNAc), galactose, and sialic acid, and for Thrombospondin repeats, may be elongated to a disaccharide by the addition of glucose. Both of these fucosyltransferases have been localized to the endoplasmic reticulum, which is unusual for glycosyltransferases, most of which function in the Golgi apparatus. O-glucose O-glucose is added between the first and second conserved cysteines of EGF-like repeats in the Notch protein, and possibly other substrates by protein:O-glucosyltransferase (Poglut). This enzyme is known as Rumi in Drosophila, and is also localized to the ER like the Ofucosyltransferases. The O-glucose modification appears to be necessary for proper folding of the EGF-like repeats of the Notch protein, and increases secretion of this receptor. O-N-acetylglucosamine (O-GlcNAc) O-GlcNAc is added to serines or threonines by O-GlcNActransferase. O-GlcNAc appears to occur on most serines and threonines that would otherwise be phosphorylated by serine/threonine kinases. Thus, if phosphorylation occurs, O-GlcNAc does not, and vice versa. This is an incredibly important finding because phosphorylation/dephosphorylation has become a scientific paradigm for the regulation of signaling within cells. A massive amount of cancer research is focused on phosphorylation. Ignoring the involvement of this form of glycosylation, which clearly appears to act in concert with phosphorylation, means that a lot of current research is missing at least half of the picture. O-GlcNAc addition and removal also appears to be a key regulator of the pathways that are disrupted in diabetes mellitus. The gene encoding the OGlcNAcase enzyme has been linked to non-insulin dependent diabetes mellitus. It is the terminal step in a nutrient-sensing hexosamine signaling pathway.

O-N-acetylglucosamine In Other Contexts Recently, O-GlcNAc was reported to occur between the fifth and sixth conserved cysteines in some EGF-like repeats from the Notch protein. It would seem unlikely that this modification would be due to the same enzyme involved with addition of O-GlcNAc to cytoplasmic and nuclear localized proteins. Considering that O-fucose and O-glucose addition to EGF-like repeats is due to ER localized enzymes, presumably an ER localized protein O-GlcNActransferase exists. O-mannose During O-mannosylation, a mannose residue is transferred from mannose-p-dolichol to a serine/threonine residue in secretory pathway proteins[2]. O-mannosylation is common to both prokaryotes and eukaryotes. Collagen Glycosylation Many lysines in collagen are hydroxylated to form hydroxylysine, and many of these hydroxylysines are then glycosylated by the addition of galactose. This galactose monosaccharide can then be further elongated by the addition of a glucose. This glycosylation is required for the proper functioning of collagen. Glycosylation of hydroxlysine occurs in the ER. Hydroxyproline Glycosylation Proline is also hydroxylated in collagen, however, no glycosylation occurs here as the hydroxyprolines are necessary for hydrogen bonding in the collagen triple helix. There is one protein named Skp1 in Dictyosteliumdiscoideum that carries a GlcNAc on hydroxyproline, but this would appear to be an extremely rare form of glycosylation. Otherwise, only plants appear to carry glycans on hydroxyproline, with both galactose and arabinoseglycans being reported in the literature. Glycosylation of Glycogenin Liver and muscle glycogenin carries a glucose on a tyrosine side chain. This is the only known example of glycosylated tyrosine in nature. Glycosylation of Ceramide Either a galactose or a glucose can be added to a hydroxyl on the lipid ceramide. The glucose can be further elongated to a disaccharide by the addition of a galactose. Proteoglycans The large and complex glycans that modify proteoglycans are initiated by addition of xylose to serine. This is the only form of glycan so far reported to begin with xylose addition directly to

protein apart from the xylose seen on phospho-serine in Dictyosteliumdiscoideum described below.

Phospho-serine glycosylation
Xylose, fucose, mannose, and GlcNAcphospho-serine glycans have been reported in the literature. Fucose and GlcNAc have been found only in Dictyosteliumdiscoideum, mannose in Leishmaniamexicana, and xylose in Trypanosomacruzi. Mannose has recently been reported in a vertebrate, the mouse, Musmusculus, on the cell-surface laminin receptor alpha dystroglycan4. It has been suggested this rare finding may be linked to the fact that alpha dystroglycan is highly conserved from lower vertebrates to mammals.[4]

C-mannosylation
A mannose sugar is added to the first tryptophan residue in the sequence W-X-X-W (W indicates tryptophan; X is any amino acid). Thrombospondins are one of the most commonly C-modified proteins, although this form of glycosylation appears elsewhere as well. C-mannosylation is unusual because the sugar is linked to a carbon rather than a reactive atom such as nitrogen or oxygen. Recently, the first crystal structure of a protein containing this type of glycosylation has been determined - that of human complement component 8, PDB ID 3OJY.

Formation of GPI anchors (glypiation)

A special form of glycosylation is the formation of a GPI anchor. In this kind of glycosylation a protein is attached to a lipid anchor, via a glycan chain. (See also prenylation.)

Protein targeting
Protein targeting or protein sorting is the mechanism by which a cell transports proteins to the appropriate positions in the cell or outside of it. Sorting targets can be the inner space of an organelle, any of several interior membranes, the cell's outer membrane, or its exterior via secretion. This delivery process is carried out based on information contained in the protein itself. Correct sorting is crucial for the cell; errors can lead to diseases.

Targeting signals
Targeting signals are the pieces of information that enable the cellular transport machinery to correctly position a protein inside or outside the cell. This information is contained in the

polypeptide chain or in the folded protein. The continuous stretch of amino acid residues in the chain that enables targeting are called signal peptides or targeting peptides. There are two types of targeting peptides, the presequences and the internal targeting peptides. The presequences of the targeting peptide are often found at the N-terminal extension and is composed of between 6-136 basic and hydrophobic amino acids.In case of peroxisomes the targeting sequence is on the C-terminal extension mostly. Other signals are composed by parts which are separate in the primary sequence. To function, these components have to come together on the protein surface by folding. They are called signal patches. In addition, protein modifications like glycosylations can induce targeting.

Protein translocation
In 1970, Gnter Blobel conducted experiments on the translocation of proteins across membranes. He was awarded the 1999 Nobel prize for his findings. He discovered that many proteins have a signal sequence, that is, a short amino acid sequence at one end that functions like a postal code for the target organelle. The translation of mRNA into protein by a ribosome takes place within the cytosol. If the synthesized proteins "belong" in a different organelle, they can be transported there in either of two ways, depending on the protein.

Cotranslational translocation
The N-terminal signal sequence of the protein is recognized by a signal recognition particle (SRP) while the protein is still being synthesized on the ribosome. The synthesis pauses while the ribosome-protein complex is transferred to a SRP receptor on the endoplasmic reticulum (ER), a membrane-enclosed organelle. There, the nascent protein is inserted into the Sec61 translocation complex (also known as the translocon) that passes through the ER membrane. In secretory proteins and type I transmembrane proteins, the signal sequence is immediately cleaved from the nascent polypeptide once it has been translocated into the ER by signal peptidase. The signal sequence of type II membrane proteins and some polytopic membrane proteins are not cleaved off and therefore are referred to as signal anchor sequences. Within the ER, the protein is first covered by a chaperone protein to protect it from the high concentration of other proteins in the ER, giving it time to fold correctly. Once folded, the protein is modified as needed (for example, by glycosylation), then transported to the Golgi apparatus for further processing and goes to its target organelles or is retained in the ER by various ER retention mechanisms.

Posttranslational translocation
Even though most proteins are cotranslationallytranslocated, some are translated in the cytosol and later transported to their destination. This occurs for proteins that go to a mitochondrion, a chloroplast, or a peroxisome (proteins that go to the latter have their signal sequence at the C terminus). Also, proteins targeted for the nucleus are translocated posttranslation. They pass through the nuclear envelope via nuclear pores.

Transmembrane proteins

The amino acid chain of transmembrane proteins, which often are transmembrane receptors, passes through a membrane one or several times. They are inserted into the membrane by translocation, until the process is interrupted by a stop-transfer sequence, also called a membrane anchor sequence. These complex membrane proteins are at the moment mostly understood using the same model of targeting that has been developed for secretory proteins. However, many complex multi-transmembrane proteins contain structural aspects that do not fit the model. Seven transmembrane G-protein coupled receptors (which represent about 5% of the genome of humans) mostly do not have an amino-terminal signal sequence. In contrast to secretory proteins, the first transmembrane domain acts as the first signal sequence, which targets them to the ER membrane. This also results in the translocation of the amino terminus of the protein into the ER membrane lumen. This would seem to break the rule of "cotranslational" translocation which has always held for mammalian proteins targeted to the ER. This has been demonstrated with opsin with in vitro experiments.[1][2] A great deal of the mechanics of transmembrane topology and folding remains to be elucidated.

Sorting of proteins to mitochondria

Most mitochondrialproteins are synthesized as cytosolic precursors containing uptake peptide signals. Cytosolicchaperones deliver preproteins to channel linked receptors in the mitochondrial membrane. The preprotein with presequence targeted for the mitochondria is bound by receptors and the General Import Pore (GIP) (Receptors and GIP are collectively known as Translocase of Outer Membrane or TOM) at the outer membrane. The preprotein is translocated through TOM as hairpin loops. The preprotein is transported through the intermembrane space by small TIMs (which also acts as molecular chaperones) to the TIM23 or 22 (Translocase of Inner Membrane) at the inner membrane. Within the matrix the targeting sequence is cleaved off by mtHsp70. Three mitochondrial outer membrane receptors are known: TOM20, TOM22 and TOM70 TOM70: Binds to internal targeting peptides and acts as a docking point for cytosolic chaperones. TOM20: Binds presequences TOM22: Binds both presequences and internal targeting peptides The TOM channel is a cation specific high conductance channel with a molecular weight of 410 kDa and a pore diameter of 21. The presequence translocase23 (TIM23) is localized to the mitochondialinner membrane and acts a pore forming protein which binds precursor proteins with its N-terminal. TIM23 acts a translocator for preproteins for the mitochondrial matrix, the inner mitochondrial membrane as well as for the intermembrane space. TIM50 is bound to TIM23 at the inner mitochondrial side and found to bind presequences. TIM44 is bound on the matrix side and found binding to mtHsp70. The presequence translocase22 (TIM22) binds preproteins exclusively bound for the inner mitochondrial membrane. Mitochondrial matrix targeting sequences are rich in positively charged amino acids and hydroxylated ones.

Proteins are targeted to submitochondrial compartments by multiple signals and several pathways. Targeting to the outer membrane, intermembrane space, and inner membrane often requires another signal sequence in addition to the matrix targeting sequence.

Sorting of proteins to chloroplasts

The preprotein for chloroplasts may contain a stromal import sequence or a stromal and thylakoid targeting sequence. The majority of preproteins are translocated through the Toc and Tic complexes located within the chloroplast envelope. In the stroma the stromal import sequence is cleaved off and folding as well as intra-chloroplast sorting to thylakoids continues. Proteins targeted to the envelope of chloroplasts usually lack cleavable sorting sequence.

Sorting of proteins to both chloroplasts and mitochondria

Many proteins are needed in both mitochondria and chloroplasts. In general the targeting peptide is of intermediate character to the two specific ones. The targeting peptides of these proteins have a high content of basic and hydrophobicamino acids, a low content of negatively charged amino acids. They have a lower content of alanine and a higher content of leucine and phenylalanine. The dual targeted proteins have a more hydrophobic targeting peptide than both mitochondrial and chloroplastic ones.

Sorting of proteins to peroxisomes

All peroxisomal proteins are encoded by nuclear genes. To date there are two types of known Peroxisome Targeting Signals (PTS): Peroxisome targeting signal 1 (PTS1): a C-terminal tripeptide with a consensus sequence (S/A/C)-(K/R/H)-(L/A). The most common PTS1 is serine-lysine-leucine (SKL). Most peroxisomal matrix proteins possess a PTS1 type signal. Peroxisome targeting signal 2 (PTS2): a nonapeptide located near the N-terminus with a consensus sequence (R/K)-(L/V/I)-XXXXX-(H/Q)-(L/A/F) (where X can be any amino acid). There are also proteins that possess neither of these signals. Their transport may be based on a so-called "piggy-back" mechanism: such proteins associate with PTS1-possessing matrix proteins and are translocated into the peroxisomal matrix together with them.

Diseases
Peroxisomal protein transport is defective in the following genetic diseases:

y y y

Zellweger syndrome. Adrenoleukodystrophy (ALD). Refsum disease

Receptor-mediated endocytosis
Several molecules that attach to special receptors called clathrincoated pits on the outside of cells cause the cell to perform endocytosis, an invagination of the plasma membrane to incorporate the molecule and associated structures into endosomes. This mechanism is used for three main purposes:
y y

Uptake of essential metabolites, for example, LDL. Uptake of some hormones and growth factors, for example, epidermal growth factor and nerve growth factor. y Uptake of proteins that are to be destroyed, for example, antigens in phagocytotic cells like macrophages.

Receptor-mediated endocytosis can also be "abused": Some viruses, for example, the Semliki forest virus, enter the cell through this mechanism. y Cholera, diphtheria, anthrax, tetanus, botulinum, and other bacterial toxins enter the cell this way.
y

Protein destruction
Defective proteins are occasionally produced, or they may be damaged later, for example, by oxidative stress. Damaged proteins can be recycled. Proteins can have very different half lifes, mainly depending on their N-terminal amino acid residue. The recycling mechanism is mediated by ubiquitin.

Protein targeting in bacteria

With some exceptions, Bacteria lack membrane-bound organelles as found in eukaryotes, but they may assemble proteins onto various types of inclusions such as gas vesicles and storage granules. Bacteria may have a single plasma membrane (Gram-positive bacteria), or an inner membrane plus an outer membrane separated by the periplasm (Gram-negative bacteria). Proteins may be incorporated into the plasma membrane, or either trapped in the periplasm or secreted into the environment, according to whether or not there is an outer membrane. The basic mechanism at the plasma membrane is similar to the eukaryotic one. In addition, bacteria may target proteins into or across the outer membrane. Systems for secreting proteins across the bacterial outer membrane may be quite complex and play key roles in pathogenesis. These systems may be described as type I secretion, type II secretion, etc.

In most Gram-positive bacteria, certain proteins are targeted for export across the plasma membrane and subsequent covalent attachment to the bacterial cell wall. A specialized enzyme, sortase, cleaves the target protein at a characteristic recognition site near the protein C-terminus, such as an LPXTG motif (where X can be any amino acid), then transfers the protein onto the cell wall. An system analogous to sortase/LPXTG, termed exosortase/PEPCTERM, is proposed to exist in a broad range of Gram-negative bacteria.

Secretory pathways
The secretory pathway includes vesicular traffic, secretion, and endocytosis. Secretory proteins follow this pathway.

Early stages
Retrograde transport is common in the early stages. Proteins that have been successfully delivered to the Golgi apparatus advance through cisternal progression.

Later stages
Coated vesicles mediate several transport steps.